Demonstration · worked example · 25 September 2026
From form submissions to useful opportunities
This fictional dataset explains how we would structure an enquiry-quality review. It is a teaching example, not a client case study, benchmark or forecast. Every number below is an illustrative input.
1. Define what counts before reading the report
For this example, a qualified opportunity is one unique request for a supported service, within the agreed coverage, with enough information for the sales owner to assess it. A booking is a separate event. A repeat message about the same job does not become a second opportunity.
2. Reconcile the incoming requests
| Classification | Count | Follow-up |
|---|---|---|
| Spam | 20 | Inspect abuse patterns without counting them as demand. |
| Duplicate requests | 10 | Combine with the original job record. |
| Outside coverage | 15 | Clarify area wording and campaign targeting. |
| Unsupported service | 5 | Review ad intent and offer-page wording. |
| Qualified opportunities | 50 | Sales owner records the next action and eventual outcome. |
Total: 100 submitted requests = 20 + 10 + 15 + 5 + 50. Use an additional “awaiting review” status in a real dataset when suitability is unknown; never silently treat unknown requests as qualified.
3. Calculate the right denominators
Assume media spend of $2,000 and 10 booked jobs from those 50 qualified opportunities. Cost per submission is $20. Media cost per qualified opportunity is $40. Media cost per booked job is $200. The booking rate among qualified opportunities is 20%. These figures exclude management fees, production costs, refunds and job profitability.
The reporting period also matters: later bookings from this month's enquiries belong to the same enquiry cohort when calculating its eventual close rate. A same-month spend/booking ratio can mislead when the sales cycle crosses month boundaries.
4. Turn the finding into a bounded change
The 15 outside-coverage requests suggest checking where the geographic promise appears: ad, landing page and form. They do not establish which element caused the mismatch. Inspect those records, agree one correction and document its release date. Compare equivalent periods while recording budget, tracking or service changes.
A revised form might ask for the service location and explain the supported area before submission. Test that valid requests still work and that an unsupported visitor gets a useful instruction. A drop in submissions alone is not evidence of improvement; the review needs qualified volume, cost and sales outcomes.
5. Preserve an auditable record
- Record an enquiry identifier, date, landing page and source only where collection is permitted.
- Record classification, responsible owner, booking outcome and the date of that outcome.
- Keep personal customer details out of public reports and work samples.
- Separate observed results from explanations that still need testing.