Your CAS Practice Has Six Workflows to Automate — Here's the Tool List

Published: March 15, 2026 | By: The Crossing Report | 7 min read

---

Summary

The Journal of Accountancy (January 2026) published a practitioner guide for accounting firms building out Client Advisory Services using AI — six specific workflow categories with named tools. This is the most concrete, tool-specific AI implementation guide published for small accounting firms this year. Here's what each workflow does, which tools do it, and where to start.

---

The Problem With "Building a CAS Practice"

Every accounting firm consultant says the same thing: shift from compliance to advisory. Build a CAS practice. Stop doing just tax returns and start telling clients what to do.

What they rarely say is exactly how.

"Client Advisory Services" is a broad category. It can mean cash flow analysis, financial forecasting, business benchmarking, scenario modeling, or just monthly calls where you tell clients what their numbers mean. The work that actually pays — the deliverables clients value enough to pay advisory rates for — requires pulling and synthesizing financial data that takes significant staff time to compile manually.

That's the bottleneck. And it's exactly what AI solves.

The Journal of Accountancy's January 2026 practitioner guide identifies six CAS workflow categories where AI tools have measurably reduced that bottleneck for small accounting firms. Not "consider AI" — six specific categories with named tools and documented results.

---

The Six Workflows

1. Client Onboarding

The problem: Collecting engagement documents, financial history, and data access from new clients is high-touch, slow, and often bottlenecks the beginning of every new engagement.

The tools: FloQast (engagement onboarding workflows) and Microsoft Power Automate (automated document request sequences and reminders).

What it does: Automates the request-and-collect cycle — sends document requests, tracks responses, follows up automatically, and routes completed documents to the right people. New client intake that used to take 2-3 weeks of back-and-forth can compress to a week or less.

Who starts here: Firms with high new-client volume (20+ new engagements per year) or those where onboarding is a staff time drain that delays getting to billable work.

---

2. Spend and Expense Management

The problem: For clients using expense management platforms, transaction data is messy — inconsistently categorized, lacking documentation, and requiring substantial cleanup before it's useful for financial analysis.

The tools: BILL (invoice processing, AP automation, expense extraction) and Ramp (expense categorization, policy enforcement, month-end automation).

What it does: Automates invoice extraction and data entry, enforces expense categorization policies, flags policy violations, and prepares clean data for bookkeeping. Ramp's Accounting Agent (launched February 2026) reports 90%+ auto-coding accuracy and 3x faster month-end close for early customers.

Who starts here: Firms managing clients who use Ramp or BILL already — the integration is frictionless. Also strong for firms where staff spends significant time cleaning up client expense data before they can do any analysis.

---

3. Bookkeeping

The problem: Client-specific transaction categorization is the highest-volume, most repetitive work in a CAS practice — and the work that makes or breaks the quality of everything built on top of it.

The tools: Keeper (AI-powered bookkeeping with client communication built in) — the JoA guide also referenced Botkeeper, though Botkeeper shut down in February 2026 and was acquired by Xendoo. For firms evaluating this category now, Keeper and Xendoo (post-Botkeeper) are the two relevant options.

What it does: Automates transaction categorization using client-specific rules, flags exceptions for human review, and handles the client communication around uncategorized or unusual transactions. The human review step remains — but it focuses effort on the exceptions rather than every transaction.

Who starts here: Firms with high bookkeeping volume (10+ clients with active books) where staff spends 60%+ of their time on categorization rather than analysis.

---

4. Client Reporting

The problem: Monthly financial reporting for advisory clients — statements, commentary, benchmarking summaries — is customized per client but structurally similar. It's high-skill work to format correctly and takes 2-4 hours per client per month to assemble manually.

The tools: Reach Reporting (auto-generated financial statements with AI-drafted narrative commentary) and Qvinci (multi-client consolidated reporting with comparison benchmarks).

What it does: Pulls data directly from accounting systems, generates formatted reports in the client's preferred format, and drafts initial narrative commentary on variances and trends. The accountant reviews and edits rather than building from scratch.

Who starts here: Almost every CAS firm. Reporting automation is the fastest return because the time savings are immediate, the output is visible to clients, and the review protocol is straightforward (check the numbers, edit the narrative).

