A complete business case — built around your firm's specific numbers, your current AI readiness level, and the realistic path from where you are today to a measurably different tax season next year.
Human Brilliance. Amplified by AI.
Package Contents — 12 Documents
| Prepared for | David Castillo, Castillo, Reeves & Associates |
| Date | March 2026 |
| Version | 1.0 |
| Prepared by | Synergi AI Sales Team |
| Contact | hello@synergiai.io |
Dear David Castillo,
Thank you for the conversation earlier this month. What you described is one of the most consistent patterns we see across mid-size CPA firms: your most experienced, highest-cost people are spending the majority of tax season on work that does not require their expertise.
Senior CPAs billing at $150–$350 per return spend a significant portion of their time on routine data extraction, input verification, and form population — tasks that an AI-assisted workflow can handle in a fraction of the time, with a full human review layer before anything goes out the door. At 35 people processing 2,400+ returns per year, that bottleneck is costing your firm real margin, real capacity, and real morale. We built Synergi to address exactly that problem.
This package is not a brochure. It is a complete business case — built around your firm's specific numbers, your current AI readiness level, and the realistic path from where you are today to a measurably different tax season next year. Every document in this package is connected: the same assumptions, the same numbers, the same recommended approach, sequenced so that you can read straight through or jump to the section most relevant to your immediate questions.
| # | Document | What It Covers |
|---|---|---|
| 00 | Cover Letter | This document. Context, framing, and next step. |
| 01 | Discovery & Strategic Assessment | AI readiness score, strategic options matrix, recommended path, and kill criteria. |
| 02 | Market Evaluation | AI adoption trends in accounting, competitive landscape, regulatory environment, and the opportunity window. |
| 03 | Solution Architecture | How Insight 360's 28-agent system maps to your 6 departments, plus integration points and data flow. |
| 04 | ROI Analysis | Your current cost baseline vs. AI-assisted economics. Margin expansion from 35% to 67% at scale. |
| 05 | Marketing Positioning | Brand strategy, audience messaging, client disclosure template, and FAQ guide. |
| 06 | Go-to-Market Strategy | 5-week rollout, channel strategy, target segments, objection handling. |
| 07 | Campaign Package | 4 ready-to-deploy campaigns with email copy, LinkedIn sequences, and referral partner program. |
| 08 | Employee AI Upskill Program | 5-module training curriculum, certification path, and change management framework. |
| 09 | Implementation Roadmap | 5-phase rollout from alignment through optimization, with decision gates and milestones. |
| 10 | Pricing Proposal | Synergi Professional at $799/month (annual), implementation package, and Proof-of-Value guarantee. |
| 11 | Appendix: Agent Reference | Complete 28-agent roster with roles, responsibilities, and activation timeline. |
We are not asking you to automate your firm. We are asking you to run three tax returns through an AI-assisted workflow — with your CPAs reviewing and signing every line — and measure the result. That is the pilot. If it works, you expand. If it does not meet your standards, you stop. No long-term commitment is required to find out.
The firms that will define the profession over the next five years are not the ones that wait for AI to become obvious. They are the ones that figure out how to use it carefully, with human judgment at every critical decision point, before their competitors do.
Your current AI readiness level — what we call Level 2, Experimenting — is actually the ideal entry point for this kind of structured, low-risk implementation. You have enough exposure to know what the technology can do. You do not yet have the systematic deployment that turns that curiosity into competitive advantage. That is exactly the gap this program is designed to close.
Schedule a 30-minute walkthrough of this package with your operations lead and one senior CPA. The goal of that call is not to make a decision — it is to walk through the ROI model with your actual numbers and identify which 3 returns would be ideal candidates for the pilot.
To schedule, email us at hello@synergiai.io or reply to the message that accompanied this package.
We look forward to the conversation.
Sincerely,
The Synergi AI Team
hello@synergiai.io
This package was prepared specifically for Castillo, Reeves & Associates based on information gathered during initial discovery. Numbers reflect publicly available industry benchmarks and firm-level estimates provided during our conversation. All figures will be validated and refined during the pilot phase.
| Prepared for | David Castillo, Castillo, Reeves & Associates |
| Prepared by | Higgins (Strategic Advisor) via Synergi Insight 360 |
| Date | March 2026 |
| Version | 1.0 |
| Classification | Confidential — Prospect Use Only |
This assessment synthesizes what we know about Castillo, Reeves & Associates after initial discovery conversations and applies Synergi's structured strategic framework to determine your AI readiness, the realistic options available to you, and the recommended path forward. The headline finding is this: your firm is operating at AI Readiness Level 2 (Experimenting), which places you at the inflection point where unstructured experimentation either consolidates into a systematic capability or fades into background noise. The recommended path — AI-assisted tax prep with mandatory human review — is moderate in risk, high in reward, and fully reversible if it fails to meet your standards. The pilot is designed to answer the only question that matters: does this actually work for your clients, at your quality bar, in your workflow?
| Dimension | Detail |
|---|---|
| Firm size | 35 people |
| Annual revenue | $8M |
| Departments | 6 (Tax Prep, Advisory, Compliance, Finance, Business Development, Administration) |
| Annual return volume | 2,400+ returns/year |
| Peak season | January–April (tax season) with secondary September–October (extensions) |
| Primary revenue driver | Tax preparation and compliance work |
| Billing model | Per-return: $150 simple / $350 complex |
The firm's core operational challenge is not a people problem or a technology problem — it is a workflow architecture problem. During the four months of peak tax season, the firm's senior CPAs function as both the highest-value client-facing advisors and the primary operators of a high-volume, time-sensitive production process. These two roles are in direct conflict.
Routine tax preparation — data extraction from source documents, input into preparation software, initial form population, and basic reconciliation checks — requires accuracy and attention to detail. It does not require the judgment, client relationships, or professional expertise that make a senior CPA valuable. Yet this work consumes the majority of senior CPA time during the period when that time is scarcest and most expensive.
The downstream effects are predictable:
If senior CPAs spend 60% of their tax season time on routine preparation tasks currently executable by an AI-assisted workflow, the math is straightforward:
| Metric | Estimate |
|---|---|
| Senior CPA effective hours in tax season | ~500 hrs/person |
| Hours on routine prep (60%) | ~300 hrs/person |
| Hours freed by AI-assisted workflow (60% reduction in prep time) | ~180 hrs/person |
| Value of freed hours | Significant reallocation to advisory and overflow capacity |
This is not about eliminating people. It is about redirecting existing capacity to work that generates more value per hour.
Level 2 — Experimenting
On the Synergi AI Readiness Spectrum
| Level | Name | Description |
|---|---|---|
| 1 | Overwhelmed | No AI usage; actively resistant or unaware |
| 2 | Experimenting | Individual use of ChatGPT or similar tools; no systematic deployment; results are inconsistent |
| 3 | Focused | One or two defined AI use cases with measurable outcomes; starting to build internal knowledge |
| 4 | Aligned | AI embedded into core workflows; team-level adoption; governance in place |
| 5 | Orchestrated | Multi-agent AI system operating across all departments; AI is a strategic asset |
Firms at Level 1 require foundational mindset change before technology implementation. Firms at Level 3+ are already building competitive advantage and have less to gain from an outside program. Level 2 firms have demonstrated enough curiosity to experiment but have not yet converted that curiosity into systematic, measurable outcomes. They are not locked in to any particular approach, and they have enough internal credibility for AI to propose a structured program rather than starting from zero.
The evidence for Level 2 at Castillo, Reeves & Associates:
Three paths are available. Each has been evaluated across five dimensions.
| Dimension | Option A: Full AI Automation | Option B: AI-Assisted Prep with Human Review | Option C: Wait and Watch |
|---|---|---|---|
| Risk level | High | Moderate | Low |
| Reward potential | High | High | None near-term |
| Reversibility | Low | High | N/A |
| Speed to value | 6–12 months | 5 weeks to pilot | 12–24 months (if ever) |
| Client trust impact | Requires careful management | Transparent and defensible | No change |
| Staff impact | Role elimination risk; attrition likely | Upskilling; CPAs move up the value chain | Frustration compounds |
| Regulatory exposure | Highest | Manageable with governance | None |
| Recommended | No | Yes | No |
Replace the human-in-the-loop for preparation and initial review with a fully automated pipeline. AI prepares, AI reviews, CPA signs. This is technically feasible for a narrow band of very simple returns (W-2 only, standard deduction) but creates unacceptable liability exposure for complex returns, is unsupportable under current AICPA guidance, and would likely trigger client trust erosion if disclosed or discovered. The reward potential is real but the reversibility is low — once clients and staff adapt to a fully automated workflow, reversing it is operationally difficult and signals instability. Not recommended at this stage.
AI handles data extraction, document parsing, form pre-population, and initial consistency checks. The CPA reviews every output, makes all judgment calls, and signs every return. The client experience is unchanged. The internal workflow changes significantly: CPAs spend time on review, exception handling, and advisory work rather than on data entry and form population. This is the path that is defensible to clients, consistent with AICPA guidance, insurable under standard E&O coverage, and fully reversible if quality metrics decline. It is also the path most likely to generate the 60% prep time reduction that underpins the ROI case.
Do nothing now; monitor the market; revisit in 12–24 months. The risk is genuinely low. The cost is the opportunity: competitors who implement now will have 12–24 months of process refinement, staff training, and client positioning before you start. In a profession where switching costs are low and referral networks are everything, a 12-month head start matters. Technology maturity in AI-assisted accounting is advancing faster than the market has priced, and the firms that wait for consensus are rarely the ones that define the next era of practice. Not recommended unless the pilot fails.
AI-Assisted Preparation with Mandatory Human Review — Positioned as a Quality Upgrade
The framing matters as much as the implementation. This is not a cost-cutting initiative. It is a capacity and quality initiative: your CPAs now have the bandwidth to review more thoroughly, catch more issues, and spend more time on the judgment work that actually protects clients. The AI handles the tedious part so the expert can focus on the consequential part.
"We are investing in tools that let our CPAs spend more time doing the work only they can do."
"We use AI-assisted document processing to ensure faster turnaround and to free up our CPAs to focus on review and advisory — your return is always reviewed and signed by a licensed CPA."
The pilot and any subsequent scale should be abandoned immediately if any of the following thresholds are crossed:
| Criterion | Threshold | Measurement |
|---|---|---|
| Client satisfaction | Drops below 90% satisfaction on post-filing survey | Client survey at filing completion |
| Error rate | Exceeds the manual baseline error rate established in the same period | QA audit of pilot returns vs. control group |
| Staff attrition signal | Any staff member cites AI implementation as a reason for departure | Exit interview + ongoing 1:1 monitoring |
| Regulatory action | Any IRS or state board inquiry linked to AI-assisted preparation | Immediate legal review trigger |
| Data incident | Any unauthorized access to or exposure of client tax data | Zero tolerance; automatic halt |
These criteria are not aspirational. They are contractual commitments in the pilot agreement. If any threshold is crossed, the pilot stops, the data is reviewed, and no further rollout occurs until the cause is identified and resolved.
| Value | Implementation Expression |
|---|---|
| Transparency | One-page AI disclosure available for clients who ask; internal documentation of what AI does and does not do in the workflow |
| Human-centered | CPAs always review and sign every return; no client-facing output is generated without licensed CPA approval |
| Trust-first | Pilot runs before any firm-wide commitment; kill criteria are defined and binding before the first return is processed |
| Professional integrity | AI is positioned as a preparation tool, not a judgment tool; all professional responsibility remains with the licensed CPA |
| Quality over throughput | Success metrics include quality measures (error rate, client satisfaction) not just speed metrics |
Your firm is at Level 2 on the AI Readiness Spectrum — the ideal entry point for a structured, low-risk implementation.
