"So what would this cost?"
You're twelve minutes into a first call and it's going fine. The business is a mess in the ordinary way, and you already know two things you'd fix in the first month. Then that question lands and something in your chest goes sideways, because you heard it as an accusation instead of a buying signal.
So one of three things happens. You go quiet and say you'd need to look at a few things first. You name a number lower than the one you meant, before anybody pushed on it. Or you say the sentence that has killed more work than any price ever has: let me send you some information.
Then nothing. And two weeks later you put it down to price, because price is the only explanation anybody ever hands you.
Here's the part that should bug you. Every March you do something harder than that call, and you do it well.
Somebody sits across from you convinced they owe money they don't have, scared and a little embarrassed about the shoebox. You take the frightening thing and put it in words they can hold. You tell them what happens next and roughly when. They leave calmer than they came in, and they tell two people about you.
Most people I know in this profession are very good at explaining something difficult to a nervous person. That is selling.
The gap isn't nerve and it isn't a personality you were supposed to be born with. You run that sequence beautifully with a client and you don't run it at all with a prospect, because nobody ever wrote it down for you. So on a sales call you improvise. Improvising under pressure looks like going quiet, or like discounting.
My own firm, SDO CPA, has been full since June and runs a waitlist, so I say no for a living now. I've also spent the last few weeks asking other firms what work they actually want, which is this same conversation pointed the other way, and most of them couldn't answer. When I laid out the system that fills a firm a few weeks back, and then the referral loop last week, both of them ended in the same place. Demand shows up. Then somebody has to have the conversation.
One thing to square with the first issue. I told you to publish your ranges on an intake form and stop taking prospect calls, and that's still what I'd do at my own firm. The conversation just moves later. Somebody who has read your ranges and filled in your form still asks what it'll cost, because what they're really asking is whether you understand their situation well enough for the number to mean anything.
How do you answer "how much do you charge"?
You answer it, in a range, on the call. Silence reads as hiding something and a fast discount reads as an opening bid. But answer it second. Before the number you want three things: what's going on in the business right now, what it costs them if nothing changes, and who signs. With those, a range is a diagnosis and it lands like one. Without them, any number is a guess you'll defend badly. If you genuinely can't range it on the first call, say exactly what you need in order to and when they'll have it. Never trade the number for a free look at the books.
Key Takeaways
You already run this sequence every March - explaining something hard to a nervous person, in order, is the skill. You just don't run it on prospects.
Qualify before you present anything - the same three questions every time, before any number and any scope.
A diagnosis beats a menu - "here's what I'd do first, and why that first" asks them to trust you. A service list asks them to be the expert, which is why they called you.
The ask is one next step with a date on it - most warm work goes cold on "let me know what you think."
Follow up twice, spaced, useful both times - then stop. Silence after the second one is an answer, and it's cheaper to get than to wonder about.
They're buying certainty about you - what you refuse to do proves that faster than a list of everything you can.
The diagnosis sequence
Four moves: qualify, diagnose, ask, follow up. Run them in that order and most of what feels like a sales problem stops happening to you.
None of this came out of accounting. I took it from businesses that sell things for a living, and one thing had to change before it survived contact with a firm owner. Somebody hiring you isn't buying a product. They're buying a few years of one person looking at their worst numbers. So the questions ask about the business rather than about budget and timeline, and the diagnosis has to be something you'd be willing to say to their face in March.
Qualify before you present anything. Three questions, the same three every time, so you're not inventing them at minute twelve.
What's going on in the business right now. What happens if nothing changes. Who else is part of this decision.
The first gets you the mess. The second gets you the price, because the cost of doing nothing is the only number your fee is ever measured against. The third is the one everybody skips and it's the expensive one. You can run a perfect call with somebody who was never going to be allowed to say yes. So ask it in the first ten minutes, awkward or not.
Present a diagnosis, not a menu. A menu hands the decision back to somebody who came to you because they couldn't make it. Fourteen bullet points of services is a menu, and so is a rate card with no opinion attached.
A diagnosis sounds like this: here's what I'd fix first, here's why it goes first, here's what happens after. Two sentences. You already do this with tax positions.
