Before I would increase a marketing budget, I would open five recent enquiries and follow what happened to each one. Who arrived? What did they need? What did we do next?
Five enquiries won’t tell you the conversion rate of your whole business. They can give you a much better question to investigate than “How do we get more leads?”
My work has involved both acquisition and retention. That is why I want the conversation to continue after someone clicks. A campaign can bring the right person to you, and the business can still make the next step difficult.
What I would check in a lead quality review
Choose five consecutive enquiries from the same offer. Don’t select only the best ones or the ones that went wrong. Use your inbox, CRM and the actual conversation. Where the record is missing, write “unknown”.
- Promise: What did the ad or page lead this person to expect?
- Fit: Did their need match what you can deliver? Decide what a suitable enquiry means before judging it.
- Response: When did someone respond, and did the answer address the question?
- Next step: Was there one clear action, with someone responsible for following up?
- Outcome: What happened? Record a booked call, purchase, explicit refusal or an unknown outcome separately.
What the review might reveal
This is an illustrative example, not a client result. Imagine a service business reviewing the following five enquiries.
| Enquiry | What the record shows | What to investigate |
|---|---|---|
| 1 | Suitable need; asked about timing; received a general brochure. | Did we answer the question? |
| 2 | Wanted a service the business does not offer. | Is the offer clear before someone enquires? |
| 3 | Suitable need; no reply recorded. | Who owns the response? |
| 4 | Had a conversation and booked a call. | What made the next step clear? |
| 5 | Asked for a quote; outcome unknown. | Is the record incomplete, or did follow-up stop? |
These records would not justify saying the campaign is failing. They point to several possible problems that need different fixes. Changing the targeting won’t repair an unanswered enquiry. Faster replies won’t make an unsuitable offer relevant.
Choose one change you can test
I would start with the clearest repeated problem, then agree who owns the change. If suitable enquiries wait without an answer, give the handoff an owner and a response standard the team can actually meet. If people misunderstand the offer, improve the page and check whether the next enquiries show the same confusion.
Keep the offer and traffic source in view when comparing results. Record the number of suitable enquiries, replies and completed next steps, alongside the total enquiries reviewed. A change in the mix of visitors can change the result too.
For a low-volume business, five enquiries are a starting point for investigation, not proof that a change worked. Continue the review across a longer period before making a major budget decision. For the broader measurement question, read how I would measure AI marketing ROI.
Where AI fits, and where I want a person involved
AI can organise an anonymised review, group repeated questions or draft a response for a team member to check. I would keep the decision about customer fit, an unusual request or a promise to the customer with a person. Use only information you are allowed to share in an approved tool.
The useful output is a decision: what to change, who will do it and what you will check afterwards. If the delay is inside the business, find the process that gets stuck before automating it.
Your five-enquiry review
Copy this line into a private spreadsheet, then complete one row for each enquiry:
Enquiry reference · Source and promise · Customer need · Fit and reason · First response · Next step and owner · Outcome · Unknowns
After five rows, write one sentence: “We will change ___, owned by ___, and review ___ on ___.” That is enough to start a useful discussion with your team.
If you want another perspective on what the review shows, book a free consultation with me. Bring the pattern you found; you don’t need to send customer details.
Research behind the approach
This five-enquiry review is my proposed starting exercise. It is not a statistically validated sample size. Nielsen Norman Group’s guide to customer journey mapping explains why teams need to look across the whole customer experience. Google’s lead acquisition documentation separates lead acquisition from qualification and later outcomes. Those distinctions matter when deciding what to improve.