The core problem is workflow architecture: senior CPAs are the bottleneck in a process that does not require all of their expertise.
AI-assisted prep with human review is the only option that is simultaneously high-reward and fully reversible.
The pilot is the right first move: 3 returns, defined success criteria, and a clear go/no-go decision point before any broader commitment.
Framing matters: this is a quality and capacity initiative, not a cost-cutting exercise.
| Prepared for | David Castillo, Castillo, Reeves & Associates |
| Prepared by | Blake (Competitive Intel) + Kieran (Account Research) via Synergi Insight 360 |
| Date | March 2026 |
| Version | 1.0 |
| Classification | Confidential — Prospect Use Only |
The AI adoption curve in accounting and tax preparation has crossed the early-adopter threshold and is accelerating into early majority territory. Approximately 15–20% of mid-size CPA firms (10–100 staff) have begun experimenting with AI-assisted workflows, and a smaller but growing cohort — estimated at 5–8% — have deployed systematic AI programs with measurable outcomes. The window for meaningful competitive differentiation through early, well-governed AI adoption is open but narrowing. Regulatory bodies have moved from silence to cautious guidance, the liability landscape is clarifying, and the technology has matured to a point where the question is no longer "can this work?" but "who will implement it well?" This document gives you the market context to answer that question with confidence.
| Firm Segment | Staff Count | Estimated AI Adoption | Typical Use Cases |
|---|---|---|---|
| Large national firms | 500+ | 70–80% | Audit analytics, tax research, document review |
| Regional/mid-size firms | 50–500 | 30–45% | Tax prep assistance, document processing, research |
| Mid-size firms (your segment) | 10–100 | 15–20% | ChatGPT for drafting, some document OCR, limited workflow integration |
| Small firms | 1–9 | 8–12% | Individual ChatGPT use; no systematic deployment |
Source estimates based on AICPA 2025 Technology Survey data, Accounting Today research, and Thomson Reuters Institute industry reports. Figures reflect North American public accounting firms.
Among the 15–20% of mid-size firms currently experimenting, the most common use cases are:
What the experimenting firms are not yet doing: running AI outputs through a structured quality review, maintaining audit trails of AI involvement in return preparation, or tracking measurable outcomes from AI use. This is the gap between Level 2 (Experimenting) and Level 3 (Focused) on the AI Readiness Spectrum.
The acceleration is driven by three converging forces:
| Position | Description | Estimated Share | Trajectory |
|---|---|---|---|
| AI Leaders | Systematic deployment with measurable outcomes; actively marketing their AI capability as a differentiator | 5–8% | Growing; likely to pull away from the market |
| AI Followers | Structured experimentation underway; 6–18 months behind leaders | 10–15% | Catching up; competitive parity likely within 2 years |
| Passive Experimenters | Individual tool use with no governance or measurement; current state of most firms | 15–20% | Risk of falling behind if they do not systematize |
| Non-Adopters | Aware but waiting; some actively resistant | 55–65% | Significant competitive exposure as the market moves |
Castillo, Reeves & Associates's current position is Passive Experimenter, near the boundary with Follower. The pilot program would move you to Follower status, and successful scale would place you in Leader territory within 12–18 months.
The 5–8% of mid-size firms leading on AI share several characteristics:
Firms with AI-assisted workflows are beginning to compete on turnaround time rather than price. Because their cost per return is lower (AI handles the time-intensive preparation steps), they can offer faster delivery without reducing their margin. This creates a subtle but growing competitive disadvantage for firms still running fully manual workflows: clients who value speed begin to notice the difference, particularly for straightforward returns where a 48-hour turnaround is meaningfully better than 2 weeks.
| Metric | Estimate |
|---|---|
| Mid-size CPA firms in the US (10–100 staff) | ~12,000 firms |
| Annual revenue of this segment | ~$45–60B |
| Returns processed annually by this segment | ~80–100M individual and business returns |
| Firms likely to adopt systematic AI workflows in next 3 years | 3,000–5,000 (25–40%) |
| Potential annual time savings if 60% prep time reduction is realized | Hundreds of millions of CPA hours |
For Castillo, Reeves & Associates specifically, the addressable opportunity is the capacity that currently sits locked inside senior CPA hours spent on routine preparation. At 2,400+ returns per year and a 60% prep time reduction, the downstream effect is a significant reallocation of existing capacity — not just cost savings, but the ability to take on more clients, expand advisory services, or improve work-life balance during tax season.
The IRS has not issued a prohibition on AI-assisted return preparation as of March 2026. The agency's focus has been on preparer accuracy and liability rather than the specific tools used to achieve that accuracy. The core principle is unchanged: a CPA who signs a return is responsible for its accuracy regardless of how the underlying data was prepared or organized. AI as a preparation tool does not change the professional responsibility framework; it changes only who (or what) does the initial data organization.
The IRS has issued informal guidance noting that preparers should document their review process, which is consistent with the AI-assisted + mandatory human review model recommended here.
State CPA licensing boards have been largely silent on AI-specific guidance, operating under existing standards that require CPAs to exercise professional judgment and maintain competence. The key existing standards that apply:
| Standard | Application to AI-Assisted Prep |
|---|---|
| AICPA Code of Professional Conduct — Due Care | Requires competence in the tools used; CPAs should understand what the AI does and does not do |
| AICPA SSTS No. 1 (Tax Return Positions) | Preparer responsibility for return accuracy is unchanged; AI does not create or absorb liability |
| State licensing board continuing education | Some states are beginning to offer (and may eventually require) CPE credit for AI literacy in tax practice |
The AICPA's 2024 and 2025 guidance on AI in accounting practice has coalesced around three principles:
The AI-assisted + mandatory human review model described in this package is fully consistent with all three principles.
| Risk Factor | Current Status |
|---|---|
| Premium impact for AI-using firms | Mixed; some carriers offer slight premium reductions for firms with documented AI governance; others have added questionnaires |
| Coverage exclusions for AI-related errors | Not yet standard, but carriers are watching; governance documentation is the primary protection |
| Recommended protection | Document the AI workflow, maintain audit trails of what AI produced vs. what the CPA reviewed and modified, and ensure the review layer is substantive (not perfunctory) |
Notify your E&O carrier before implementing AI-assisted workflows. Provide them with your governance documentation. This is both good risk management and potentially beneficial for your renewal.
| Risk Category | Specific Risk | Likelihood | Impact | Mitigation |
|---|---|---|---|---|
| Regulatory | State board issues new restrictive guidance mid-implementation | Low | Moderate | Monitor AICPA and state board communications; governance documentation provides a defensible record |
| Liability | AI-assisted return contains an error that leads to client harm | Low (with review layer) | High | Mandatory CPA review of every AI output; no return signed without substantive review |
| Client trust | Client discovers AI was used and objects | Moderate | Moderate | Proactive disclosure policy; emphasize CPA review and sign-off |
| Data security | Client tax data exposed through AI platform | Low (with proper vendor selection) | Very High | SOC 2 compliant platform; data isolation; no training on client data |
| Staff resistance | CPAs resist or undermine AI workflow adoption | Moderate | Moderate | Involve senior CPAs in pilot design; frame as capacity relief, not replacement |
| Technology | AI performance degrades on complex returns | Moderate | Low (with review layer) | Pilot selection includes a mix of complexity levels; kill criteria include error rate monitoring |
Three conditions are currently aligned that will not remain aligned indefinitely:
1. Technology has crossed the "good enough" threshold. For structured document processing and form pre-population, AI performance in 2025–2026 is materially better than it was in 2022–2023. The gap between AI-extracted data accuracy and junior-staff-entered data accuracy has closed to the point where AI-assisted is defensible in a production environment with a review layer. This was not true three years ago.
2. The market has not yet priced in the advantage. The 5–8% of firms leading on AI are gaining an advantage, but they have not yet made that advantage visible enough to reshape client expectations market-wide. There is still time to join the leading cohort before the laggard cohort becomes large enough to normalize non-adoption in the eyes of clients.
3. Your firm's current scale is ideal for implementation. At 35 people and 2,400+ returns, you are large enough to benefit meaningfully from a 60% prep time reduction and small enough for a 5-week implementation to touch every relevant part of the workflow. Larger firms face change management complexity that slows adoption; smaller firms lack the volume to generate compelling ROI data from a pilot.
| Year | Scenario: Implement Now | Scenario: Wait 24 Months |
|---|---|---|
| 2026 | Pilot complete; 50+ returns/month at 67% margin | Manual workflow; 35% margin |
| 2027 | Full scale; competitive differentiation on speed | Begin implementation as laggard; 18 months behind |
| 2028 | 2 years of compounding margin advantage | Catching up; limited differentiation opportunity |
The cost of waiting is not dramatic in year one. It compounds in years two and three as early adopters reinvest their margin advantage into client acquisition, staff retention, and further capability development.
Approximately 15–20% of mid-size CPA firms are experimenting with AI; 5–8% have systematic deployment. The window for first-mover advantage in your market segment is open but narrowing.
The regulatory environment is permissive for AI-assisted prep with mandatory human review. Professional responsibility remains unchanged; the tool changes, not the liability.
E&O insurers are watching but not restricting. Documentation and governance are the primary protection.
The technology has crossed the threshold where AI-assisted document processing is defensible in a production workflow with a review layer.
At 35 people and 2,400+ returns per year, your firm is in the ideal size range to implement quickly and generate compelling before/after data within a single tax season.
| Prepared for | David Castillo, Castillo, Reeves & Associates |
| Prepared by | Riley-O (Operations Lead) + Alfred (Chief of Staff) via Synergi Insight 360 |
| Date | March 2026 |
| Version | 1.0 |
| Classification | Confidential — Prospect Use Only |
This document maps the Synergi Insight 360 platform to the specific operational structure of Castillo, Reeves & Associates. Insight 360 is a multi-agent AI orchestration system built around 28 specialized agents organized into 6 functional streams plus an executive layer. The platform runs on a three-phase pipeline — Align, Strategy, Execute — that mirrors how a well-run firm thinks through any significant initiative: understand the situation first, decide what to do, then execute with clear ownership and accountability. For Castillo, Reeves & Associates, the immediate deployment focus is the Tax Prep and Operations stream, with the platform architected to expand into Advisory, Compliance, Finance, Business Development, and Administration as the firm builds confidence and capability. The data flow is designed from the ground up for professional service firm requirements: client data is isolated per engagement, never used to train models, and handled in compliance with SOC 2 standards.
Every Insight 360 engagement runs through the same three-phase pipeline. The phases are sequential but iterative — the firm can return to an earlier phase if new information changes the picture.
