Direct answer first: choosing an accounting partner is a decision about scope, accountability, and process — not about logos or lowest price. You are buying three things: a named person who owns your books, a defined routine (close, reporting, tax calendar) that runs without you, and a team that will tell you what your numbers mean. This guide covers what to actually evaluate, the five checks to run before you sign, what the first conversation should reveal, and the red flags that should end it early.
What you are actually buying
Most owners shop for an accountant the way they shop for a utility: cheapest reliable option, switched rarely, thought about never. That framing is wrong in one specific way — accounting is not a commodity, because the cost of a bad provider is not the fee; it is the unreconciled year you discover at tax season, the deal that stalls in diligence, the filing that arrives late. When the downside is that large, the buying decision deserves more structure than a price comparison.
Here is what a good engagement actually consists of — and therefore what you are paying for:
- A named person, not a queue. You should be able to name the human who owns your books and reach them. The most common failure in small-business accounting is the anonymous assembly line: your work moves through hands, no one remembers your business, and accountability evaporates. Ask for the person; if you cannot get a name, you have your answer.
- A defined routine. The engagement should come with a calendar you can see: what happens monthly (close, reports), quarterly (review, filings), and annually (statements, tax returns, planning). A partner who cannot describe their own cadence cannot run yours.
- A reporting commitment. You are buying reports that get read and explained — not a file you have to chase. The engagement should say what you receive, when, and in what form, and who walks you through it.
- A scope boundary. The contract should say what is included (bookkeeping? close? tax filings? advisory?) and what changes cost. Ambiguity here is where "this was extra" surprises come from.
The five checks before you sign
Before you sign anything, run these five checks. They take an afternoon, and they catch most bad engagements early:
1. Who does the work — and who owns it
Ask directly: who will be your day-to-day contact, who does the actual work, and who is accountable if something is wrong? The answer should be a specific person or a small team, not "our team handles everything." Then ask how long that person has been with the firm. High churn is how continuity dies.
2. The process, on paper
Ask for their close process and their reporting calendar. A real partner will have both documented and will walk you through them. A vague answer ("we handle everything as it comes") is a confession that your books will be handled reactively — which is exactly the failure mode you are trying to avoid.
3. Software compatibility
Your books will live in software — theirs or yours. Ask what they use, whether they will work in your existing system or migrate you, and what the migration involves. A partner who refuses to work in your stack, or insists on migrating you with no explanation, is a partner who values their convenience over your continuity.
4. The scope in writing
Ask for the engagement letter or scope of services before you commit. It should list what is included month by month, what is extra, how changes are agreed, and how you terminate. If they cannot produce a scope on request, that is the answer — an engagement that was never scoped will be renegotiated at every surprise.
5. A reference you can call
Ask for a reference from a business of roughly your size and stage — not their flagship client. Then call it and ask three questions: does the named person actually do the work, do reports arrive on schedule, and what has gone wrong in the relationship? The third question is the one that tells you the truth.
What to look for in the first conversation
The first conversation is diagnostic — for you and for them. You are both evaluating fit, and a good partner will ask more questions than they answer. Specifically:
- They ask about your business before they talk about theirs. What do you sell, how do you bill, where does cash get complicated, what keeps you up at night? A partner who starts with their brochure is selling a service; one who starts with your business is solving a problem.
- They tell you what you do not need. The single strongest signal in the whole conversation: a partner who says "you do not need that yet" — a CFO layer you are not ready for, software you do not need to buy, a service you can defer. It is also the exact opposite of the typical sales behaviour, which is why it is so informative.
- They are honest about scope and price. They name what is included, what is not, and what drives the fee. Vagueness about price is a red flag dressed in politeness.
- They talk about process, not people. "We have thirty years of experience" is a fact about the firm, not about your books. "Here is how your first month will run, step by step" is a fact about your engagement. Process talk predicts outcomes; people talk predicts marketing.
The red flags
These deserve a hard stop — not a discount, not a second meeting:
- No named owner. You cannot find out who will actually do your work, or the answer changes every time you ask.
- Vague scope, elastic price. The engagement cannot be described in writing, or the fee is "whatever it takes" with no boundary.
- Promises with no process. Big assurances ("we will make sure you never have a problem") with no calendar, no checklists, no mechanism. Flattery is cheap; routines are expensive, and you are buying the routines.
