One Accounting Firm Cut Tax Return Time by 87% — Here's the Workflow They Used
One Accounting Firm Cut Tax Return Time by 87% — Here's the Workflow They Used
It's tax season. Your team is doing what your team does every April: manually entering W-2 data, cross-referencing 1099s, chasing clients for missing documents, and opening returns that look exactly like they did in 2024.
Meanwhile, LBMC — a Southeast US accounting and advisory firm — is spending 30 minutes per return.
Not 30 minutes total. Thirty minutes of review time. The data entry, document processing, and initial field population happen automatically, before a CPA opens the file.
Get the full picture. Go premium.
Weekly intelligence briefings, deeper analysis, and direct access to the full archive.
That's down from four hours.
If your first reaction is skepticism, that's reasonable. But the workflow exists, it's running at real firms right now, and the gap between what you're spending per return and what firms like LBMC are spending is widening every tax season you wait.
Here's how it works, who it's for, and what you need to know before deciding whether it's worth implementing.
The 4-Hour Return: Why It Takes That Long
The four hours per return that LBMC was spending before AI automation wasn't inefficiency — it was the standard workflow. A typical individual return requires:
- Receiving and organizing client documents (W-2s, 1099-DIV, 1099-INT, 1099-B, K-1s, mortgage interest statements, charitable contribution records)
- Manually reviewing each document for completeness
- Entering or mapping data into your tax software
- Cross-checking totals, catching mismatches, flagging missing items
- Generating workpapers and linking source documents for review
That's work. Necessary work. But most of it is document processing, not accounting judgment. And document processing is exactly where AI automation has the clearest ROI.
The Tool: SurePrep
SurePrep is a tax workflow automation platform now owned by Thomson Reuters, which acquired it in 2023 for $500 million. That acquisition tells you something: this isn't a startup experiment. Thomson Reuters, which also owns UltraTax, is building AI tax automation into the core infrastructure of accounting workflows.
The platform has two primary components relevant to this case study:
TaxCaddy handles document collection. Instead of email attachments and dropped-off folders, clients upload their documents through a secure portal. The system auto-categorizes what comes in — a W-2 goes in the W-2 bucket, a 1099-DIV goes in its own — and tracks what's missing against a checklist you define.
1040SCAN handles the automation. It uses AI and machine learning to scan every uploaded document, extract the relevant data, and populate your tax software automatically. It integrates with UltraTax CS, Lacerte, Drake, ProSeries, and CCH Axcess, so you're not switching your core platform.
The output: a return that arrives in your hands partially completed, with all source documents linked and an audit risk summary generated. What's left for the CPA is judgment work — reviewing the populated return, catching exceptions, handling complex items, and making the calls that require a licensed professional.
That's the 30 minutes.
The Math for a Small Firm
If you're running a 10-person accounting practice processing 400 individual returns per year, the capacity math is straightforward.
At four hours per return, your team spends 1,600 hours on individual return processing annually. At 2,000 working hours per year per person, that's 0.8 full-time equivalents — almost one entire person — doing document processing.
At 30 minutes per return, that same 400 returns takes 200 hours. You've recovered 1,400 hours.
What do you do with 1,400 hours? Three options:
- Serve more clients without adding headcount. At four hours per return, a 10-person firm has a hard ceiling on return volume. At 30 minutes per return review, you can process significantly more.
- Move your team up the value chain. The recovered hours don't have to go into more returns — they can go into advisory work: tax planning conversations, cash flow analysis, business advisory services that bill at higher rates and deepen client relationships.
- Reduce stress during tax season. Smaller firms often lose staff to burnout or to larger firms that have invested in automation. A practice that runs April without forced overtime retains staff better.
What SurePrep Doesn't Solve
Three honest limitations before you get too far down this path.
Complex returns still require real time. 1040SCAN excels at high-volume individual returns with standard document types. Partnership returns with complex allocations, returns with significant capital gain complexity across hundreds of transactions, or multi-state returns with unusual sourcing issues still require substantial senior judgment. LBMC's 30-minute figure likely reflects their streamlined individual return workflow, not their most complex engagements.
Document chaos is a client problem, not a tech problem. SurePrep's TaxCaddy portal is excellent at organizing documents once clients send them. It does not solve clients who send documents in three batches over six weeks, who lose their 1099-B until February 28, or who forget to mention they started a rental property. Client behavior is still a process management challenge.
The review matters more than ever. Reducing review time to 30 minutes only works if the reviewer is trained to catch what the AI missed. Firms that implement AI automation and then underinvest in review quality are moving risk, not removing it. The malpractice exposure from a misclassified 1099 that slips through 30-minute review is the same as it would have been from a misclassified 1099 that slips through four-hour review.