(Who are we?)
(What should we do?)
(Who does what, by when?)
Duration: ~120 minutes of structured AI-assisted work (not elapsed time)
The Align phase establishes shared context. For a CPA firm beginning AI-assisted tax prep, this means:
Output: A shared context document that every agent in the system references, ensuring consistent understanding across all functions.
Duration: ~120 minutes of structured AI-assisted work
The Strategy phase produces the implementation roadmap. Higgins (Strategic Advisor) and the department-specific agents analyze the options, model the outcomes, and produce a concrete recommendation with explicit assumptions. For the CPA firm implementation, the Strategy phase output includes:
Output: A signed-off strategic plan with clear owner assignments and decision documentation.
Duration: ~120 minutes per execution cycle, repeated as needed
The Execute phase converts strategy into action. Riley-O (Operations Lead) coordinates task assignment, tracks milestones, monitors SLA compliance, and surfaces issues before they become problems. For ongoing tax season operations, the Execute phase is a recurring cycle that covers each batch of returns from intake through filing.
Output: Completed work products, progress reports, and exception flags for human review.
Castillo, Reeves & Associates has 6 departments. Each maps to one or more Insight 360 agent streams. The mapping below shows the primary and secondary stream assignments for each department.
| Firm Department | Primary Agent Stream | Secondary Agent Stream | Primary Use Cases |
|---|---|---|---|
| Tax Prep | Operations | Support | Document ingestion, data extraction, form pre-population, QA review |
| Advisory | Sales | Marketing | Client meeting prep, proposal drafting, engagement letter generation |
| Compliance | Legal | Operations | Regulatory monitoring, filing deadline tracking, audit support |
| Finance | Finance | Cross-functional | Internal P&L, budget vs. actual, revenue ops |
| Business Development | Marketing | Sales | ICP targeting, outreach content, competitive positioning |
| Administration | Operations | Cross-functional | Vendor management, resource planning, SLA monitoring |
For the pilot and initial scale, the deployment is focused on the Tax Prep department. The Operations and Support agents handle the full document-to-draft workflow. The Legal and Finance streams provide oversight and governance. The Sales and Marketing streams are activated in Phase 2 of the broader rollout (post-pilot), supporting Business Development and Advisory growth enabled by the capacity freed up in Tax Prep.
28 agents across the executive layer, cross-functional team, and 6 department streams.
| Agent | Role | Primary Function |
|---|---|---|
| Jarvis | Executive Orchestrator | Overall system coordination; routes complex multi-department requests; escalation handler for anything that crosses stream boundaries |
| Alfred | Chief of Staff | Calendar of initiatives; progress reporting to managing partner; keeps all streams aligned to the firm's strategic priorities |
| Higgins | Strategic Advisor | Strategic options analysis; kill criteria monitoring; quarterly readiness assessments; the agent that produced Document 01 |
| Agent | Role | Primary Function |
|---|---|---|
| Kendall | Pricing Strategist | Return pricing analysis; margin modeling; fee schedule optimization |
| Cameron | Finance Analyst | ROI modeling; cost-per-return tracking; budget vs. actual for the AI implementation |
| Morgan-L | Legal Advisor | Regulatory compliance review; E&O documentation; client disclosure language |
| Marley | Follow-Up Manager | Client communication tracking; deadline reminders; post-filing satisfaction check-ins |
| Agent | Role | Primary Function |
|---|---|---|
| Dakota | Marketing Orchestrator | Coordinates all marketing stream agents; routes requests; manages content calendar |
| Harper | Brand Voice | Ensures all external communications maintain consistent tone and brand standards |
| River | Content Agent | Drafts articles, newsletter content, client education materials, and thought leadership |
| Avery-M | ICP Adapter | Adapts messaging and content for different client segments (individuals, small business, high-net-worth) |
| Blake | Competitive Intel | Monitors competitor positioning, industry news, and market shifts relevant to the firm |
| Emery | Brand Reviewer | Final review layer for all external content before publication or distribution |
| Agent | Role | Primary Function |
|---|---|---|
| Tatum | Sales Orchestrator | Coordinates all sales stream agents; manages pipeline visibility; routes requests |
| Drew | Proposal Writer | Drafts service proposals, engagement letters, and scope-of-work documents |
| Reese-S | Sales Enablement | Maintains objection playbooks, talk tracks, and supporting materials |
| Kieran | Account Research | Pre-meeting prospect and client research; identifies expansion opportunities |
| Remy | Outreach Agent | Drafts personalized outreach emails, LinkedIn messages, and follow-up sequences |
| Jules | Call Prep | Prepares briefing documents before client or prospect calls |
| Hayden | Pipeline Analyst | Tracks pipeline stage, velocity, and conversion metrics |
| Logan | Deal Strategist | Analyzes complex or competitive deals; recommends approach and pricing |
| Agent | Role | Primary Function |
|---|---|---|
| Marlowe | Finance Team Lead | Coordinates Finance stream; owns internal financial reporting and management dashboards |
| Noel | Budget & Forecast | Monthly budget vs. actual analysis; rolling forecast; scenario modeling |
| Harley | Revenue Ops | Revenue recognition, billing accuracy, and collection tracking |
| Agent | Role | Primary Function |
|---|---|---|
| Morgan-L | Legal Advisor | (Also cross-functional) Regulatory compliance, E&O documentation, engagement letter legal review |
| Riley | Compliance Auditor | Ongoing monitoring of IRS guidance, state board positions, and AICPA standards relevant to AI-assisted practice |
| Agent | Role | Primary Function |
|---|---|---|
| Riley-O | Operations Lead | Primary owner of the Tax Prep AI workflow; coordinates document ingestion, processing, and QA handoff to CPAs |
| Sage-O | Vendor Manager | Manages relationships with tax software vendors, document management providers, and AI platform integrations |
| Nico | Resource Planner | Allocates CPA review capacity across the return queue; flags bottlenecks before they become delays |
| Kai | Process Analyst | Documents workflow steps, identifies inefficiencies, and recommends process improvements |
| Devon | SLA Monitor | Tracks turnaround time commitments; alerts when a return is at risk of missing a deadline |
| Agent | Role | Primary Function |
|---|---|---|
| Kris | Knowledge Manager | Maintains the firm's internal knowledge base: past returns (anonymized), common issues, CPA review notes |
| Toni | Escalation Handler | Routes exceptions (unusual return complexity, data quality issues, client disputes) to the appropriate CPA or partner |
| Dallas | Policy Tuner | Updates AI workflow instructions based on new IRS guidance, firm policy changes, or lessons from QA reviews |
| Frankie | Quality Reviewer | Final QA check on AI-prepared outputs before they reach the CPA review queue |
| System Category | Common Tools | Integration Type | Purpose |
|---|---|---|---|
| Tax preparation software | ProConnect Tax, Lacerte, UltraTax CS, Drake, CCH Axcess | API or structured export | Receives AI-extracted data for form pre-population; sends completed return data back for QA comparison |
| Document management | Canopy, TaxDome, ShareFile, Box | API or secure file sync | Secure intake of client source documents; output storage of AI-prepared draft returns |
| Practice management | Karbon, Financial Cents, Jetpack Workflow | API or webhook | Syncs return status, milestone completion, and CPA assignment with existing workflow tracking |
| CRM / Client portal | Salesforce, HubSpot, or firm-specific client portals | API | Updates client communication history; triggers follow-up workflows post-filing |
| Accounting / Finance | QuickBooks, Xero (for internal firm financials) | Read-only API | Cameron (Finance Analyst) pulls firm P&L data for internal ROI tracking and margin analysis |
| Communication | Email (Google Workspace / Microsoft 365) | API | Marley (Follow-Up Manager) drafts and schedules client communications; Jules (Call Prep) pulls email history for briefings |
For the 3-return Proof-of-Value pilot, deep integration is not required. The pilot can run with:
Full integration with tax preparation software is implemented in Week 3–4 of the broader 5-week rollout.
How client tax data moves through the Insight 360 system, from initial upload through filing.
Client submits source documents (W-2s, 1099s, K-1s, prior returns, etc.) via your existing secure client portal or document management system.
Riley-O (Operations Lead) triggers the ingestion workflow. Documents are transferred to an isolated, encrypted workspace. Each client's data is logically separated from all other clients. No document content is retained after the return is completed and archived.
Insight 360 processes each document: OCR and structured data extraction from PDFs and images, entity identification, income item categorization, deduction and credit identification, and consistency checks. Extraction confidence scores are generated for each data point. Low-confidence items are flagged for CPA attention.
Extracted data is organized into a structured draft output: line-by-line population of applicable forms, supporting schedule preparation, open questions flagged for CPA judgment, and prior-year comparison for anomaly detection. Frankie (Quality Reviewer) runs automated QA checks before handoff.
Draft and all source documents are delivered to the assigned CPA via the review queue. CPA reviews every line, resolves all flagged items, and makes all judgment calls. CPA documents their review. No return proceeds to filing without explicit CPA approval.
CPA-approved return is sent to client for review via existing client portal. Client approves; return is filed through your standard process. Devon (SLA Monitor) confirms filing and updates the practice management system.
Marley (Follow-Up Manager) triggers post-filing client satisfaction check. Outcome data (prep time, error rate, client response) is logged for ROI tracking. Client workspace is archived; no data retained in active AI processing environment.
| Commitment | Implementation |
|---|---|
| Data isolation | Each client's documents and extracted data are stored in a logically isolated workspace. No cross-client data access is possible within the system. |
| No training on client data | Client tax data is never used to train, fine-tune, or improve the AI models. Processing is inference-only. |
| SOC 2 compliance | The Synergi Insight 360 platform operates under SOC 2 Type II controls. Documentation available on request. |
| Encryption | All data in transit is encrypted via TLS 1.3. All data at rest is encrypted via AES-256. |
| Retention limits | Client data is retained only for the duration of the active engagement plus a firm-defined archival period. Deletion is confirmed and documented. |
| Access controls | Role-based access controls ensure that agents only access the data required for their specific function. |
| Audit trail | Every AI action on client data is logged with timestamp, agent identity, and action type. This log is available to the CPA and the firm at any time. |
Insight 360's 28-agent system maps directly to Castillo, Reeves & Associates's 6 departments, with the Operations and Support streams owning the Tax Prep workflow from document intake through CPA handoff.
The three-phase pipeline (Align 120 → Strategy 120 → Execute 120) provides a repeatable structure for every implementation decision — not a one-time setup but an ongoing operating rhythm.
Integration with existing tools (tax software, document management, practice management) is achievable within the 5-week implementation timeline, with the pilot requiring minimal integration to run.
The data flow architecture is designed for professional service firm requirements: client isolation, no model training on client data, full audit trails, and SOC 2 compliance.
The CPA review layer is not an optional add-on — it is architecturally embedded in the workflow. No return reaches filing without explicit CPA approval.