You'll be wrong sometimes and that's survivable. Being agreeable and unspecific is not.
Ask for one next step, and put a date on it. Not "let me know what you think." That's a wish.
An ask sounds like: I'll send the engagement letter Thursday morning, and if it's signed by Monday I start on the first. Or, when they're clearly not ready: let's take twenty minutes Tuesday and decide whether this is a fit, and if it isn't I'll tell you who to call instead.
One next step, one date. You'd never end a client meeting without one, and both versions are written out in the annex.
Follow up twice, spaced, useful each time. The first goes three business days later and carries the thing you promised on the call, because you did promise something. The second goes about a week later and carries one observation about their business you didn't say out loud the first time.
Both dates get set the moment the proposal leaves your outbox, not on the day they come due. Two calendar events, booked while you're still thinking about the person, messages already inside them. The annex has the shape; it's a two-minute job.
Then you stop. Two is the rule. A third starts costing you the thing you were trying to protect, and by then the silence has told you what you needed to know.
What they're actually buying
Nobody has ever hired a bookkeeper because the service list was thorough.
What a prospect is working out, the whole call, is whether you tell them the truth when the truth is inconvenient. Everything else is detail. Which is why the fastest trust move you have is naming what you don't do.
"I don't do tax returns, and I'll hand you to someone who does" costs you nothing and buys you a person who believes the rest of your sentences. A capability tour buys the opposite, because everybody they've ever hired said yes to everything.
What you can hand off, and what you can't
Later, and gladly: the scheduling link, the reminder that fires, the notes landing somewhere other than the back of an envelope, the second follow-up going out on time. That's plumbing, and plumbing belongs with software or with somebody who isn't you. Most of it you can build yourself in an afternoon now.
If somebody else takes the inquiry before you do, the sequence still holds. It just starts one desk earlier: they collect the three answers, you make the diagnosis. Capture sheet for that is in the annex.
Not ever: the diagnosis conversation, and the call on who's a fit. Your judgment is the product in both. Hand them off and you've hired someone to be you badly, in front of the people deciding whether to trust you.
Do this before Friday:
Write your three qualifying questions on one card and keep it where you take calls. Same three, every time.
Open your info email, the one you send after a first call. Rewrite it as four lines: their problem in their words, what you'd do first and why, a range, one next step with a date.
Send one overdue follow-up today, the one that's been sitting longest and quietly bothering you. Then before you close the tab, put its second date on your calendar with the message already written into it. That's the part that keeps working after this week.
What's the sales moment you dread most: the price question, the follow-up, or the no? Hit reply and tell me which one. Whichever pile gets biggest becomes a future issue, and I read every one of them.
Operator annex
Below is the file I open before a first call.
The call sheet
Print it, duplicate it, whatever. The point is that you're reading questions instead of remembering them while somebody is talking at you.
Prospect:
Date:
Came from:
1. What's going on in the business right now?
Answer:
Read: usable / vague / not a real problem yet
2. What happens if nothing changes?
Answer:
Cost of doing nothing, in their words:
3. Who else is part of this decision?
Answer:
Decider on this call? yes / no
If no, next step is getting them on one.
Diagnosis, two lines max:
Range I gave:
Next step + date:
Follow-up 1 due:
Follow-up 2 due:What the answers actually tell you
Question one. "What's going on in the business right now?"
A usable answer has a date in it. "We're four months behind and the lender wants statements by the end of October." Now you have a mess and a clock, and the clock is doing your selling.
A vague answer sounds like "we just want to get more organized." Nothing to diagnose there. Ask the follow-up that saves the call: what made this the month you decided to deal with it? Something happened. It always did.
Question two. "What happens if nothing changes?"
A usable answer: "I miss the extension deadline again and I'm guessing at estimates for the third year running." Write it down in their words. That sentence is what your fee gets compared against, and you're going to quote it back to them.
A weak answer: "I guess it'd just carry on being annoying." Believe them. Low pain, low budget, and no amount of discounting turns that into a good client. It turns it into a client who calls you in April.
Question three. "Who else is part of this decision?"
Usable: "Just me." Or "me and my partner, and she'll want to see the number before anything."
