How do you start a tax preparation business in 2027?

Two ways in: the volume storefront versus the credentialed boutique

Nearly every tax preparation business that survives past year three resolves into one of two shapes, and the founder who refuses to choose usually ends up doing both badly. The first shape is the volume storefront. You rent visible retail space, you staff it with seasonal preparers, you offer bank products — refund transfers, pay-by-refund fee collection, refund advances — and you win on foot traffic and turnaround speed. The average fee per return is low, somewhere in the $200 to $350 band for a wage-earner return with a state attached, but the count is high. Six hundred returns at $275 is $165,000 of gross in eleven weeks. The model lives or dies on throughput per preparer-hour and on how cheaply you can acquire a return.

The second shape is the credentialed boutique. You work from home or a small suite, you take almost no walk-ins, and every client arrives by referral or by search for something specific — "enrolled agent multi-state," "CPA for short-term rental owners," "crypto tax preparer." Average fee per client runs $600 to $2,500 and up, counts stay under 200, and much of the revenue is not the return at all. It's the quarterly estimated-payment work, the S-corp election analysis, the reasonable-compensation study, the IRS notice you resolve in August. Same industry, almost nothing in common operationally.

The comparison matters because the two models fail differently. The storefront's exposure is demand-side: the simple return is being commoditized from underneath by TurboTax and H&R Block's online product, by IRS Direct File's continued state and form-type expansion, and by AI-assisted preparation tools that handle Schedule A and B conversationally. Every year, a slice of the storefront's addressable book leaves for a $39 to $129 DIY option that takes half an hour. The storefront responds by cutting price, which compresses the only variable that made the model work.

The boutique's exposure is supply-side and personal. You are the product. Your capacity ceiling is your own hours plus whatever you can genuinely supervise, and complex work does not delegate cleanly to a seasonal hire. A partnership return with three K-1s, basis tracking, and a state apportionment question is not a task you hand to someone you met in November. The boutique fails by burning out, by taking one client too complicated for its expertise, or by never building the year-round revenue that carries May through December.

There is a third shape worth naming even though it's really a variant: buying an existing book. Retiring preparers sell client lists regularly, typically at some multiple of annual revenue negotiated privately, often with an earnout tied to retention through the first season under new ownership. You inherit relationships, a filing history, and a software conversion headache. For someone with credentials but no marketing appetite, acquisition beats a cold start — though you should read every engagement letter and check for open IRS notices before you sign anything.

How to decide which model fits you

The decision is not primarily about capital. It's about which constraint you're willing to live inside. Ask four questions in order, and answer them honestly rather than aspirationally.

First: do you already have the credential, or are you starting the study clock now? The Enrolled Agent path runs through the three-part Special Enrollment Examination, and a working adult typically needs six to twelve months to pass all three parts. If you're twelve months from a credential, the boutique model is not available to you yet — you cannot sell representation you don't have. The Annual Filing Season Program gets you a Directory listing and limited representation rights for continuing-education hours rather than an exam, which is a legitimate bridge into a first season.

Second: where does your client flow actually come from? Be specific. "Referrals" is not an answer; "my brother-in-law manages twelve rental properties and knows forty other landlords" is. If you cannot name the first fifteen clients, the storefront's foot traffic is doing work your network can't do yet — but you're also paying rent for that privilege before you know whether the location converts.

Third: what does your cash position tolerate? A home-based solo launch lands roughly $11,000 to $15,000 all-in for the first year. A virtual boutique with better software and real marketing runs $15,000 to $25,000. A storefront adds $15,000 to $30,000 in lease, buildout, signage, and seasonal payroll before a single return is filed, pushing total first-year exposure to $35,000 to $50,000. That's not a small difference when 70 to 80 percent of your revenue arrives inside an eleven-week window.

Fourth: what do you want to be doing in year five? The storefront is a management business — you'll spend your time on hiring, scheduling, quality control, and lease negotiation. The boutique is a craft business — you'll spend it on tax research and client conversations. Both are respectable. They are not interchangeable, and people routinely pick the one that doesn't match their temperament because the spreadsheet looked better.