This analysis quantifies the financial impact of adopting Synergi Professional at a 35-person CPA firm processing 2,400+ returns per year across 6 departments. The core opportunity: AI-assisted preparation compresses per-return labor costs by reducing prep time 60%, allowing the same team to handle 50% more volume — and improving gross margins from 35% to 67% at scale. With a total Year 1 investment of approximately $22,000 (or $19,600 on annual billing), the practice reaches break-even within 3 months of scaled operation. The math is straightforward; the strategic upside is compounding.
| Cost Item | Monthly | Annual (Monthly) | Annual (Prepaid) |
|---|---|---|---|
| Synergi Professional Plan | $999/mo | $11,988/yr | $9,588/yr ($799/mo) |
| Implementation Package (one-time) | — | $7,500 | $7,500 |
| Executive Training (one session, one-time) | — | $2,500 | $2,500 |
| Total Year 1 Cost | ~$22,000 | ~$19,600 | |
| Total Year 2+ Cost (recurring only) | $11,988 | $9,588 |
Notes:
| Metric | Current Value |
|---|---|
| Total returns per year | 2,400+ |
| Blended average revenue per return | ~$225 |
| Gross annual revenue (tax prep only) | ~$540,000 |
| Average prep time per return (simple) | 3–4 hours |
| Average prep time per return (complex) | 8–12 hours |
| Staff capacity (tax season peak) | Fully utilized |
Synergi enables a tiered pricing structure that is both more competitive and more profitable:
| Return Type | Current Price | AI-Assisted Price | Prep Time Reduction |
|---|---|---|---|
| Simple (1040 + W-2s) | ~$175 | $150 | 60% |
| Complex (Schedule C, partnerships) | ~$350 | $350 | 60% |
| Blended average | ~$225 | ~$250 (mix shift) | 60% |
Key insight: simple returns decrease slightly in price (competitive positioning) while complex returns hold price but drop in cost. The net effect is margin expansion, not revenue reduction.
A 60% reduction in prep time does not mean a 60% reduction in staff — it means the same team can handle significantly more volume:
| Scenario | Returns/Year | Revenue (at $250 blended) |
|---|---|---|
| Current capacity | 2,400 | ~$540,000 |
| 25% volume increase | 3,000 | ~$675,000 |
| 50% volume increase | 3,600 | ~$810,000 |
| Advisory time unlocked | +$50,000–$150,000 | (CFO-lite, planning services) |
The advisory upside is particularly significant: when CPAs spend 60% less time on data entry and formatting, they have capacity for higher-margin engagements — financial planning, business advisory, and proactive strategy work that clients will pay premium rates for.
Current gross margin of 35% is driven primarily by labor cost. With 2,400 returns at an average blended rate of $225, gross revenue is approximately $540,000. Labor and overhead consume ~65% of that, leaving ~$189,000 in gross profit.
AI-assisted preparation changes the cost structure fundamentally:
| Cost Category | Current | AI-Assisted | Change |
|---|---|---|---|
| Labor per return (simple) | $95 | $38 | -60% |
| Labor per return (complex) | $210 | $84 | -60% |
| Blended labor cost per return | ~$147 | ~$59 | -60% |
| Software cost per return (Synergi) | $0 | ~$5 | +$5 |
| Net cost per return | ~$147 | ~$64 | -56% |
| Scale | Revenue/Return | Cost/Return | Gross Margin |
|---|---|---|---|
| Current (manual, any volume) | $225 | $147 | 35% |
| AI-Assisted (<50 returns/mo) | $225 | $110 | 51% |
| AI-Assisted (50+ returns/mo, steady state) | $250 | $83 | 67% |
At 50+ returns per month with optimized workflows, gross margin approaches 67%. This is the "steady state" achieved after 2–3 months of operational refinement post-launch.
Total Year 1 investment: ~$22,000 (monthly billing) or ~$19,600 (annual prepaid).
At 67% gross margin on $250/return blended:
More conservatively, accounting for ramp-up (Weeks 1–5 at reduced throughput):
| Month | Returns Processed | Cumulative Profit (at 67% margin) | Cumulative Investment |
|---|---|---|---|
| Month 1 (pilot) | 30 | $5,010 | $11,000 (onboarding) |
| Month 2 (ramp) | 100 | $11,700 (cumulative) | $13,000 |
| Month 3 (scale) | 200 | $28,100 (cumulative) | $14,999 |
| Month 3 end | Break-even achieved | ||
| Month 6 | 200/mo steady | $67,200 (cumulative) | ~$17,000 |
Break-even on full Year 1 investment is reached in approximately 3 months of scaled operation, or months 6–8 from project start (accounting for 5-week rollout).
| Metric | Year 1 (Ramp) | Year 2 (Scale) | Year 3 (Expand) |
|---|---|---|---|
| Base Case | |||
| Returns processed | 2,400 (same as today) | 3,000 (+25%) | 3,600 (+50% vs Y1) |
| Blended revenue/return | $225 | $240 | $250 |
| Gross revenue (tax prep) | $540,000 | $720,000 | $900,000 |
| Gross margin | 45% (blended ramp) | 60% | 67% |
| Gross profit | $243,000 | $432,000 | $603,000 |
| Synergi investment | $22,000 | $11,988 | $11,988 |
| Net improvement vs. current | +$54,000 | +$243,000 | +$414,000 |
| Best Case | |||
| Returns processed | 2,800 | 3,600 | 4,200 |
| Advisory revenue added | $50,000 | $120,000 | $200,000 |
| Gross profit | $320,000 | $550,000 | $780,000 |
| Worst Case | |||
| Returns processed | 2,400 (flat) | 2,400 (flat) | 2,600 |
| Gross margin achieved | 40% | 55% | 60% |
| Gross profit | $216,000 | $316,800 | $390,000 |
| Still an improvement vs. current | +$27,000 | +$127,800 | +$201,000 |
Even in the worst-case scenario — zero volume growth, slower margin improvement — the practice improves gross profit every year. The downside is limited; the upside is structural.
| Factor | Doing Nothing | Adopting Synergi |
|---|---|---|
| Gross margin | Stays at 35% | Grows to 67% at scale |
| Tax season labor pressure | Continues or worsens | Significantly reduced |
| Competitive position | Eroding (competitors adopt AI) | Differentiated |
| Client capacity | Capped at ~2,400/yr | Expandable to 3,600+ |
| Advisory time available | Minimal | Substantial |
| Staff burnout risk | High (peak season) | Reduced |
| Year 1 cost of inaction | $0 spent | ~$22,000 invested |
| Year 1 profit difference | $189,000 gross profit | $243,000+ gross profit |
| 3-year cumulative difference | Flat/declining | +$414,000+ vs. current |
| Quarter | Cash Out | Cash In (Incremental) | Net Quarterly Position |
|---|---|---|---|
| Q1 (Onboarding + Pilot) | -$12,997 | +$5,010 (pilot returns) | -$7,987 |
| Q2 (Ramp) | -$2,997 | +$28,100 (cumulative, 50% volume) | +$25,103 |
| Q3 (Scale) | -$2,997 | +$50,100 (full volume, 67% margin) | +$47,103 |
| Q4 (Steady State) | -$2,997 | +$50,100 | +$47,103 |
| Full Year 1 Net | -$21,988 | +$133,310 | +$111,322 vs. current baseline |
Note: "Cash In (Incremental)" represents profit above the current 35% margin baseline — the additional profit generated by AI-assisted operations, not total revenue.
The investment is modest, the return is not. At $22,000 Year 1 all-in, Synergi Professional pays for itself within 3 months of scaled operation.
Margin transformation, not marginal improvement. Going from 35% to 67% gross margin is not an optimization — it is a structural change to the business model.
Volume capacity is the hidden asset. Processing 3,600+ returns with the same team means growth without proportional hiring.
Advisory revenue is the long game. Hours freed from data extraction become hours available for higher-value, higher-margin services.
The worst case still wins. Even with no volume growth and slower margin improvement, Year 3 gross profit outperforms current by $200,000+.
Competitive timing matters. Early adoption locks in differentiation; delayed adoption means catching up to firms that already have the workflow advantage.
Adopting AI is only half the work. How a CPA firm communicates that adoption determines whether clients see it as a competitive advantage or a cause for concern. This document provides a complete positioning framework for a 35-person firm introducing AI-assisted tax preparation — covering brand strategy, audience-specific messaging, a client disclosure template, and a FAQ guide. The through-line is simple: technology that serves people must be explained in human terms. Done right, AI adoption becomes a differentiator that deepens client trust rather than eroding it.
AI handles the routine so your CPA can focus on what matters most — your financial future.
This single sentence does three things:
The firm is not positioning itself as a tech company. It is positioning as a trusted advisor that uses technology to deliver better outcomes for clients. The distinction matters enormously to the 45+ demographic that represents the majority of established CPA client bases.
These framings commoditize the service and invite price competition. The correct frame keeps the CPA at the center of the relationship, with AI in a supporting role.
| Pillar | What It Means | How to Communicate It |
|---|---|---|
| Trust | The CPA relationship remains human | "Same team, stronger tools" |
| Transparency | Clients know exactly what AI does and doesn't do | Clear disclosure, proactive FAQ |
| Quality | AI reduces errors, not review rigor | "Every return reviewed by a licensed CPA" |
| Human Oversight | A CPA checks and signs every return | Emphasize in every touchpoint |
Emotional state: Possible anxiety about change, loyalty to the current team, concern about data privacy.
Primary message: "Nothing changes about who manages your account. We're just giving your CPA better tools."
Supporting messages:
Tone: Warm, reassuring, familiar. Reference the relationship first, the technology second.
What to avoid: Technical jargon, promises of perfection, any language that sounds like a cost-cutting announcement.
Emotional state: Evaluating options, may have a poor experience with previous firm, price-sensitive but also quality-conscious.
Primary message: "Faster turnaround, deeper insights, same personal touch — because we invest in tools that give your CPA more time for what actually matters."
Supporting messages:
Referral partners: Attorneys, financial advisors, bankers, business coaches, HR consultants.
Primary message: "We're a forward-thinking firm investing in quality — which means your clients get faster service, better outcomes, and a team that has time to actually think about their situation."
Supporting messages:
The following is a template for a client-facing disclosure document. Customize with firm name, CPA signatures, and specific platform references as appropriate.
We believe you have the right to know exactly how we prepare your returns. Here is what you can expect when you work with us.
Every return prepared with AI assistance at Castillo, Reeves & Associates is reviewed line-by-line by a licensed Certified Public Accountant before submission. We do not submit any return without full human review and professional sign-off. This is not optional and not subject to change.
The following Q&As are designed for use on your website, in email communications, and as a leave-behind during client conversations.
A: AI assists in the preparation process — specifically with data extraction, formatting, and initial calculations. But your licensed CPA reviews everything, makes all judgment calls, and signs your return. Think of it like a GPS: it helps navigate, but a skilled driver is still behind the wheel.
A: Absolutely. Your CPA relationship does not change. What changes is how much of their time gets spent on data entry versus actually thinking about your tax situation. AI handles the routine; your CPA handles everything that matters.
A: Yes. We use enterprise-grade, SOC 2 compliant platforms. Your data is encrypted, never shared with third parties, and never used to train any AI model. The only people with access to your data are your engagement team — same as always.
A: No — in many cases, AI-assisted preparation allows us to offer more competitive pricing on straightforward returns while giving your CPA more time for complex work. Simple returns start at $150. Complex returns at $350. No surprise fees.