The one to watch for: "I'll take it back to the team." Don't let that pass. Who's on the team, what will they want to know, and should you be on that call. Half the proposals that die quietly die in a room you weren't in.
When they ask the price before you've scoped it
This is the moment people discount. The reflex is to soften the number before anyone has objected to it, which sets the number at the bottom of your range and tells them the range was decorative.
The move is to answer, and to answer with the two things that move it:
Fair question, and you'll have a range before we hang up. Two things move it:
[thing one] and [thing two]. Tell me about [the one that moves it most] and
I'll narrow it while we're talking.Then deliver it with the edges attached, so the number reads as a diagnosis:
For what you've described, [range]. It's the bottom of that if [condition],
and the top if [condition]. I'll know which once I've seen [the specific
thing you need to see].And the pair worth reading twice. The flinch:
...which I know is probably more than you were expecting. I could
probably start at [lower number] and we can revisit in a few months.Holding the range:
That's the range. If it's more than you want to spend right now, the smaller
version is [narrower scope] at [lower range]. Same standard of work, less of
it. What I won't do is the full job at the smaller number, because then I do
it badly and we both regret it.Nothing here is a fixed price. Ranges and estimates, always, and the number moves when the scope moves. What the number should actually be is a different problem from how you say it out loud, and that one has its own issue now.
Turning the info email into a diagnosis
Three real rewrites. Different people, different lengths, different amounts of patience.
One. The cleanup.
What most firms send:
Hi [Name],
Great speaking with you today! As discussed, please find attached an overview of
our services. We offer monthly bookkeeping, cleanup work, payroll support, sales
tax filings, and year-end tax preparation. Pricing depends on scope of work and
transaction volume.
Let me know if you have any questions and I'd be happy to discuss further.What actually gets a reply:
Hi [Name],
Short version of what I heard: [the mess, in their words], and the thing setting
the clock is [the deadline they named].
I'd start with [the one fix], because until that's right everything downstream is
guesswork. That piece runs [range]. Once it's clean, ongoing is [range] a month.
Engagement letter comes over [day]. Signed by [day] and I start [date].Two. The one who's already annoyed. Short call, short email, no throat-clearing.
[Name],
You've got [the specific problem] and somebody's already priced you for [the
thing that doesn't fix it]. That's why it hasn't gone away.
[The one fix], [range], done by [date].
Say go and I'll send paperwork today.Three. The one who isn't ready. Don't fake a proposal. Say the real thing.
Hi [Name],
Being straight with you: [what they described] isn't work I'd take on right now,
and I don't think you'd be happy with what it cost.
What I'd do in your position is [the smaller, cheaper move], which is a
[timeframe] job, not a [timeframe] one.
If [the trigger] happens, come back and I'll pick it up. I'll remember the file.The shape underneath all three:
[One line: their problem, their words.]
[One line: what I'd do first, and why that goes first.]
[Range, and what it covers.]
[One next step, with a date.]The two follow-ups
Both short on purpose. Long follow-ups don't get read and they make you sound worried.
Follow-up one, three business days out:
Hi [Name],
Here's [the thing I said I'd send]. The part that matters for you is [one
sentence].
If you want to move, I can start [date]. If the timing's wrong, just say so and
I'll stop filling up your inbox.Follow-up two, about a week after that:
[Name],
One thing I kept thinking about after we spoke: [the observation, specific to
their business].
Worth knowing either way. If this isn't the season for it, no hard feelings and
I'll leave it there.Two. Then stop. The permission to stop is most of the value here, because what keeps people from sending the first follow-up is knowing there's no agreed end to it.
The part that runs while you're with a client
Everything above only fires if you remember. So don't remember.
The moment the proposal leaves your outbox, before you close the tab, make two calendar events with the messages already inside them:
Event 1
Title: Follow up 1 · [Prospect] · send [the thing I promised]
When: 3 business days from today
Notes: [follow-up one, already filled in with their details]
Event 2
Title: Follow up 2 · [Prospect] · one observation, then stop
When: 7 days after Event 1
Notes: [follow-up two, already filled in]Two minutes at send time. Then it happens on a Tuesday you've forgotten all about, with the email already written, so the only thing left is to press send. No prompt for this one and none needed. It's two minutes and a habit.