The numbers behind each path

Start with the fixed costs, because they're the same regardless of model and they're small enough that people underestimate the rest. A Preparer Tax Identification Number renewal runs $19.75. State registrations vary widely — a handful of states, including California, Oregon, Maryland, and New York, impose their own preparer registration or licensing regimes on top of the federal PTIN, while most impose none. Budget roughly $200 for PTIN plus EFIN plus state paperwork combined.

Software is the first real decision. Drake Tax has historically been the independent preparer's default at roughly $1,800 to $2,000 for an unlimited 1040-plus-business package, which makes it cheap per return once you clear about 60 returns. Intuit's ProConnect is priced per return, which inverts the math: excellent for a first season at 40 returns, expensive at 400. Lacerte and Thomson Reuters UltraTax CS sit at the firm end, with CCH Axcess serving larger practices. The right choice depends almost entirely on your expected count and whether you need multi-state and entity modules. Confirm current pricing directly with the vendor — these figures move year to year and vendors restructure tiers regularly.

Errors-and-omissions insurance runs roughly $400 to $1,200 annually for a $1 million policy for a solo preparer. This is the line item new operators skip and should not. A single missed estimated-payment schedule or an unreported K-1 line generates penalty and interest exposure that dwarfs the premium, and the client who receives an IRS notice does not care that it was an honest error. Pair the policy with engagement letters that define scope, fee, document-retention obligations, and what happens if the client hands you incomplete records in April.

Now the revenue side, by model. A solo EA filing 200 returns at a $400 blended average grosses $80,000. Subtract $2,000 software, $800 insurance, $2,500 marketing, and $3,000 in miscellaneous overhead and you're netting somewhere in the $50,000 to $60,000 range before self-employment tax — respectable for year two, thin for year one when you'll likely file 60 to 100. A two-preparer storefront at 600 returns can gross $200,000 to $300,000, but seasonal payroll, rent, and bank-product costs consume a large share, and your net margin is a management outcome rather than a pricing outcome.

The boutique math looks different because the fee schedule does. Schedule A and B individual returns support $200 to $600. A Schedule C sole proprietor with depreciation and a home office supports $500 to $1,500. A multi-state S-corp or partnership return with K-1s supports $1,500 to $5,000. Add a monthly bookkeeping retainer at a few hundred dollars a month and a quarterly planning engagement, and 80 to 150 clients can produce revenue that a 600-return storefront cannot, with a fraction of the headcount.

The variable nobody models correctly is seasonality. Roughly 70 to 80 percent of a tax preparation firm's revenue lands between late January and April 15. That means October is where solo operators run out of money — not because the business failed, but because they spent April's cash in June. Reserve a specific percentage of every spring dollar into a separate account before you touch it, and treat the reserve as non-negotiable operating capital rather than profit. The firms that reach year three almost always did one of two things: built a year-round book of bookkeeping, advisory, notice resolution, and quarterly work, or ran a genuinely disciplined reserve. Most did both.

Adjacent to this, if you're evaluating a services business generally, the same seasonality shape shows up in landscaping, tax-adjacent bookkeeping, and event-driven trades. The countermeasure is identical everywhere: convert episodic transactions into recurring commitments. In RevOps terms, you're moving from a one-time-sale motion to a retention motion, and the metric that matters shifts from returns filed to revenue per client per year.

Sequencing the launch so nothing blocks in January

The single most common failure in starting a tax preparation business is doing the right things in the wrong order. Software and office space feel like progress because you can see them; credentials and IRS applications don't, and they're the ones with hard lead times. Work backward from the third week of January.

Begin with the PTIN. It's a same-session online application through the IRS, it's required for every paid preparer, and nothing downstream works without it. Do it the week you decide, not the week before the season.

Then start the EFIN application, because it is the long pole. Becoming an Authorized e-file Provider runs through IRS e-Services and takes roughly 45 days, with fingerprinting required for applicants who aren't already credentialed as an EA, CPA, or attorney. Suitability checks can extend that. Starting the application in August for a January launch is comfortable; starting in November is a gamble; starting in January means you're paper-filing your first season or working under someone else's EFIN, which introduces its own problems.