Is This Worth Implementing Before This Filing Season Is Over?
Probably not for April 15. That ship has sailed for most firms.
But April 15 is not the end of tax season — it's the beginning of extension season. Firms that file Form 4868 or 7004 for clients have until October 15. That's six months of extended returns that could run through AI-assisted workflows if you start implementation now.
And for the 2027 filing season, the implementation window is this spring and summer.
SurePrep implementation typically takes 4-8 weeks for initial setup, integration with your existing software, and staff training. Firms that start in May or June are operational for the September-October extension crunch.
Where to Start This Week
If you want to evaluate whether AI tax automation makes sense for your practice, start with one question: How many individual returns does your firm process per year, and what is your current average hours per return?
If your number is above 150 returns and more than two hours per return average, the ROI conversation is worth having. Contact Thomson Reuters about a SurePrep demo — the product page is at corp.sureprep.com.
If your number is below that threshold, your priority isn't document automation yet — it's either growing volume or identifying higher-leverage AI investments (client communication drafting, advisory memo generation, engagement letter automation) that pay off at lower return counts.
The LBMC number is real. The workflow exists. The question is whether it fits your firm's profile right now — and if not now, when.
Source: GiaSpace, "5 AI Tools Every Accounting Firm Should Use in 2026" (February 5, 2026). LBMC is a Tennessee-headquartered accounting, HR, technology, and security advisory firm.
Frequently Asked Questions
How did LBMC cut tax return time from 4 hours to 30 minutes?
LBMC implemented SurePrep's AI-powered tax document automation platform alongside Intuit Assist. SurePrep's 1040SCAN tool automatically scans and classifies incoming client documents (W-2s, 1099s, K-1s), extracts the data, and populates the return fields — all before a CPA opens the file. What previously required manual data entry and cross-referencing now arrives pre-populated with audit risk flags already generated. The 30 minutes remaining represents senior review time, not rework. According to GiaSpace's 2026 analysis, firms using SurePrep in combination with AI-assisted review tools typically see 40-70% reduction in per-return time, with LBMC representing the high end of that range.
Is SurePrep right for a small accounting firm (under 20 employees)?
SurePrep is used by firms ranging from sole practitioners to Big Four, but the ROI calculus depends on volume. For a firm processing fewer than 150 returns per year, the setup and subscription cost may not justify the time savings. For firms doing 300+ individual or business returns per year, the math typically works: if SurePrep saves 2-3 hours per return at an average billing rate of $75-100 per hour, a 500-return practice saves $75,000-$150,000 in staff time annually. The platform integrates with major tax software including UltraTax, Drake, Lacerte, and ProSeries, so implementation doesn't require switching your core tax platform.
What does SurePrep actually do, step by step?
The workflow has three stages. First, document collection: clients upload their documents through SurePrep's TaxCaddy portal, which replaces email back-and-forth and the paper pile on your desk. Second, automated processing: 1040SCAN uses AI and machine learning to scan, classify, and extract data from every document — W-2s, 1099s, brokerage statements, K-1s, real estate forms — and maps the extracted data to the correct fields in your tax software. Third, review handoff: the system generates a completed return workpaper with all source documents linked, plus an audit risk summary, ready for CPA review. The 30-minute review LBMC achieved is what's left after stages one and two run automatically.
What does SurePrep cost for a small accounting firm?
SurePrep's pricing is not publicly listed and is negotiated based on firm size and return volume. Thomson Reuters, which acquired SurePrep in 2023 for $500 million, positions it as an enterprise tool but serves practices of all sizes. Industry sources suggest that pricing for small firms (under 500 returns annually) typically runs $5,000-$15,000 per year depending on volume and modules selected. The ROI breakeven for a firm billing at $85/hour in staff time is reached at roughly 100-200 returns, depending on current manual prep time per return.
What are the limitations of AI tax automation for small firms?
Three honest limitations. First, complex returns require more review time — AI handles high-volume individual returns well, but partnership returns, complex trust returns, and multi-state filings with unusual fact patterns still require significant senior judgment. Second, client document chaos is not fully solved — AI can classify documents once they arrive, but getting clients to upload complete documentation (rather than sending partial batches over three weeks) remains a human process management problem. Third, errors in AI-extracted data require trained review — the 30-minute review that LBMC achieved requires a reviewer who knows what to look for. Firms that reduce review time without maintaining quality oversight increase, not decrease, their malpractice exposure.
Get the weekly briefing
AI adoption intelligence for accounting, law, and consulting firms. Free to start.
Free weekly digest. No spam. Unsubscribe anytime.
Related Reading
This is the kind of intelligence premium subscribers get every week.
Deep analysis, cross-sector patterns, and the frameworks that help professional services firms make the crossing.