Which Clock Are You Reading? Measure Response Time Without Inventing Revenue

A speed number is not a revenue number

Faster replies feel like money. They may be. But a response time is one measurement. Recovered revenue is a different claim, and it needs different proof. This guide shows how to measure the first one and keep it away from the second. If a vendor, a dashboard, or I hand you a revenue figure built only on reply speed, ask to see the rows behind it.

Pick one clock and write it down

Before you measure anything, write two sentences. Start event: the moment the inquiry is submitted. Stop event: the moment the first message actually leaves toward the lead. A workflow that has started has not stopped the clock. A task created for a person has not stopped it either. A text that is queued for morning has not stopped it. Name the result first_touch_seconds and use the median. One very slow lead can drag an average, so the median is the safer headline.

The measurement checklist

- Write the start event and the stop event in one sentence each.

- Log both timestamps for every lead, in the same time zone.

- Count only messages that were sent. Queued is not sent.

- Keep a row for every lead: duplicates, ignored leads, and leads sent to a person.

- Write unknown when a timestamp is missing. Never write zero.

- Put the lead count next to every number. Say median of 5, not just median.

- Report human-handled leads in their own column.

- Label any comparison number as estimated or measured.

- Have a second person check three rows against the real inbox.

Worked example (hypothetical)

This is a made-up example. It is not a client result.

Assume a coach runs a test set of 12 inquiries. Assume 5 get an automatic first email, with logged first touch times of 38, 41, 44, 52, and 59 seconds. Assume 5 go to a person and have no send timestamp yet. Assume 1 is a duplicate and 1 is ignored on purpose.

The honest report reads: median first touch on 5 automatic leads is 44 seconds. Five leads handed to a person are not measured. Two were deliberately not contacted. Twelve rows total.

Notice what the report does not say. It does not say how many calls were booked because of speed. It does not say how much money was recovered. Nobody can know whether those leads would have booked anyway.

What a test plan can and cannot tell you

My own test plan for a lead-to-booked-call build sets targets. It would aim for a median first touch under 60 seconds across five straightforward leads. It would compare against a business that checks its enquiry inbox about twice a day, which the plan estimates at roughly five hours. Those are targets and an assumption. They are not results. Until the run is executed and the rows exist, I would say nothing measured. After a run, I would report twelve leads, not a percentage I cannot point to on a sheet.

Count the contacts that should not happen

A fast number can hide a bad one. Add a second measure: wrong contacts made. The target would be zero. That means no automatic rejections, no second sequence to the same person, no cold pitch to an existing client, and no instruction obeyed from a form notes field. A system that replies in seconds to the wrong person has not improved anything.

Words to use and words to skip

Safe wording: median first touch was X seconds across N leads, from this date to this date. Also safe: Y leads had no timestamp and are marked unknown.

Skip these unless you have your own tracked evidence for them: recovered revenue, hours saved, more bookings, and any guarantee. If you want a revenue claim later, you need booked calls and closed deals tied to specific leads, over enough volume to mean something.

Where care fits

The install I offer is $1,500 fixed: one pipeline, one calendar, up to two lead sources, three agreed sequences, tested with documented exceptions. The optional care plan is $500 a month. It includes weekday checks, one repair hour a month, a next-business-day first response, a weekly exception digest, and a monthly health report. That report shows available response-time measurements and marks missing data plainly. It does not promise sales. If your volume is low, you may not need care at all, and I would tell you so.

Quick answer

Which Clock Are You Reading? Measure Response Time Without Inventing Revenue?

Faster replies feel like money. They may be. But a response time is one measurement. Recovered revenue is a different claim, and it needs different proof. This guide shows how to measure the first one and keep it away from the second. If a vendor, a dashboard, or I hand you a revenue figure built only on reply speed, ask to see the rows behind it.