- Dependence on one person's head. The partner who keeps everything in their memory, with nothing documented, is the partner whose departure will destroy your continuity. Documentation is a feature; secrecy is a risk.
- Over-promising the layers. The firm that says "we are your CFO" when they are selling you bookkeeping — title inflation is not a discount, it is a misrepresentation of what you are buying.
- Guarantees that mean nothing. "We guarantee accuracy" sounds wonderful and is worth exactly nothing; errors are corrected, not guaranteed away. What you want instead is the process that makes errors surface early — reconciliation, review, a named reviewer.
How the layers change what you should buy
Your stage decides the shape of the engagement, and a good partner will say so out loud:
| Stage | What the engagement should look like | What to weight most |
|---|---|---|
| Early — books still simple | Monthly bookkeeping + reconciliation, a quarterly review, tax-ready records | Process and named person; price matters at this stage, but not first |
| Growing — decisions getting real | Monthly close, management reporting, budget vs actual, proactive tax planning | Reporting quality and the advisory relationship |
| Scaling — financing, multiple entities | Consolidation, cash forecasting, CFO-level input, due-diligence support | Depth of team, sector familiarity, availability under pressure |
One consequence of the table: the partner that fits you at the first stage may not fit you at the third. That is normal, and it is worth saying at the start of the relationship, not at the end — it keeps the conversation honest about when and how you might grow apart.
Fee structures: what they tell you
You will be offered one of three ways to pay, and each one carries information about the relationship:
- Hourly. You pay for time. It is transparent about effort but gives the partner no reason to be efficient — and it converts every question you ask into an invoice. Hourly makes sense for projects with genuinely unknown scope; it is a poor fit for the ongoing routine, where you want the partner to get faster over time, not more billable.
- Fixed monthly fee. You pay for a defined scope of work. This aligns incentives: the partner profits by making the routine efficient, which is exactly what you want, and surprises are bounded by the scope document. The discipline it demands — a written scope — is itself the feature.
- Value-based or advisory. The fee tracks the outcome — typically for CFO-type work. It is the hardest to evaluate, precisely because the outcome is partly the partner's judgement. Fine at the scaling stage; easy to over-buy early.
A useful question to ask at any fee structure: what would happen to the fee if your business grew by a third? The answer shows whether the pricing is honest about the work or just elastic. And whatever the structure, get the fee's drivers in writing — the fixed fee should come with the scope that justifies it.
When to switch — and how to do it cleanly
Two honest signals that it is time to switch, not to renegotiate:
- The routine never arrived. You have been with the partner for two full quarters and there is still no monthly cadence you can see — reports arrive late, reconciliation is a rumour, and you still cannot name the person who owns your books. The gap is not going to close with more time; it is the nature of the engagement.
- The advice has stopped being honest. Your questions get agreeable answers, the numbers are never challenged, and the first conversation's candour has quietly become a sales performance. That is the sign the partner has stopped treating you as the client and started treating you as the account.
Switching cleanly is a project, and doing it right takes about four to six weeks: agree the transition with the old partner in writing, obtain a full handover — journals, reconciliations, filings history, open items — and insist the new partner starts with a baseline (reconciled accounts, a documented differences list) rather than a blind takeover. The businesses that lose a year of history at a switch are the ones that accepted a handover of files instead of a handover of truth.
The decision, honestly
When the five checks have passed and the red flags are absent, the final decision usually comes down to two things you cannot get from a checklist: chemistry on the routine — will this person make your monthly reporting something you look forward to or something you dodge — and the honest sense of whether they will tell you what you need to hear. The partner who tells you what you want to hear is not a partner; they are a vendor with good manners.
One more thing worth doing before you decide: run the comparison against not hiring anyone. If your books are already complete, current, and read, a partner adds advisory value. If your books are none of those things, the partner is not optional — the only question is how much longer you will pay the hidden cost of books that do not reconcile. Our guide to what good accounting actually includes is the checklist to hold any candidate against, and bookkeeping vs accounting vs CFO settles which layer you need first.
The bottom line
Choose the partner the way you would choose a doctor or a lawyer: by the quality of the routine, the honesty of the first conversation, and the named person who will own the outcome. The fee is the smallest part of the decision — the difference between a good partner and a bad one shows up in the year-end you do not have to reconstruct, the filing that is never late, and the numbers you can trust without checking.
