Why an afternoon is enough
You do not need a six-week consulting engagement to find your biggest leak. You need four hours, honesty, and the discipline to trace one real path from stranger to conversation instead of reading five dashboards that each describe a fragment.
Block the afternoon. Turn off notifications. Work through these five stages in order.
Stage 1: Be your own lead (45 minutes)
Open your phone on mobile data, not office wifi. Find your business the way a customer would — search your category and city, or click your own ad.
Note, without excusing anything:
- How long the page took to become usable.
- Whether the first screen said what you do, for whom, and what to do next.
- How many taps to reach an enquiry action.
- Whether the form asked for anything you would resent giving a stranger.
Then actually submit an enquiry using a personal number your team will not recognise. Write down the timestamp.
Now wait. Do not intervene. Record when a genuine human responds. This number alone frequently explains a quarter of flat revenue.
Stage 2: Pull four numbers (45 minutes)
You need only four figures for the last full month:
- Clicks or sessions from your main acquisition source.
- Landing page views for the page those clicks hit.
- Form starts or enquiry initiations.
- Qualified conversations — real discussions with someone who could buy.
Calculate the percentage drop between each adjacent pair. One gap will be dramatically worse than the others. That is your priority, and everything else is a distraction this quarter.
Common patterns and what they mean:
- Clicks high, page views much lower: page speed, broken tracking, or a redirect problem.
- Page views high, form starts near zero: the offer, clarity, or trust is failing above the fold.
- Form starts high, completions low: form length, required fields, or validation errors on mobile.
- Completions high, qualified conversations low: lead quality or response speed.
Stage 3: Read twenty raw leads (45 minutes)
Not the summary — the actual records. Open the last twenty enquiries and read them like a human.
For each, mark: real prospect or not, in service area or not, budget plausible or not, responded within an hour or not, and what the final outcome was.
Twenty records is enough to see the shape. You will usually find one of three stories: the leads are wrong, the leads are fine but slow-handled, or the leads are fine and handled and the pitch is losing.
Stage 4: Check the plumbing (45 minutes)
Verify the unglamorous things that silently break:
- Does the form actually deliver? Send a test and confirm it arrives in the CRM, not just in an inbox someone forgot.
- Is conversion tracking firing? Compare platform-reported conversions against CRM records for the same week. A large mismatch means your optimisation is being fed noise.
- Do WhatsApp and phone enquiries get recorded anywhere, or do they exist only in one person's phone?
- Is there an owner for leads that arrive after hours and on Sunday?
- Does anything auto-acknowledge the enquiry so the person knows they were heard?
Missing acknowledgement is the cheapest fix in marketing and one of the most commonly absent.
Stage 5: Write the one-page verdict (45 minutes)
Force yourself into a single page with four lines:
Biggest leak: the step with the worst drop, stated as a number.
Most likely cause: your best current hypothesis, stated plainly.
Next change: one specific action, owned by one named person, with a date.
How we will know: the metric that should move, and by when.
One change. Not seven. Systems improve through sequential fixes with clean measurement, and collapse into confusion when five things change at once.
What not to do after the audit
Do not increase ad budget because the audit was uncomfortable. If the leak is at the page or the follow-up, more traffic multiplies the loss.
Do not redesign the entire website because one section underperforms. Redesigns reset your learning and often lose what was working.
Do not change your agency in the same week you change the offer. You will never know which move mattered.
Reasonable targets to aim at
Rough benchmarks for a services business, useful as direction rather than gospel:
- First response to a new enquiry: under fifteen minutes in working hours, automated acknowledgement instantly at all hours.
- Landing page to form start: healthy is usually mid single digits to low double digits, depending on offer and traffic quality.
- Form start to completion: above seventy percent, or your form is too long.
- Qualified rate: depends entirely on your definition, which is why you must write the definition down first.
The point of the exercise
An audit is not an evaluation of your team. It is a map of where value escapes the system. Most businesses discover that the expensive top-of-funnel work is competent and the cheap unglamorous middle — response time, form fields, routing — is where the money leaves.
That is good news, because those fixes are fast, cheap, and entirely within your control.