A: AI can make errors — which is exactly why every return goes through a full human review by a licensed CPA before submission. Our process is designed so that AI output is a starting point for your CPA's review, not a final product.
A: A licensed Certified Public Accountant at Castillo, Reeves & Associates signs every return. This is non-negotiable and will never change regardless of how our technology evolves.
A: We respect that. You can opt out of AI-assisted preparation and request a fully manual process. Just let your CPA know. We will never apply AI to your return without your knowledge.
A: Not even close. We're a 35-person firm with licensed CPAs at the center of everything we do. AI reduces the time they spend on administrative tasks — which means they have more capacity to think about your situation, answer your questions, and provide genuinely useful advice. We see this as an investment in our people, not a substitute for them.
A: Those are self-service tools with no professional oversight. We are a licensed CPA firm where AI is one tool among many, always under the supervision of a credentialed professional. The two are not comparable.
A: We launched our AI-assisted workflow in January 2026 after a rigorous evaluation, pilot testing, and staff training process. We did not adopt AI quickly — we adopted it carefully, and our protocols reflect that.
| Capability | Castillo, Reeves & Associates with Synergi | Traditional Manual Firm |
|---|---|---|
| Return turnaround time | 40–60% faster | Standard/slow during peak season |
| Data extraction accuracy | AI-assisted + CPA review | Manual entry (error-prone) |
| CPA advisory time per client | Significantly more | Limited during peak season |
| Capacity for volume growth | 3,600+ returns/yr (same team) | Capped at current staffing |
| Pricing (simple returns) | $150 | $175–$250+ |
| Proactive planning conversations | More common | Less available |
| Client disclosure on process | Full transparency | Rarely addressed |
| Staff burnout during tax season | Reduced | Ongoing concern |
The competitive advantage is not "we have AI." It is: our CPAs have more time for you because AI handles what doesn't require a CPA.
Lead with trust, follow with technology. Clients hire CPAs because they trust them. AI adoption must reinforce — not disrupt — that foundation.
Radical transparency is a competitive advantage. Firms that are vague about AI use create anxiety. Firms that are specific and honest about their process build confidence.
Segment your messaging. Existing clients need reassurance. Prospects need differentiation. Referral partners need credibility. One message does not serve all three.
The FAQ is a conversion tool. A well-crafted FAQ preempts objections before they become sales blockers.
"Your CPA, Amplified" is a durable position. It does not age out as AI improves. The more AI can do, the more it amplifies — and the more valuable the human judgment in the loop becomes.
Opt-out matters. Giving clients the right to opt out of AI-assisted preparation is not a liability — it is a trust signal. Very few clients will exercise it. All clients will feel better knowing they could.
A 35-person CPA firm does not need a $500,000 marketing budget to successfully launch AI-assisted tax preparation. It needs a disciplined 5-week rollout, a clear segmentation strategy, and a sales methodology that earns trust before it asks for commitment. This document provides the operational playbook: from internal announcement to public launch, from client segmentation to objection handling. The goal is a first client rollout in 5 weeks with zero client attrition and measurable margin improvement by Week 8.
| Week | Phase | Key Actions | Owner |
|---|---|---|---|
| Week 1 | Internal Announcement + Training | All-staff briefing on AI strategy and positioning; begin CPA workflow training with Synergi; establish internal FAQ and communication protocols | Managing Partner + Synergi Onboarding |
| Week 2 | Pilot Selection | Identify 3 pilot clients (mix: 1 simple 1040, 1 Schedule C, 1 complex); obtain verbal consent; brief pilot clients on what to expect; set up Synergi workflows for pilot use cases | Engagement Leads |
| Week 3 | Pilot Execution | Run 3 returns through AI-assisted workflow; CPA reviews each output against manual preparation; document time savings, error flags, and client feedback; capture side-by-side comparison data | Pilot CPAs + Operations |
| Week 4 | Refinement + Expand | Incorporate Week 3 feedback; expand to 10 clients (mix of tiers); refine prompts and workflows; finalize client disclosure and FAQ materials; prep external communications | Synergi Config + Marketing Lead |
| Week 5 | Public Launch | Send announcement email to full existing client base; post LinkedIn announcement; brief referral partners; open AI-assisted workflow to all new engagements | All Staff |
| Priority | Channel | Rationale | Timeline |
|---|---|---|---|
| 1 | Direct Email to Existing Clients | Highest trust, lowest cost, fastest conversion | Week 5 (ongoing) |
| 2 | Partner/Referral Network | Attorneys, advisors, bankers — each refers multiple clients | Week 5–6 |
| 3 | LinkedIn (Firm + Individual CPAs) | Professional credibility, thought leadership surface | Week 5–12 |
| 4 | Local Business Events / Chambers | In-person trust-building, referral network expansion | Month 2–3 |
| 5 | Website / SEO | Long-term lead generation, lower urgency | Month 2+ (ongoing) |
Why first: Highest volume, lowest complexity, easiest to validate AI output. AI performs most reliably on structured, predictable data.
Why second: Moderate complexity, but common enough that AI workflows can be standardized after Tier 1 refinement.
Why last: These require the most nuanced judgment and the most robust CPA review. AI assists with data extraction and formatting, but human oversight is highest here.
Goal: Understand client concerns and explain the approach before they need to ask.
Key principle: Clients who feel heard are clients who stay.
Goal: Demonstrate quality through the 3-return Proof-of-Value pilot.
Key principle: Proof closes more than promises.
Goal: Convert the engagement into an ongoing relationship where AI is a standard part of the service.
Key principle: The relationship expands when the CPA has time to think about the client's future, not just their past year's numbers.
| # | Objection | Response |
|---|---|---|
| 1 | "I don't trust AI with my taxes." | "We don't ask you to. Your CPA reviews every line of your return before it's submitted — AI just handles the data entry and initial organization. Nothing gets filed that your CPA hasn't personally reviewed and signed." |
| 2 | "Will my CPA still review everything?" | "Yes, without exception. Every return — regardless of complexity or how it was prepared — is reviewed and signed by a licensed CPA at our firm. That is not optional and will never change." |
| 3 | "Is my data safe?" | "Absolutely. We use enterprise-grade, SOC 2 compliant platforms. Your data is encrypted, never shared with third parties, and never used to train any AI model." |
| 4 | "Will this cost more?" | "No — AI-assisted preparation is either the same price or more competitive than our standard rates. Simple returns are $150. Complex returns are $350. There are no surprise fees." |
| 5 | "What if AI makes a mistake?" | "It can — which is exactly why every AI-generated output is reviewed by your CPA before anything is submitted. Our process is designed so that AI creates a draft and your CPA creates the final product." |
| 6 | "I've used you for 15 years. I don't want things to change." | "Nothing about who manages your account is changing. The only thing that changes is that your CPA has more time for the parts of your return that actually require their expertise." |
| 7 | "Can I opt out?" | "Yes, absolutely. If you'd prefer a fully manual preparation process, just let us know and we'll accommodate that." |
| 8 | "How is this different from TurboTax?" | "It's fundamentally different. TurboTax is a self-service tool with no professional oversight. We are a licensed CPA firm. AI is one tool we use — under full CPA supervision." |
| 9 | "Are you replacing staff?" | "No. We adopted AI to give our CPAs more time and capacity, not to reduce our team. Our 35-person firm is intact." |
| 10 | "How do I know the AI is accurate?" | "We have data. During our pilot, we ran AI-prepared returns side-by-side with manually prepared ones and documented the comparison. We're happy to walk you through that data." |
| Metric | Baseline | Target (Month 3) | Target (Month 6) |
|---|---|---|---|
| Returns processed/month | ~200 | 200 (same, optimized) | 250–300 |
| Average prep time (simple return) | 3–4 hrs | 1.5–2 hrs | 1–1.5 hrs |
| Average prep time (complex return) | 8–12 hrs | 4–6 hrs | 3–5 hrs |
| Gross margin | 35% | 50%+ | 60–67% |
| Client opt-out rate | N/A | <5% | <3% |
| CPA-reported satisfaction with workflow | N/A | 7+/10 | 8+/10 |
| Metric | Target (Month 1) | Target (Month 3) |
|---|---|---|
| Existing clients informed | 100% | 100% |
| Client disclosure document sent | 100% of active clients | 100% |
| Pilot clients completed | 3 | 10+ |
| LinkedIn announcement reach | 500+ impressions | — |
| Referral partner briefings | 5+ | 10+ |
| New prospects engaged via AI story | 5 | 20 |
| Risk | Likelihood | Impact | Mitigation |
|---|---|---|---|
| Client backlash to AI announcement | Low–Medium | High | Proactive communication, opt-out option, transparent FAQ |
| Staff resistance to new workflow | Medium | Medium | Early staff briefing, involve CPAs in pilot design, training sessions |
| AI output errors during pilot | Low | High | Full CPA review at every step; no AI output submitted unreviewed |
| Data security concern or breach | Very Low | Critical | SOC 2 compliant platform, encryption, access controls, insurance review |
| Slower-than-expected adoption | Medium | Low | Extended ramp timeline is fine; break-even is still reached within 6 months |
| Referral partner confusion | Low | Medium | Dedicated partner briefing packet and Q&A session |
| Pricing pushback | Low | Medium | Clearly communicated pricing tiers; AI pricing is competitive or better |
Five weeks to first client rollout is achievable. The timeline is intentionally aggressive enough to build momentum and deliberate enough to avoid quality failures.
Tier your rollout. Starting with simple 1040s protects the firm while building operational confidence.
Align, Prove, Partner is the sales motion. No client should be surprised by AI.
The objection handling table is a training tool. Every CPA-facing client should be able to answer these 10 questions fluently before Week 5 launch.
Internal buy-in precedes external launch. If your CPAs are not confident in the workflow, clients will feel it.
Metrics create accountability. Without KPIs, "it's going well" is not a performance signal.
This campaign package translates the positioning and go-to-market strategy from Documents 05 and 06 into ready-to-deploy marketing assets. Four campaigns cover the full stakeholder map: existing clients, new prospects, thought leadership, and referral partners. Total 90-day budget ranges from $5,000 to $8,000 — a modest investment for a firm making a significant operational leap. Most of the value here is not spend; it is execution. Done consistently over 90 days, these campaigns establish the firm as a trusted, modern practice and generate both retention and new business.
Objective: Ensure 100% of existing clients understand the AI adoption before they hear about it elsewhere, feel reassured about quality and their CPA relationship, and are invited to participate in the pilot if eligible.
Format: Email sequence, 3 touches over 3 weeks
Budget: $500 (email platform, design, staff time)
Subject line options:
Send timing: Week 5, Day 1 of public launch
Hi Sarah,
After 15+ years working with clients like you, we've always asked ourselves one question: how do we give our CPAs more time to actually think about your situation — not just fill out forms?
This year, we have an answer.
We've implemented Synergi's AI-assisted tax preparation platform at Castillo, Reeves & Associates. Here is what that means in plain terms:
• AI handles data extraction and initial formatting — the mechanical parts of return preparation
• Your CPA reviews every line, makes every judgment call, and signs every return
• Nothing about your account relationship changes
• Turnaround times get faster. Your CPA has more time for the conversations that matter.We know "AI" can sound unsettling when it's your financial information. That's exactly why we're being transparent about this from day one.