---

5. Business Insights and Scenario Modeling

The problem: The analysis clients actually pay for — scenario modeling, variance explanations, forecasting, benchmark comparisons — requires synthesizing data across time periods and making interpretive judgments. This is the work that's hardest to scale.

The tools: Datarails (FP&A automation: budget-to-actual analysis, rolling forecasts, scenario models directly in Excel).

What it does: Automates the data aggregation and comparison work that underlies scenario modeling — pulling actuals vs. budget, generating variance analysis, running "what if" scenarios — so the accountant spends time on interpretation rather than data assembly.

Who starts here: Firms where partners spend 4+ hours per client per month on financial analysis that could be automated. Also firms trying to move upmarket into CFO-level advisory work — Datarails is the tool category that enables the conversation.

---

6. Client Communication

The problem: The back-and-forth with clients — answering questions, following up on document requests, summarizing meeting outcomes — is time-consuming and often falls to senior staff who have more important uses for their attention.

The tools: Gmail Smart Reply and Microsoft Copilot (AI-assisted email drafting and follow-up).

What it does: Drafts routine client responses, generates meeting summaries, and composes follow-up emails based on prior context. Senior staff review and send rather than composing from scratch.

Who starts here: Firms where client communication volume is high and where a partner or senior accountant is spending 1-2 hours per day on email that doesn't require their expertise.

---

Where to Start

The JoA guide and practitioner follow-up data consistently point to the same starting sequence for small accounting firms building CAS:

Start with reporting (Workflow 4), then bookkeeping (Workflow 3), then insights (Workflow 5).

The logic: reporting automation produces visible results for clients immediately, which justifies the transition to advisory pricing. Bookkeeping automation cleans the data foundation that reporting and insights depend on. Insights automation is where the real advisory margin lives — but it requires the data quality that the earlier workflows establish.

Onboarding (1), expense management (2), and communication (6) are real wins, but they're support infrastructure rather than the core of the CAS value proposition. Build those in parallel once the core three workflows are running.

---

The Pricing Implication

This is the part most CAS conversations skip.

Automated reporting and bookkeeping don't just save staff time — they change the economics of what you can profitably offer at what price.

When a monthly reporting package requires 4 hours of staff time to assemble, it's hard to price under $2,000/month per client while maintaining margin. When the same package requires 45 minutes of review after AI assembly, $750-$1,200/month becomes viable — and you can serve more clients with the same headcount.

That pricing compression is both an opportunity (you can compete for clients who couldn't previously afford advisory services) and a competitive pressure (other firms are making the same calculation). The accounting firms raising prices in 2026 — 80% of US accounting firms, per the Ignition 2026 pricing benchmark — are doing so because AI is improving their margins, not despite it.

---

Your Next Step

Pick one of these six workflows and map it against your current practice:

• If you have monthly reporting work: Sign up for Reach Reporting's trial (reachreporting.com) and import one client's QuickBooks data. Run their last three months of reporting through the AI commentary feature. Time how long it takes compared to your current process.

• If bookkeeping cleanup is your biggest time drain: Evaluate Keeper (keeper.app) — they offer a demo with your own chart of accounts and client data. The test question: what percentage of transactions would have been categorized correctly without human input?

• If you're already running clean data and want to move upmarket: Request a Datarails demo focused specifically on budget-vs-actual and rolling forecast use cases. Ask them to show you a model for a client your size.

The CAS practice you're trying to build is already running at firms similar to yours. The tools are accessible. The question is which workflow you start automating this quarter.

---

The Crossing Report covers AI adoption for professional services firm owners every Monday. Subscribe here.

---

Related Reading

• Stuck in Pilot Mode? The 4-Step AI Adoption Plan for Accounting Firm Leaders
• Best AI Tools for Small Accounting and Law Firms (2026)
• Basis AI: Accounting Unicorn Raises $87M — What It Means for Small Firms
• AI Can Now Read Your Client's Entire Financial Year in One Session — Here's How to Use It
• PwC Expects End-to-End AI Audit Automation by Year-End — 3 Moves for Small CPA Firms
• Finance Platforms Are Going 'Bring Your Own AI' — What This Means for Your Accounting Practice