A partner at a small CPA firm looks at the first month of Google Ads and does the obvious math. The account spent $1,500 and brought in six new clients. That's $250 a client, and the average individual return the firm files costs about that much. "We paid for the work to do the work," the partner says, and wants to turn the ads off.
The math isn't wrong. It's just one year long. A tax client doesn't buy once and disappear. They come back next February with a new W-2, and the February after that, and some of them stay for decades. Price the click against one return and almost every click looks expensive. Price it against the years a client stays and the same click can be the best money the firm spends all year.
Accounting Firms Buy a Client Once and Bill Them Every Year
Every Google Ads budget for an accounting firm comes down to one chain of numbers:
- Cost per click. What Google charges each time someone clicks your ad.
- Conversion rate. The share of clicks that become a lead: a call, a form, a booked consultation (the conversions Chapter 1 set up).
- Cost per lead. Cost per click divided by conversion rate.
- Close rate. The share of leads that sign an engagement letter.
- Cost per client. Cost per lead divided by close rate.
- Lifetime value. What that client pays you over every year they stay.
A click price on its own means nothing. A $20 click can be cheap and a $5 click can be ruinous. The chain decides which, and the last link, lifetime value, is where accounting firms have an advantage most advertisers would envy.
At $12 a Click, One Itemized Return Doesn't Pay for the Client Who Filed It
Here's the chain worked through with real inputs. Every number is labeled, because a budget built on borrowed guesses is worse than no budget.
- Cost per click: $12. Keyword Planner's top-of-page range for "cpa near me" in Boston, our example market, runs from $4.50 to $23.89. Twelve dollars sits comfortably inside it. Your market's range will differ; the math works the same way.
- Conversion rate: 4.85%. LocaliQ's 2026 search advertising benchmarks don't list accounting as its own category. The closest, Business Services, averages a 4.85% conversion rate. It's a vendor's figure drawn from its own customers' campaigns, so treat it as a starting point until your own account has data.
- Cost per lead: about $247. Twelve dollars divided by 4.85%.
- Close rate: 1 in 3. This one is an illustration, not a benchmark. No reliable national figure exists. Count your own: of the last 30 consultations, how many signed?
- Cost per client: about $742. $247 divided by one-third.
Now set that against what the client pays. In the National Society of Accountants' most recent fee survey, an itemized Form 1040 with a state return averaged $323. The survey collected those fees in 2020, and your own fees are the better number, but the shape doesn't change: one year of that client's fees covers less than half of what it cost to win them.
The click price swings this more than anything else. Run the same chain at the bottom of Keyword Planner's range, $4.50 a click, and a client costs about $278. At the top, $23.89, it costs about $1,478. That spread is why a firm should never copy a competitor's budget: the same number of clients can cost five times more in one market, or one month, than another.
If the partner in the opening scene stopped here, turning the ads off would be the right call. They shouldn't stop here.
The Third April Pays the Click Back
At $323 a return, a client who stays covers the $742 early in their third year. From the fourth April on, every return is money the ad already paid for.
How many clients stay that long? Nobody publishes a national client-retention rate for accounting firms, and the numbers that circulate on vendor blogs come with no method attached. The honest answer is in your own client list. Pull last year's returns and count how many of those clients have filed with you three years or more. That share is your retention, and it's worth more than any benchmark.
The reviews suggest it's high. Of 622 Google reviews we read for Boston-area firms, 110 mention how long the writer has been a client, and 20 say ten years or more. "I have been going to Simon for my tax planning for 20 years now." "I've relocated half way across the country, but would never consider switching accountants." Even at storefront tax chains, loyalty holds: "Charlene has been my tax preparer for about 12 years now."
A Business Client Pays It Back Before the First Extension Deadline
Individual returns are the slow payback. Business clients are fast. In the same NSA survey, a Form 1120-S for an S corporation averaged $903 and a Form 1065 partnership return $733. The owner's own 1040 usually comes along too, so an S corporation client can bring in more than $1,200 in year one. At a $742 cost per client, that's paid back before the first extension deadline.