Credential study runs in parallel with everything, not after it. The Special Enrollment Examination has three parts — individuals, businesses, representation and practice procedures — and the parts can be taken in any order and passed across a window rather than in one sitting. Enrolled Agent status carries unlimited representation rights before the IRS with no state board involvement, which is why it's the highest-leverage credential for an independent preparer who doesn't need attest work. The CPA track requires state licensure, 150 credit hours, and the Uniform CPA Examination, and it opens doors the EA doesn't — but it's a multi-year commitment, not a launch step.

Entity formation and insurance come next, and they're quick. An LLC filing plus an operating agreement plus a bank account is a week of work. Whether to elect S-corp treatment is a question of profit level, not of launch timing; it's easier to make that election in year two once you know what the business actually earns. The Small Business Administration's guidance on structure selection is a reasonable starting point, but the election analysis is exactly the kind of work you'll eventually sell to clients, so it's worth doing carefully on your own books first.

Software selection should happen after you have a defensible count estimate, which means after you've named your first fifteen to twenty-five clients. Demo two packages, run a prior-year return you already know the answer to through both, and time yourself. Speed on a familiar return is a decent proxy for speed on unfamiliar ones. If you serve refund-advance or pay-by-refund clients, bank-product enrollment is a separate signup with its own timeline — handle it in the fall.

Marketing is last in sequence and first in importance, which is the paradox that trips people up. Tax preparation is a search-and-referral business, not a paid-advertising one. A Google Business Profile, a site that names your niche in the title tag, listings in the National Association of Tax Professionals and National Association of Enrolled Agents directories, and the IRS Directory of Federal Tax Return Preparers cover most of the discovery surface. Then pick a lane — real-estate investors, restaurant owners, physicians, content creators, traders — and own its long-tail queries. Generic "tax preparation near me" is a bloody keyword; "short-term rental depreciation enrolled agent" is not.

Once returns start moving, instrument two numbers weekly: throughput and accuracy. Throughput is returns completed per preparer-week against a target. Accuracy is e-file reject rate plus post-filing notices per hundred returns. Investigate any week that misses throughput within seven days — the cause is usually intake, not preparer speed, and a better organizer fixes more than a faster typist does. But treat accuracy as the dominant metric. One notice for a missed K-1 line costs more in remediation time, insurance exposure, and referral damage than ten additional returns earn.

Where the durable margin actually sits

The industry generates roughly $11 billion annually across something on the order of 110,000 firms in the United States. Paid preparers e-file in the neighborhood of 63 million individual federal returns out of roughly 165 million total. Those two ratios tell you the whole strategic story: a large majority of firms are small, and a large minority of taxpayers still pay someone. But the taxpayers leaving for DIY tools are overwhelmingly the simple ones, which means the paid-preparer share is shrinking in count while shifting upward in complexity.

That shift is the opportunity. Judgment-heavy work resists commoditization for structural reasons, not sentimental ones. Multi-state apportionment requires knowing which states use market-based sourcing and which use cost-of-performance. Partnership allocations require reading the operating agreement. Crypto reporting requires reconciling wallet-level cost basis across exchanges that report inconsistently. Reasonable-compensation analysis for an S-corp shareholder requires defensible documentation. None of these are form-filling problems, and software that autocompletes forms doesn't touch them.

Representation is the other durable line. An Enrolled Agent can represent any taxpayer before the IRS on any matter — examinations, collections, appeals. Notice resolution, installment agreements, penalty abatement requests, and audit representation are billed hourly or on flat engagement fees, and they arrive year-round rather than in an eleven-week crush. A book with meaningful representation revenue is a fundamentally more stable business than one that only files.

Then there's the adjacent-services question, which is where tax preparation stops being a standalone business and starts being the front door to a small-business advisory practice. The client who hands you a shoebox of receipts in March is a bookkeeping client eleven months of the year. The one who asks whether to buy the truck in December or January is a planning client. The one forming a second entity needs structure advice. Each of these is higher-margin than the return itself and each smooths the revenue curve. This is the same motion a RevOps practitioner would recognize as expanding within an existing account — the acquisition cost is already sunk, so incremental revenue carries near-full margin.

