How to Do a Business Audit Yourself: A Founder's Step-by-Step

Most founders know something is off before they can name it. Revenue is fine. The team is working hard. But close rates are slipping, or margins are compressing, or a competitor is gaining ground that did not used to threaten you. A self-conducted business audit is the structured way to find out what is actually happening — before you pay someone else to tell you.

This is a practical guide to running one. Four areas, a specific method for each, and an honest account of where a self-audit reaches its limits.

Area 1: Financial health

Start with the numbers. Pull gross margin, net margin, and operating cash flow for the last four quarters. Then calculate two ratios: CAC payback (total sales and marketing spend ÷ new customers acquired ÷ monthly gross profit per customer) and Net Revenue Retention (starting ARR + expansions − churn − downgrades ÷ starting ARR × 100).

If your NRR is below 90% and you are not sure why, that is a finding. If your CAC payback is above 18 months, that is a finding. Do not try to explain them yet — mark them and move on. The explanation comes after you have collected data across all four areas.

Area 2: Competitive position

Pick your top three competitors. For each one, answer: what is their entry price, their mid-market price, and their positioning? What channels do they use to acquire customers? What has changed in their product or marketing in the last six months? What is their primary claim against the category — and what does it imply about their target customer?

The data for this is mostly public: their website, their pricing page, their LinkedIn company page activity, recent press releases, and third-party review sites. A thorough competitive audit takes two to three hours per competitor. If you find yourself saying "I do not know" more than once per competitor, the self-audit is surfacing a real gap.

Area 3: Acquisition channels

Build a simple channel attribution table: list every acquisition channel, the approximate spend or time investment per month, the leads generated, and the close rate. Which channel generates the most qualified pipeline? Which has the worst unit economics? What would happen to pipeline if you doubled the budget on your best channel?

This exercise frequently surfaces that one or two channels are responsible for most of the quality pipeline, while several others absorb meaningful resources with poor returns. The self-audit does not need to produce a recommendation here — it needs to surface the data clearly enough that the recommendation becomes obvious.

Area 4: Product and retention

Look at your usage data. What features do your best customers use most? What features do churned customers use least? What was the most common reason for churn in the last 12 months? If you do not have clean answers to these questions, the audit is telling you something about your data infrastructure, not just your product.

Cross-reference with the sales conversation notes from the last 20 deals — won and lost. What objections came up most often? What competitor names appeared? What pricing comments came up? The qualitative data from sales conversations is often the most direct signal of where the product or positioning needs work.

Synthesising the findings

Once you have collected data across all four areas, look for the pattern. Are the findings independent problems, or do they point to a single root cause? Compressing margins, rising CAC, and increasing competitor mentions in sales conversations often point to the same cause: a competitor is undercutting your positioning and you have not responded.

Write one sentence that names the pattern: "The core finding is that [specific dynamic] is happening, which explains [observed symptoms]." Then name the decision this finding supports. That is the named next move — the specific thing to do, not a range of options.

Where self-audits run out of steam

A self-audit is limited in three ways. First, you can only audit what you can see — and the most important competitive moves are often happening out of sight: a competitor's new enterprise tier, a partnership not yet announced, a pricing experiment running on a segment you do not serve. External research can surface these. Internal research cannot.

Second, a self-audit suffers from confirmation bias. Founders tend to find the evidence that confirms existing hypotheses and underweight evidence that contradicts them. This is normal — but it means self-audit findings should be treated as strong hypotheses, not conclusions.

The self-audit is the right first step. It sharpens the question. Then decide whether that question is important enough to warrant an external investigation.

If the self-audit produces a finding you can act on immediately, act on it. If it produces a question you cannot answer from internal data — or if the stakes are high enough that confirmation bias is a real risk — the Free Decision Diagnostic is a useful bridge: send the specific question the self-audit surfaced, and we will return a sourced 5-page brief on what the public evidence says. 24 hours, no cost.