P.S. — We're offering a small group of existing clients the chance to experience AI-assisted preparation as a pilot this season. If you'd like to be considered, reply to this email and I'll follow up directly.
Send timing: Week 6 (7 days after Email 1)
Hi Sarah,
We received a lot of responses to our last email — thank you. The most common question was some version of: "But is AI actually doing my taxes?"
Here's our honest answer: partly, and deliberately.
What AI does: Reads your uploaded documents (W-2s, 1099s, statements), organizes the data, and creates a structured draft for your CPA to review.
What AI does not do: Make any final decisions. Sign anything. Substitute for professional judgment. Or access your data outside of your engagement with us.
What your CPA does: Reviews the AI's output line by line, applies professional judgment to every situation your return involves, makes all final determinations, and signs your return. This step is non-negotiable.
Think of it this way: AI is to your CPA what a GPS is to an experienced driver. It helps with navigation — but the driver is still making every real decision.
Send timing: Week 7 (14 days after Email 1)
Hi Sarah,
We've shared what AI-assisted preparation is and how it works. Now we'd like to invite you to experience it.
We're opening our AI-assisted workflow to existing clients for this tax season. Here's what you can expect:
• Same CPA. Same quality. Faster turnaround.
• Full transparency — you'll receive a summary of how your return was prepared
• Priority scheduling for pilot participants
• Pricing: Simple returns start at $150. Complex returns at $350.If you'd prefer to stick with our traditional manual process, that's completely fine — just let us know and we'll continue as always.
Either way, we're grateful for your trust.
Objective: Engage 20 targeted new prospects using the AI differentiation story as a conversation opener. Generate 5–8 qualified consultations within 90 days.
Format: LinkedIn connection + message sequence, email follow-up
Budget: $1,000 (LinkedIn premium, research, staff time)
Step 1 — Connection Request:
"Hi Michael — I'm a CPA at Castillo, Reeves & Associates and came across your profile through the Austin CPA Society. We work with a lot of business owners in the Austin area and recently made some significant investments in our practice that I think would be relevant to you. Happy to connect."
Step 2 — First Message (2–3 days after connection accepted):
"Thanks for connecting, Michael. Quick context: we're a 35-person CPA firm in Austin, TX and earlier this year we implemented AI-assisted tax preparation — which means faster turnaround and more time for our CPAs to focus on strategy rather than paperwork. We're opening up availability for a small number of new clients this season. If that's relevant to your situation, I'd love to have a 20-minute conversation."
Step 3 — Follow-up (7 days later if no response):
"Michael — following up briefly. I know timing matters. If now isn't the right moment, I'm happy to reconnect later in the year. I'll also share a short article we wrote about how we're thinking about AI in tax preparation — might be useful regardless of whether we work together."
Objective: Establish Castillo, Reeves & Associates as a credible voice on AI in accounting. Build organic reach, attract referral partners, and generate inbound prospect inquiries over a 90-day horizon.
Format: 2 articles (500–800 words each), 1 speaking engagement
Budget: $1,500–$2,500
Target audience: Existing clients, prospective clients, general public
Publish timing: Week 5 (coincides with launch)
Outline:
Target audience: Referral partners, peer CPAs, business owners
Publish timing: Week 8 (month 2 of launch)
Outline:
Target venue: Local CPA society, chamber of commerce, small business association, or financial advisors' meetup
Format: 20-minute presentation + 10-minute Q&A
Title: "AI-Assisted Tax Prep: What It Is, What It Isn't, and What It Means for Your Clients"
Timing: Month 2–3 (post-launch, with live data to share)
Objective: Brief 10+ referral partners on the firm's AI adoption and establish a structured referral arrangement with 5+ active partners within 90 days.
Audience: Attorneys (estate planning, business law), financial advisors, bankers, HR consultants, business coaches
Budget: $1,000–$2,000
| Referral Type | Incentive |
|---|---|
| New individual client (simple return) | $50 referral credit to partner's client or partner's choice |
| New business client (complex return) | $100 referral credit |
| Retained client after Year 1 | $25 annual retention credit |
| Reciprocal referral arrangement | Mutual referral without cash incentive (preferred for peers) |
| Campaign | Item | Estimated Cost |
|---|---|---|
| Campaign 1: Client Announcement | Email platform (90 days), design, staff time | $500 |
| Campaign 2: Prospect Outreach | LinkedIn Premium (2 seats, 3 months), research, staff time | $1,000 |
| Campaign 3: Thought Leadership | Writing support (2 articles), speaking prep, graphics | $1,500–$2,500 |
| Campaign 4: Referral Partner Program | Materials, events/coffees, incentive reserve | $1,000–$2,000 |
| Contingency (10%) | $400–$800 | |
| Total 90-Day Budget | $4,400–$6,800 |
| Week | Campaign | Channel | Action |
|---|---|---|---|
| 1 | Internal | All-staff | Briefing, Q&A, workflow training |
| 2 | Internal | Engagement leads | Pilot client selection and consent |
| 3 | Campaign 1 prep | Draft and approve 3-email sequence | |
| 4 | Campaign 3 prep | Article | Draft Article 1; LinkedIn post planned |
| 5 | Campaign 1 Email 1 | Send launch announcement to full client list | |
| 5 | Campaign 3 Article 1 | LinkedIn + website | Publish "Why We Chose AI-Assisted Tax Prep" |
| 6 | Campaign 2 | Begin prospect connection + message sequence | |
| 6 | Campaign 1 Email 2 | Send "What AI-Assisted Tax Prep Means for You" | |
| 6 | Campaign 4 | Phone + email | Begin referral partner personal outreach |
| 7 | Campaign 1 Email 3 | Send "Ready to Try It?" opt-in email | |
| 7 | Campaign 4 | In-person | Partner briefing meetings (5 targets) |
| 8 | Campaign 2 | Email follow-up to LinkedIn prospects (non-responders) | |
| 8 | Campaign 3 Article 2 | LinkedIn + website | Publish "Human Judgment + AI Precision" |
| 8 | Campaign 4 | Materials | Distribute co-branded materials to active partners |
| 9–10 | Campaign 2 | LinkedIn + Email | Continue prospect sequence; follow up consultations |
| 9–10 | Campaign 4 | Events | Attend 1–2 networking events with referral story |
| 11 | Campaign 3 | Speaking | Submit speaking proposal to local CPA society or chamber |
| 12 | All campaigns | Review | 90-day performance review; plan Q2 campaigns |
Sequence matters. Existing clients hear from you first, before they read about AI in a news article or hear about it from a competitor.
Thought leadership has compounding returns. Two articles and one speaking engagement in 90 days is achievable and generates durable credibility.
Referral partners are a multiplier. Five engaged partners can outperform 20 cold prospects.
Budget discipline. Keeping the 90-day campaign to $5,000–$7,000 means the program pays for itself within the first month of scaled operations.
Content is the connective tissue. Every campaign links back to Article 1 and the client FAQ.
Consistency over intensity. Three emails, two articles, one speaking slot, and 20 LinkedIn conversations — executed consistently over 90 days — outperforms a one-week burst followed by silence.
Deploying AI tools without preparing your people is one of the most common and costly mistakes firms make. This program ensures all 35 members of your team — from partners to administrative staff — are equipped to work alongside AI confidently, responsibly, and effectively. The curriculum is built around a single governing idea: AI is your junior associate. It handles the repetitive; your people handle the judgment. Over four weeks of structured training and ongoing monthly sessions, your staff will move from uncertainty to competency, and your firm will have a defensible, documented process for AI-assisted work.
AI does not replace CPAs. It removes the friction that prevents CPAs from doing their best work. The framing throughout this program is intentional: every agent in your Synergi environment is a junior team member. It needs direction. It needs review. It will occasionally be wrong. Your people provide the professional judgment, the client relationship, and the signature on the return.
| Component | Format | Duration |
|---|---|---|
| Core Training Modules (1–5) | Self-paced + live workshops | 4 weeks |
| Certification Assessment | Online + practical review | End of Week 4 |
| Ongoing Monthly Sessions | 1-hour live update | Monthly, indefinitely |
| Manager Check-ins | 1:1 or team debrief | Bi-weekly during rollout |
| Track | Participants | Focus |
|---|---|---|
| Partner/Senior CPA | Partners, Senior CPAs | Strategic oversight, sign-off authority, client communication |
| Preparer | Staff CPAs, associates | Workflow execution, AI-assisted prep process |
| Administrative | Admin, scheduling, intake | Document handling, client intake, tool navigation |
Timing: Week 1 | Duration: 2 hours | Format: Live workshop (all staff)
Timing: Week 1–2 | Duration: 3 hours | Format: Guided walkthrough + practice time
Timing: Week 2–3 | Duration: 4 hours | Format: Process walkthrough + supervised practice return
| Stage | Action | AI Role | Human Role |
|---|---|---|---|
| 1. Document Intake | Client uploads docs to secure portal | None | Confirm receipt, completeness check |
| 2. AI Extraction | Agent processes documents | Extracts income, deductions, entities | Verify extraction accuracy |
| 3. Draft Preparation | Agent populates return template | Drafts return based on extracted data | Review line by line |
| 4. Discrepancy Flagging | Agent highlights anomalies | Flags year-over-year changes, missing items | Investigate each flag |
| 5. CPA Review | Preparer reviews full draft | Available for questions | Final judgment on all line items |
| 6. Senior Review | Partner/senior CPA reviews | None | Signs off and approves for filing |
| 7. Client Communication | Engagement summary sent to client | Drafts summary for CPA review | CPA edits and sends |
| 8. Filing | Return submitted | None | CPA submits and retains records |
Timing: Week 3 | Duration: 2 hours | Format: Script review + role-play
"Is AI doing my taxes?"
"AI assists our team by handling the data extraction and initial organization — the same way a calculator assists with arithmetic. Your return is reviewed, verified, and signed by a licensed CPA. The judgment and the accountability are ours."
"Is my data safe?"
"Your data is processed within our private, encrypted environment. It is not shared with third-party AI services in a way that exposes your personal information. We can walk you through our data handling policy if you'd like."
"What if AI makes a mistake?"
"That's exactly why every AI-assisted return goes through a multi-stage CPA review before anything is filed. AI can and does make errors — which is why we treat its output as a starting point, not a finished product."
