Enquiry-to-booking conversion: the metric that lies at six months

Enquiry-to-booking conversion means one thing: of the enquiries that arrived in a given month, what share eventually became bookings. The denominator is fixed when the month closes and never changes. The numerator grows for as long as that cohort keeps closing, which in a tailor-made trip business is often 6 months or more.

The ratio most operators report is different: bookings closed this month over enquiries received this month. Those are unrelated populations. Almost none of this month's bookings came from this month's enquiries, so the number describes current volume against past volume, not how well anyone is selling.

Numerator and denominator, precisely

• Denominator: qualified enquiries stamped with an arrival date inside the month. Fixed once the month closes. Spam, duplicates and enquiries for products you do not sell come out before it closes, never after.
• Numerator: bookings from that same fixed set, counted whenever they close. A booking confirming 9 months later belongs to the month the enquiry arrived.
• Neither side moves when a booking is later cancelled. Track cancellation separately — folding it in makes the metric change retrospectively, and nobody can reconcile a report printed last week.

The distortion, in one table

Here is the arithmetic, with round numbers chosen for legibility rather than to represent any real business. Every cohort converts identically: per 100 enquiries, 2 book in the arrival month, then 4, 5, 4, 3, 1 and 1 over the following 6 months. Each cohort ends at 20 per 100, a flat 20%, and nothing about sales performance changes anywhere in the table.

Only volume changes: 60 enquiries in January, then 80, 120, 160, 200, 240, and a collapse to 120 in July.

June is the best month the business has had — 240 enquiries, 29 bookings closed — and reports 12.2%. July is the month the pipeline halved, and reports 27.8%, because July's bookings are mostly June's large cohort divided by July's small one. A manager watching the calendar ratio congratulates the team in the month the business started dying.

Three definitions the number depends on

A conversion rate is only as stable as the 3 terms underneath it, each written down before the first report runs.

1. Qualified enquiry. One carrying enough to be worked: a destination, a dated window, a contactable person. Everything else is traffic. Apply the rule at arrival, because applying it later lets someone shrink a denominator after seeing the result.
2. Quoted enquiry. One where a priced proposal reached the client. Quoted-to-booked is the consultant's number; enquiry-to-booked is the business's. Reporting one and calling it the other is how these figures get misread.
3. Lost enquiry. A cohort member no longer a live prospect, under a rule that fires on its own rather than when someone tidies the pipeline.

Declaring an enquiry lost, on a rule rather than a mood

Without a cut-off a cohort never finishes: an 18-month-old enquiry books, the historic rate ticks up, and last year's reports quietly change. Two shapes are defensible — a fixed age since the last client-initiated contact, or a departure-window rule closing any enquiry whose dates have passed. Either way a lost enquiry that comes back is a new enquiry joining today's cohort, not a resurrection.

The other half is making booking state flow back to the enquiry record. Where it does not, the enquiry sits open in the report and open in the follow-up tooling, producing a wrong metric and the awkwardness in the chase sequence keeps emailing clients who booked.

Reporting a cohort that has not finished

You cannot wait 6 months to know anything, so report every cohort at a stated lag: this month at lag 0, last month at lag 1, and so on. Comparing April at lag 3 with January at lag 6 is the second most common error after the calendar ratio, and the one that makes recent months always look worse. Dates carry this hidden semantics throughout a trip business — the same trap where a day index is not a calendar date is in days are not dates in a trip model.

A cohort rate at lag 3 and one at lag 6 are different metrics with the same name. Label the lag or do not publish it.

Once cohorts are correct the breakdowns become worth having. Cohort rate by consultant separates skill from assignment, which every enquiry lands on the same consultant traces, and tells you whether the method in assigning enquiries by destination and load is working. None of it needs a warehouse: one enquiry table with an arrival date, a lag dimension and a view is a few days of product build work.

Two adjacent questions belong elsewhere: attributing a booking to a channel, and chase timing, which is derived from booking lead time rather than from this metric. Both sit in from enquiry to confirmed booking, with the rest of our travel and tourism work.

• conversion
• cohort reporting
• sales metrics
• enquiry data