Three places it can live, in order of how little setup they need:
Your calendar. Works everywhere, no setup, and it's already the first thing you open. Default choice.
Send-later, if your mail has it. Write both follow-ups the same minute you send the proposal and schedule them for their dates. Now it isn't even a decision on the day, and you can still pull one back if they reply.
Whatever sent the proposal. If it does reminders, put the two dates there so the record and the reminder sit in one place. Only worth it if more than one person at your firm touches a deal.
The audit is event-triggered. Any time a prospect surfaces in your head and isn't on the calendar, put them on it right then, with the message written in. Carrying them around in your head is the failure mode this replaces.
There's no tracking spreadsheet in here on purpose. Your calendar already opens every morning. A spreadsheet opens when you remember it, which tends to be the same moment you remembered the follow-up anyway.
The "what I don't do" block
This goes in two places: a few lines near the bottom of a proposal, and a short section on whichever page prospects land on. The labels below are mine, for you. Don't paste them.
Six versions, so you can hear the range. Pick one voice and stay in it.
1. I don't do tax returns. I'll hand you to someone who does, and I'll stay on
the file so nothing falls between us.
2. No cleanups shorter than three months. Under that you don't need me, you need
a free afternoon.
3. One new client a month. If this month's taken I'll tell you the month I have.
4. I won't take work that needs to be finished in a week. It'll be wrong, and
we'll both know why.
5. Pre-revenue startups aren't for me. Come back the month payroll starts.
6. I don't take clients who want the books to say something the bank statements
don't.Six is more than anybody needs. Pick two, and pick the two you've genuinely said no to this year.
If someone else takes the inquiry first
Once your firm is past three or four people, the owner usually isn't the first voice a prospect hears. The sequence still works. It just starts one desk earlier, and the person at that desk needs to know what to collect.
INQUIRY CAPTURE · give to [owner] same day
Who called / wrote, and when:
How they found us:
What they said is going on (their words, not a summary):
What happens if nothing changes:
Who signs. Were they on the call? yes / no
Anything we promised them, and by when:Two rules for whoever holds that sheet. Write their words, not a tidied version, because the tidied version is where the useful detail goes to die. And escalate same day, not Friday, if there's a deadline in the answer to question two.
Two prompts (review before you apply anything)
Both of these hand back a draft you'd otherwise write from scratch: the after-call email, and the block of work you don't take.
Same rule on both: the agent drafts, you publish. Nothing it writes reaches a client, a page, or a file until you've read it end to end and decided you'd sign your name to it.
Prompt one: rewrite the email you send after a first call.
I want to rewrite the email I send after a first sales call. Draft only. Don't
publish anything, don't send anything, and don't create or change anything in any
account of mine.
The email I send today: [paste it].
A real call it would have followed: [paste your notes, or answer these three:
what's going on in the business, what happens if nothing changes, who decides].
1. Rewrite it as four lines: their problem in their words, what I'd do first and
why that first, a range and what it covers, one next step with a date.
2. Give me three versions at different lengths. Say which one you'd send and why.
3. List every sentence you cut and what it was doing.
4. Flag anything that reads as a free review, a free look at the books, or a
promise about a result. I can't say any of those.
Hand it all back. I'll choose one and send it myself.Prompt two: write the block of work you don't take.
Help me write the "what I don't do" section for my firm. Draft only. Don't touch
my website, my forms, or anything that's live.
Work I've turned down in the last year: [paste the list, or paste the emails
where you said no].
1. Sort those into the three or four real reasons I said no. Ignore the polite
wording I used at the time.
2. Write each reason as one plain sentence a prospect could read without feeling
insulted. No apology and no hedging.
3. Give me a short version for a proposal and a slightly longer one for a page on
my site.
4. Tell me which of these will cost me the most work, and say it plainly rather
than talking me out of it.
Then stop there. I'll decide what goes up and I'll put it up myself.So read what it hands you before any of it goes up. You'll be holding those lines on days when holding them costs you the job.