Timing: Week 3–4 | Duration: 3 hours | Format: Case review + documentation practice
Override is required when:
| Severity | Definition | Action |
|---|---|---|
| Low | Minor formatting or phrasing error | Correct and note in workpaper |
| Medium | Incorrect figure, correctable with source review | Override, log, flag for quality review |
| High | Potential regulatory violation or material misstatement | Escalate to Partner immediately, do not file |
| Level | Title | Who | Requirements | Authority Granted |
|---|---|---|---|---|
| Level 1 | AI-Aware | All 35 staff | Complete Modules 1–2, pass 20-question assessment (80% passing score) | Understands AI use at the firm; may interact with agents for general tasks |
| Level 2 | AI-Practitioner | Preparers, associates | Complete Modules 1–4, complete 1 supervised practice return, pass practical assessment | May prepare AI-assisted returns under senior review |
| Level 3 | AI-Reviewer | Senior CPAs, Partners | Complete all 5 modules, complete 2 supervised reviews, pass advanced assessment | May sign off on AI-assisted returns; authorized to approve client disclosures |
| Fear | What Staff Are Really Asking | How to Address It |
|---|---|---|
| "AI will take my job." | "Am I still valuable here?" | Emphasize that AI expands the firm's capacity — more returns, more advisory work, more growth. AI creates the bandwidth for staff to move up, not out. |
| "What if I rely on AI and it's wrong?" | "Am I protected?" | The override protocol and certification process exist precisely for this. Responsibility remains with the CPA. AI is a tool, not a colleague with a license. |
| "I'm not a tech person." | "Will I be able to learn this?" | Modules are designed for non-technical users. The interface is conversational. If you can write an email, you can work with these agents. |
| "Clients won't trust it." | "Will this hurt our relationships?" | The transparency scripts and disclosure process were designed to protect and strengthen client trust. We lead with honesty, not concealment. |
Training is not optional — all 35 staff complete the core curriculum before AI-assisted returns go live.
The "AI as Associate" framing reduces resistance and reinforces appropriate oversight.
Three certification levels ensure the right people have the right authority at each stage.
Monthly ongoing sessions prevent the program from going stale as the platform evolves.
Change management is built into the curriculum — fears are addressed proactively, not reactively.
Documentation requirements at every stage create an audit trail that protects the firm and its clients.
This roadmap defines the five-phase path from initial onboarding to scaled AI-assisted operations at your 35-person, $8M CPA firm. The sequence is deliberate: alignment before strategy, strategy before execution, proof before scale. No phase launches until the previous phase's decision gate is cleared. Your first AI-assisted client returns go live in Week 5. Full-scale operations — 50+ returns per month at 67% margin — are achievable by Month 6. The entire program is coordinated by 28 agents working across 6 department streams, activated in sequence according to the timeline below.
Duration: Week 1–2 | Purpose: Establish the foundation.
Deliverable: Complete Company Profile — validated and approved by firm leadership before Phase 1 begins.
| Agent | Role | Phase 0 Contribution |
|---|---|---|
| Higgins | Strategic Advisor | Facilitates readiness assessment, drafts initial values framework |
| Alfred | Chief of Staff | Documents all intake sessions, produces Company Profile draft |
Duration: Week 2–3 | Purpose: Translate the Company Profile into a strategic plan.
| Scenario | Monthly Returns | Monthly Revenue | Margin | Annual Incremental Margin |
|---|---|---|---|---|
| Base | 30 | $7,500 | 52% | +$145,000 |
| Best | 50 | $12,000 | 67% | +$180,000+ |
| Worst | 15 | $3,750 | 42% | +$40,000 |
Deliverable: Strategic Plan + Decision Artifact DEC-2026-014 — approved by partners before Phase 2 launches.
| Agent | Role | Phase 1 Contribution |
|---|---|---|
| Higgins | Strategic Advisor | Market analysis, strategic options, kill criteria |
| Jarvis | Executive Orchestrator | Cross-functional dependency mapping |
| Alfred | Chief of Staff | Decision Records, Communication Plan |
| Cameron | Finance Analyst | ROI model, scenario analysis |
| Morgan-L | Legal Advisor | Preliminary E&O and disclosure review |
Duration: Week 3–5 | Purpose: Activate all 6 department streams simultaneously. Deploy 28 agents.
| Stream | Lead Agent | Focus in Phase 2 |
|---|---|---|
| Marketing | Dakota | "Your CPA, Amplified" positioning, client communication rollout |
| Sales | Tatum | Proposal templates, pilot client identification, outreach sequencing |
| Finance | Marlowe | Budget tracking, margin baseline, forecast model |
| Legal | Morgan-L | Client disclosure finalization, E&O policy review |
| Operations | Riley-O | Workflow SOPs, technology integrations, staff scheduling |
| Support | Kris | Knowledge base build, escalation procedures |
Deliverable: Department Command Centers Live — all 6 streams operational, all agents active, all workflows running.
Duration: Week 5–8 | Purpose: Validate the AI-assisted workflow on 3 real client returns.
| Return | Type | Complexity | Why Selected |
|---|---|---|---|
| Return 1 | 1040 (individual) | Simple | Most common return type; validates basic workflow |
| Return 2 | Schedule C (sole proprietor) | Moderate | Tests business income handling and deduction classification |
| Return 3 | Partnership (1065) | Complex | Highest-risk return type; validates advanced extraction and review |
Deliverable: Proof-of-Value Report — documents pilot results, lessons learned, and go/no-go recommendation for Phase 4.
Duration: Week 8–12 | Purpose: Ramp from pilot volume to 20+ returns/month. Complete staff certification. Roll out client communications firm-wide.
Deliverable: Scaled Operations Playbook — updated SOPs, refined quality checklists, staff certification records, client communication templates.
Duration: Month 3–6 | Purpose: Reach 50+ returns/month, hit 67% margin, and begin expanding into advisory services.
Deliverable: Quarterly Business Review — Month 3 and Month 6 reviews with firm leadership, covering margin performance, client satisfaction, staff confidence, and next-phase opportunities.
| Milestone | Owner | Target Date | Status |
|---|---|---|---|
| Company Profile complete | Alfred | End of Week 2 | Pending |
| Values confirmed by leadership | Higgins | End of Week 2 | Pending |
| Strategic Plan approved | Jarvis, Partners | End of Week 3 | Pending |
| Decision Record DEC-2026-014 issued | Alfred | End of Week 3 | Pending |
| All 6 department streams active | Riley-O | End of Week 5 | Pending |
| All 28 agents deployed | Jarvis | End of Week 5 | Pending |
| Staff training modules 1–2 complete | Training Lead | End of Week 4 | Pending |
| Staff training modules 3–5 complete | Training Lead | End of Week 6 | Pending |
| Level 2 certifications issued | HR/Training | End of Week 5 | Pending |
| Level 3 certifications issued | HR/Training | End of Week 6 | Pending |
| Pilot Returns 1–3 filed | Preparer + Senior CPA | Week 6–7 | Pending |
| Proof-of-Value Report issued | Frankie, Marlowe | End of Week 8 | Pending |
| Client communication campaign launched | Dakota, Harper | Week 9 | Pending |
| 20+ returns/month achieved | Riley-O | Month 2 | Pending |
| 50+ returns/month achieved | Riley-O | Month 4 | Pending |
| 67% margin confirmed | Harley | Month 4–5 | Pending |
| Month 3 QBR delivered | Jarvis, Alfred | Month 3 | Pending |
| Month 6 QBR delivered | Jarvis, Alfred | Month 6 | Pending |
| Phase | Touchpoint | Frequency |
|---|---|---|
| Phase 0 | Onboarding kickoff call | Once |
| Phase 0 | Company Profile review | Once (end of Phase 0) |
| Phase 1 | Strategic Plan review call | Once (end of Phase 1) |
| Phase 2 | Weekly check-in with Implementation Lead | Weekly during Phase 2 |
| Phase 3 | Pilot review call (after each return) | 3x during Phase 3 |
| Phase 3 | Proof-of-Value debrief | Once (end of Phase 3) |
| Phase 4 | Scale review call | Bi-weekly |
| Phase 5 | Quarterly Business Review | Month 3 and 6 |
| Ongoing | Monthly platform update briefing | Monthly |
| Ongoing | Priority support channel (Slack or email) | As needed |
| Gate | Criteria | Who Decides |
|---|---|---|
| Phase 0 → Phase 1 | Company Profile approved; values confirmed | Firm Partners |
| Phase 1 → Phase 2 | Strategic Plan and business case approved; Decision Record issued | Firm Partners + Synergi |
| Phase 2 → Phase 3 | All 6 department streams confirmed ready; staff training scheduled | Riley-O + Firm Partners |
| Phase 3 → Phase 4 | All 3 pilot returns accurate; client satisfaction ≥ 90%; prep time reduction ≥ 40% | Senior CPAs + Synergi |
| Phase 4 → Phase 5 | Error rate ≤ manual baseline; staff confidence ≥ 80%; no unresolved compliance issues | Firm Partners |
The phased structure is not bureaucracy — it is risk management. Each gate exists to catch problems before they reach clients.
Five weeks to the first AI-assisted client return is aggressive but achievable with focused execution.
All 28 agents activate in Phase 2, but your team interacts with a smaller set until training is complete.
The Proof-of-Value pilot is your risk-free evaluation. Three returns before any long-term commitment is required.
Month 6 is where the economics fully materialize: 50+ returns/month at 67% margin represents $180,000+ in incremental annual margin.
Synergi is not a vendor that disappears after setup. Touchpoints are built into every phase.
This document outlines the complete investment required to implement Synergi at your 35-person, $8M CPA firm — and the return that investment is designed to generate. The recommendation is the Professional plan at $799/month (annual billing), combined with a one-time $10,000 implementation package. Total Year 1 investment on the annual plan: $19,588. Projected Year 1 return: $180,000+ in incremental margin. That is a 9:1 ROI with a break-even point in Month 3–4 of scaled operations. Before any annual commitment is made, you complete a three-return Proof-of-Value pilot. If it does not meet quality standards, no long-term contract is required.
| Feature | Professional Plan |
|---|---|
| Monthly price (month-to-month) | $999/month |
| Monthly price (annual commitment) | $799/month |
| Annual savings vs. monthly | $2,400/year (20% discount) |
| Users included | Up to 50 (covers all 35 staff) |
| AI models available | Claude (Anthropic), GPT-4o (OpenAI), Gemini (Google) |
| Context assets | Unlimited |
| Agent library | 40+ agents (full library) |
| Onboarding framework | Full Align 120 + Strategy 120 + Execute 120 |
| Research Studios | Included |
| Automated daily briefings | Included |
| Support tier | Priority support + guided onboarding |
| Service | Description | Investment |
|---|---|---|
| Implementation Package | Platform setup, data migration, integration configuration, custom agent configuration for tax prep workflow | $7,500 |
| Executive Training Session | Half-day workshop for partners and senior staff covering platform strategy, agent oversight, and sign-off authority | $2,500 |
| Total One-Time | $10,000 |
Before you commit to an annual subscription, you complete a three-return pilot.