A few failure modes worth naming plainly. Opening a storefront before holding a credential inverts the value proposition: the marginal client cares whether you can fix an IRS notice, not how the lobby looks. Competing on price against DIY software for simple returns means fighting for the segment with the fastest structural decline. Hiring seasonal preparers before you have documented process and a review workflow converts every hire into liability exposure. And underbilling complex work — multi-state, multi-entity, crypto, representation — gives away the exact margin that justified getting credentialed in the first place.

Related questions

Do I need an EFIN if I only file a handful of returns?

Yes, if you e-file for compensation. Preparers who file more than a small threshold of returns are required to e-file, and clients generally expect it. The application takes about 45 days, so treat it as a fixed lead-time item regardless of expected volume.

Can I run this part-time alongside another job?

Commonly, yes. Many preparers start with 40 to 80 returns in a first season while employed elsewhere. The constraint is the compression — most of the work arrives in eleven weeks, so plan evenings and weekends deliberately and cap your intake before you overcommit.

Is buying an existing client book better than starting cold?

If you already hold a credential but lack a network, often yes. You inherit relationships and filing history. Diligence matters: review engagement letters, check for open notices or amended returns in progress, and structure retention-based earnout terms rather than paying entirely upfront.

What's the fastest legitimate credential to start with?

The Annual Filing Season Program. It requires continuing-education hours including a federal tax refresher rather than an exam, and it gets you into the IRS preparer directory with limited representation rights. Treat it as a bridge while studying for the Enrolled Agent exam.

FAQ

Do I need a license to prepare taxes in 2027?

Federally, you need a PTIN to be paid for preparing returns, and an EFIN to e-file — but there is no general federal license for preparers. A minority of states, including California, Oregon, Maryland, and New York, impose their own registration or licensing requirements. Check your state board of accountancy or department of revenue directly, since these rules change.

How much does it actually cost to start?

A home-based solo launch typically lands between $11,000 and $15,000 for the first year, including software, insurance, entity formation, marketing, and a working-capital reserve. A virtual boutique runs $15,000 to $25,000. Adding a storefront pushes the total to roughly $35,000 to $50,000 once lease, buildout, and seasonal payroll are included.

Which software should a first-year preparer buy?

It depends on volume. Per-return pricing like Intuit ProConnect suits a first season under about 60 returns; an unlimited package like Drake becomes cheaper above that. Demo two, run a prior-year return you already know through both, and compare speed and multi-state handling before committing to a full year.

How do I find my first clients?

Referrals and search, in that order. Name fifteen people in your existing network who either need a return or know people who do. Then build a Google Business Profile, list in the NATP and NAEA member directories and the IRS preparer directory, and publish content targeting one specific niche rather than generic local terms.

Should I specialize immediately or stay general?

Stay general in season one to learn what you enjoy and where you're efficient, then specialize in season two. Specialization drives fee levels — a preparer known for short-term-rental or crypto returns competes on expertise rather than price, and those clients rarely shop on cost.

How do I survive May through December?

Build a year-round revenue base before you need it. Monthly bookkeeping retainers, quarterly estimated-tax planning, S-corp election and reasonable-compensation analysis, and IRS notice resolution all bill outside the season. Simultaneously, reserve a fixed percentage of spring revenue in a separate account and treat it as untouchable operating capital.

Sources

• https://www.irs.gov/tax-professionals/ptin-requirements-for-tax-return-preparers
• https://www.irs.gov/e-file-providers/become-an-authorized-e-file-provider
• https://www.irs.gov/tax-professionals/enrolled-agents/become-an-enrolled-agent
• https://www.irs.gov/tax-professionals/annual-filing-season-program
• https://www.irs.gov/newsroom/filing-season-statistics
• https://www.sba.gov/business-guide/launch-your-business/choose-business-structure
• https://www.natptax.com/
• https://www.naea.org/
• https://www.aicpa-cima.com/resources/landing/cpa-exam
• https://directfile.irs.gov/

Related on PULSE

• Should I open a tax preparation business in 2027?
• How do you start a gig-worker tax prep business in 2027?
• Should I open or buy a Liberty Tax franchise in 2027?
• How do you start a digital marketing agency in 2027?
• How do you start a home cleaning service business in 2027?
• How do you start a landscaping business in 2027?