How long each stage should take, measured against your own record

5 days ago
3 min read

Updated: 4 days ago

Introduction

Funnels are almost always counted and almost never timed. The conversion rate between two stages is recorded, and the number of days it took is not, which means the most correctable form of loss is invisible. A buyer does not usually decide against you; they wait, lose momentum, get a faster answer somewhere else, and the funnel records a non-conversion with no explanation.

Timing each stage is no harder than counting it, because the dates are already in the records. What emerges is usually a small number of stages where days accumulate for no reason anybody would defend, and those are the cheapest improvements available in the whole model.

1. How long each stage should take is set by buyer patience, not your workload

The reference point.

The relevant question is how long a buyer will wait before assuming you are not interested or finding someone else. That interval is short at the start of the process and longer once they are committed.

2. Measure your medians before setting any target

The baseline.

Median days per stage from fifty recent records, plus the spread. Use the median rather than the mean, because one dormant enquiry from March will distort an average beyond usefulness.

3. Enquiry to first response is measured in hours

The one stage with a hard answer.

Buyers contact several suppliers and commit early. Same-day is the working standard and within the hour is better. This is the only stage where the target does not really depend on your industry.

4. First response to quote depends on what a quote requires

Where industries differ.

A price from a rate card should take a day. A quote requiring a survey, supplier prices and a drawing may reasonably take a week. What matters is that the buyer knows which, and when.

5. Say how long the next step will take

The intervention that costs nothing.

An expected date removes most of the damage of a genuinely slow stage. Buyers tolerate waiting they were told about and react badly to silence of the same length.

6. Watch the stage where time accumulates with no owner

The usual culprit.

Waiting for a supplier price, waiting for the one person who signs off, waiting for a site visit slot. These are queues rather than work, and queue time is the largest component of most elapsed times.

7. Set an internal service level per stage and report against it

Making it stick.

A target per stage, measured weekly as a median and an outlier count. Speed improvements decay without measurement, reliably and within a quarter.

8. Watch the stages after the sale too

The forgotten half.

Time from order to start, from completion to invoice, from invoice to payment. These affect cash and satisfaction, and they are omitted from most funnels because the funnel is assumed to end at the order.

9. Compare fast and slow deals for their outcomes

The check that motivates the work.

Split your last hundred enquiries by whether the response was fast or slow and compare the conversion rates. The difference is usually large, and it is the argument that gets the change made.

Be careful about compressing a stage that needs its time. Rushing a survey or a proposal to hit a target produces errors that cost more downstream, and the objective is removing waiting rather than removing work.

Conclusion

Time your stages as well as counting them, because delay is loss that no conversion rate explains.

Establish median days per stage from fifty recent records, treat enquiry-to-first-response as a target measured in hours, set the quote timescale by what a quote genuinely requires and tell the buyer which, give an expected date at every stage because told-about waiting is tolerated, find the queues where time accumulates with no owner, report against a per-stage service level weekly, extend the timing to the post-order stages, and compare conversion rates for fast and slow responses.