Monthly work changes the math again. In the 2024 client advisory services benchmark survey from CPA.com and the AICPA's PCPS, participating firms reported monthly fees that imply a median around $3,000 a month for ongoing bookkeeping and advisory clients. That's a vendor survey of firms that chose to take part, based on 2023 data, but even a fraction of it makes a $742 client look like a bargain.
This is why every service line needs its own landing page and, eventually, its own budget. A click on "small business accountant" and a click on "tax preparation near me" can cost the same and be worth ten times apart. Send both to one generic services page and you can't see the difference, let alone bid for it. Chapter 4 shows how those pages also lower what each click costs.
A Voicemail Box in March Raises Your Cost per Client
The close rate is the link a firm has its own hands on, and the one that slips first in tax season.
Responsiveness was the most common subject in the 622 reviews: 178 of them, including 33 complaints. The complaints read the same way: "Nobody ever answers the phone or calls you for an appointment." The praise is the mirror image: "He actually answered the phone several times when I called." Every unanswered call came from someone who had already decided to reach out.
Run it through the example. At $247 a lead, closing 1 in 3 means about $742 a client. Let enough March calls roll to voicemail that you close 1 in 4 instead, and the same ads now cost about $990 a client. Nothing changed in Google Ads. The price went up at the front desk.
Tax season makes this worse, because it's exactly when calls pile up and people are busiest. Some firms simply stop taking new clients by March, and one five-star review in our sample said of a preparer, "She is too busy to take new clients I should mention." If that's your firm in March, the ads shouldn't be running for new 1040 clients in March either. Chapter 15 covers budgeting around tax season. A firm that can't answer in March should expect its cost per client to rise in March, and budget for that, or fix the phones first.
Half of Tax Clients Arrive by Referral, So a Google Client Can Bring the Next One for Free
In a 2025 poll of 2,000 US taxpayers published by TaxSlayer Pro, 49% said they found their preparer through someone they knew. That's a vendor's poll, and it describes how people found preparers in general, not Google clients in particular. But it points at value the worksheet leaves out. A client who arrives through Google and has a good experience tells a coworker, a sibling, a business partner. That next client costs nothing.
Reviews work the same way. Every client who writes "they explained everything clearly" makes the next click more likely to turn into a call. Don't put referrals and reviews into the bid math, where they'd tempt you to overpay. Count them as upside you earn by getting the first client right.
A Budget Built From Leftover Cash Buys Clients by Accident
Most small firms set an ad budget by asking what they can spare. The chain lets you ask a better question: how many new clients do you want, and what does each one cost?
If the goal is ten new clients a month at a $742 cost per client, the monthly budget is about $7,400. A two-partner firm that only wants four new clients a month needs about $2,970, or roughly $98 a day. Neither number is a recommendation. Both are what the chain says a goal costs, which is more than most budgets are based on. Google Ads works in daily budgets, so divide by 30.4, the average number of days in a month Google uses, for about $245 a day. On a busy day Google is allowed to run past your daily figure, as far as double it, yet the month's bill stops at 30.4 days' worth.
Start where the numbers are strongest. That means Isolation first: our label for exact-match keywords with bids set by hand, limited to searches already proven to bring in clients, pointed at the hiring searches from Chapter 2. Run it for a couple of months, replace the illustration numbers with your own conversion and close rates, and the budget stops being a guess.
Next: Why the Best Ad, Not the Biggest Bid, Wins the Top Spot
Chapter 4 looks at Ad Rank and Quality Score: why Google doesn't simply hand the top position to whoever bids the most, and how tight ad groups and matching service pages can lower what a firm pays for the same click.
New here? The tracking behind these numbers is in Chapter 1, and the searches worth buying are in Chapter 2. The accounting guide's home page lists every chapter.
Want the chain worked out on your own numbers? That's where our Google Ads management work begins, before a single keyword goes live.