Structure:
Quality Standards:
If the pilot does not meet all quality standards, no annual commitment is required. You may continue month-to-month, pause the engagement, or request a process revision before re-running the pilot. The Implementation Package fee is non-refundable (the work has been done), but the annual subscription commitment does not activate until the pilot clears.
| Term | Monthly Plan | Annual Plan |
|---|---|---|
| Monthly platform fee | $999/month | $799/month |
| Billing cycle | Monthly | Monthly (annual commitment) |
| Minimum commitment | None | 12 months after pilot completion |
| Annual cost | $11,988 | $9,588 |
| Cancellation notice | 30 days | 30 days (end of annual term) |
| Pilot period | Month-to-month until pilot complete | Month-to-month until pilot complete |
| Price lock | 12 months | 24 months (price guarantee) |
Begin on month-to-month billing during the pilot period (Phases 0–3). Once the Proof-of-Value pilot clears, convert to annual billing for the 20% discount. The annual commitment does not start until you have confirmed the platform delivers.
| Component | Annual Plan | Monthly Plan |
|---|---|---|
| Platform subscription | $9,588 ($799/mo x 12) | $11,988 ($999/mo x 12) |
| Implementation Package | $7,500 | $7,500 |
| Executive Training Session | $2,500 | $2,500 |
| Total Year 1 | $19,588 | $21,988 |
| Savings vs. monthly | $2,400 | — |
| Metric | Value |
|---|---|
| Year 1 projected incremental margin | $180,000+ |
| Total Year 1 investment (annual plan) | $19,588 |
| Year 1 ROI | 9:1 |
| Break-even point | Month 3–4 of scaled operations |
| Current margin on tax prep | 35% |
| Target margin at 50+ returns/month | 67% |
| Prep time reduction per return | 60% |
| Returns needed per month to hit target | 50+ |
| Synergi Professional | Build Internally | Do Nothing | |
|---|---|---|---|
| Year 1 cost | $19,588 | $150,000–$300,000 | $0 |
| Time to first AI-assisted return | 5 weeks | 12–18 months | Never |
| Ongoing maintenance | Included | Dedicated technical staff required | N/A |
| Agent library | 40+ pre-built, firm-configured | Must be designed and built from scratch | N/A |
| Staff training program | Included | Must be designed internally | N/A |
| Risk | Low (pilot before commitment) | High (build risk, timeline risk) | Low short-term |
| Opportunity cost of inaction | — | — | $200,000+/year in lost margin |
| 3-year total cost of ownership | ~$47,000 | $500,000+ | $0 direct cost |
| 3-year margin improvement | $540,000+ | $540,000+ (if it works) | $0 |
| 3-year net position | +$493,000+ | +$40,000 to +$90,000 | $0 |
The Professional plan at $799/month (annual) is the right fit. It covers all 35 staff, all 28 agents, and the full pipeline framework.
Total Year 1 investment on the annual plan is $19,588. That includes implementation, training, and platform access.
The Proof-of-Value pilot means you do not make an annual commitment until three real returns demonstrate the process works.
The ROI is 9:1 on the annual plan. Break-even is Month 3–4. The math is straightforward.
Building internally costs $150,000–$300,000 and takes 12–18 months. Doing nothing costs $200,000+/year in lost margin opportunity.
The annual plan saves $2,400 versus monthly billing and includes a 24-month price guarantee.
This appendix provides the complete reference for all 28 agents deployed in this engagement. It is intended as a working document — something your team will return to during onboarding, during the pilot, and as you scale. Each agent is listed with its role, its specific responsibility within this CPA firm engagement, and the phase in which it activates. The roster is organized by team: Executive (3), Cross-Functional (4), Marketing (6), Sales (8), Finance (3), Legal and Compliance (2), Operations (5), and Support (4). Understanding who does what — and when — will make your team more effective from Day 1.
| Agent | Role | Responsibility in This Engagement |
|---|---|---|
| Jarvis | Executive Orchestrator | Coordinates all 6 department streams; maps cross-functional dependencies; prepares weekly executive briefings summarizing progress, blockers, and decisions needed; flags tensions between streams |
| Alfred | Chief of Staff | Tracks all decisions and produces Decision Records; consolidates status updates from all streams into the weekly briefing for firm leadership; manages the Communication Plan |
| Higgins | Strategic Advisor | Produces market landscape analysis and competitive assessment; generates strategic options with structured upside/downside/reversibility scoring; defines kill criteria; available for on-demand strategic questions |
Jarvis is your go-to for "how are all the pieces fitting together?" Alfred is your go-to for "where does this decision live and who approved it?" Higgins is your go-to for "should we be doing this, and what are we missing?"
| Agent | Role | Responsibility in This Engagement |
|---|---|---|
| Kendall | Pricing Strategist | Models the economics of AI-assisted return pricing; validates $150/simple and $350/complex pricing against local market rates; identifies pricing opportunities for advisory service upsells |
| Cameron | Finance Analyst | Builds and maintains the ROI model; produces 3-year margin projections; runs base/best/worst case scenarios; tracks actual performance against projections during Phases 4 and 5 |
| Morgan-L | Legal Advisor | Reviews all client-facing disclosure documents for legal adequacy; assesses E&O implications; advises on state board requirements; serves as the compliance backstop for both the Legal stream and the broader engagement |
| Marley | Follow-Up Manager | Tracks all persistent action items across all phases; maintains the master follow-up log (23 items at engagement launch); sends reminders, flags overdue items, and escalates to Alfred when blockers emerge |
Central theme: "Your CPA, Amplified."
| Agent | Role | Responsibility in This Engagement |
|---|---|---|
| Dakota | Marketing Orchestrator | Coordinates all marketing activities; sequences campaigns to align with operational readiness; reports stream status to Jarvis weekly |
| Harper | Brand Voice | Establishes consistent messaging tone; defines the "Your CPA, Amplified" positioning; ensures all client-facing language reinforces the firm's values |
| River | Content Creator | Produces articles, email sequences, LinkedIn content, and client newsletter sections; drafts the AI disclosure cover letter; writes the FAQ document |
| Avery-M | ICP Adapter | Adapts messaging for three distinct audiences: existing tax prep clients, prospects, and referral partners |
| Blake | Competitive Intel | Monitors competitor AI adoption and positioning; identifies messaging gaps; alerts Dakota when a competitor makes a significant AI-related announcement |
| Emery | Brand Reviewer | QA gate for all client-facing materials; checks consistency with brand voice, accuracy of AI-related claims, and alignment with Legal's approved disclosure language |
| Agent | Role | Responsibility in This Engagement |
|---|---|---|
| Tatum | Sales Orchestrator | Coordinates all sales enablement and outreach activities; sequencing client communication rollout; reports pipeline status to Jarvis weekly |
| Drew | Proposal Writer | Drafts updated engagement letters for AI-assisted returns; creates proposal templates for advisory service upsells; produces opt-in consent documentation |
| Reese-S | Sales Enablement | Develops training materials, battle cards, and objection-handling guides for CPAs having client conversations about AI |
| Kieran | Account Research | Conducts client research; identifies which existing clients are most likely to embrace AI-assisted services |
| Remy | Outreach Creator | Writes email sequences and LinkedIn messages for the client communication campaign |
| Jules | Call Prep | Produces pre-call briefs; includes client history, known concerns, suggested talking points, and objection responses |
| Hayden | Pipeline Analyst | Tracks the pilot-to-conversion pipeline; monitors opt-in rates, consent form returns, and client satisfaction scores |
| Logan | Deal Strategist | Qualifies opportunities for advisory service expansion using the MEDDIC framework; identifies clients ready for advisory conversations |
| Agent | Role | Responsibility in This Engagement |
|---|---|---|
| Marlowe | Finance Team Lead | Oversees all financial modeling within the stream; produces the monthly financial summary for Alfred's executive briefing |
| Noel | Budget and Forecast | Maintains the engagement budget; tracks actual spend; updates forward forecast as volume ramps |
| Harley | Revenue Ops | Tracks revenue performance by return type; calculates realized margin against the 67% target; produces the margin report |
| Agent | Role | Responsibility in This Engagement |
|---|---|---|
| Morgan-L | Legal Advisor | (Also cross-functional) Leads all regulatory review; finalizes client disclosure document and engagement letter language; advises on E&O coverage implications |
| Riley | Compliance Auditor | Conducts IRS compliance review of the AI-assisted workflow; audits state board requirements; reviews data handling procedures; produces the Compliance Audit Report |
| Agent | Role | Responsibility in This Engagement |
|---|---|---|
| Riley-O | Operations Lead | Oversees the operational rollout across all five phases; owns the master project timeline; produces the go/no-go recommendation at each decision gate |
| Sage-O | Vendor Manager | Manages technical relationships with all technology vendors; owns the integration testing process |
| Nico | Resource Planner | Manages staff allocation and capacity planning during the ramp; identifies when additional bandwidth is needed |
| Kai | Process Analyst | Conducts workflow optimization analysis; reviews override logs and error patterns; produces process improvement recommendations |
| Devon | SLA Monitor | Tracks service level metrics: return turnaround time, CPA review time, client satisfaction scores, error rates |
| Agent | Role | Responsibility in This Engagement |
|---|---|---|
| Kris | Knowledge Manager | Builds and maintains the engagement knowledge base: SOPs, FAQ documents, training materials, override logs, and lessons learned |
| Toni | Escalation Handler | Manages exceptions and complex cases; acts as the first escalation point for preparers encountering unusual situations |
| Dallas | Policy Tuner | Refines AI agent policies based on real-world pilot feedback; adjusts agent instructions to prevent recurring errors |
| Frankie | Quality Reviewer | QA gate for all AI-assisted outputs before CPA review; checks draft returns for completeness and mathematical accuracy; produces quality metrics reports |
| Stream | Lead Agent | Total Agents |
|---|---|---|
| Executive | Jarvis | 3 |
| Cross-Functional | Marley (Follow-Up) | 4 |
| Marketing | Dakota | 6 |
| Sales | Tatum | 8 |
| Finance | Marlowe | 3 |
| Legal and Compliance | Morgan-L | 2 |
| Operations | Riley-O | 5 |
| Support | Kris | 4 |
| Total | 28 unique agents |
| Phase | Agents Activated | Total Active |
|---|---|---|
| Phase 0: Align 120 | Higgins, Alfred | 2 |
| Phase 1: Strategy 120 | + Jarvis, Cameron, Morgan-L, Kendall | 6 |
| Phase 2: Execute 120 | + All remaining 22 agents | 28 (all) |
| Phase 3: Proof-of-Value | All 28 active; Riley-O, Kai, Kris, Frankie, Toni, Marlowe, Noel, Riley, Hayden most active | 28 |
| Phase 4: Scale | All 28 active; Dakota, Harper, River, Tatum, Remy, Harley, Devon most active | 28 |
| Phase 5: Optimize | All 28 active; Harley, Kai, Dallas, Frankie, Logan, Tatum most active | 28 |
Agents activate when their function is needed — not before. Activating all 28 agents on Day 1 before the Company Profile is complete would generate noise, not value. The phased activation ensures each agent has the context it needs to be useful when it goes live.
28 agents work across 8 teams to execute this engagement. No single person needs to manage all of them — Jarvis and Alfred do that.
The Executive team (Jarvis, Alfred, Higgins) is your primary interface for strategy, coordination, and documentation.
Morgan-L's dual role (cross-functional and Legal stream lead) reflects how compliance works in practice — it is never siloed.
The activation timeline is deliberate. Agents that go live before their function is needed create noise. Agents that activate at the right moment create value.
The Support stream (Kris, Toni, Dallas, Frankie) is the continuous improvement engine. Their work intensifies as volume scales and patterns emerge.
Sales has the largest stream (8 agents) because client adoption is the rate-limiting factor in scaling.
Schedule a 30-minute walkthrough to review the ROI model with your actual numbers and identify your ideal pilot candidates.
Talk to Us