How to Time a Follow-Up So It Actually Lands

Why this matters

Two shops send the same follow-up message with the same offer to the same kind of customer, and one gets several times the response. The difference is almost always when it arrived. Timing is not a refinement applied after the message is written; it is the larger of the two variables, and it is the one a shop can set from its own records instead of guessing. The common failure is a fixed drip, every customer contacted at 3 days and 30 days and 90 days, which guarantees that most contacts land at a moment when the customer has no reason to act.

The principle: find the clock the decision runs on

A customer is not deciding continuously. They decide when something changes: the annoyance returns, the season turns, the budget opens, the warranty ends. Your follow-up lands when it arrives shortly before that change, not at a fixed offset from your last visit.

So the work is identifying which clock a given follow-up sits on. There are three that cover almost all field-service follow-ups, and they have very different intervals.

The annoyance clock. A deferred repair the customer chose not to do. The decision changes when the symptom comes back or gets worse. Interval is short, days to a few weeks, and it is governed by how often the fault recurs.

The demand clock. Seasonal or interval-driven service. The decision changes when the load arrives or when the interval comes due. Interval is long and is set by your own booking curve, not by the calendar date of the season.

The budget clock. A large capital item the customer wants but cannot fund yet. The decision changes at a predictable financial event they told you about, or at the point their existing equipment forces it. Interval is months, and the follow-up should be scheduled to the event they named rather than to a cadence.

Sorting a follow-up onto the wrong clock is what produces the polite non-answer. A capital replacement quote followed up on the annoyance clock, at 3 days, reads as pressure. A deferred repair followed up on the demand clock, six months later, arrives after they have already had somebody else do it.

Step 1: Tag every follow-up with its clock at the moment it is created

This has to happen at the visit, not later, because the tech is the only person who knows which clock applies. Add one field to the deferred-item record: annoyance, demand, or budget. The office cannot infer it from the job type; the same quoted item sits on different clocks for different customers depending on what they said.

If the tech cannot tell, the default is annoyance, because it is the shortest interval and a follow-up that is slightly early can be repeated, while one that is months late cannot be undone.

Step 2: Set the annoyance-clock interval from how often the fault shows itself

The follow-up should arrive after the customer has experienced the problem again at least once, and before they have gotten used to it or called someone else.

Ask the tech one question at closeout: how often will they notice this? A fault that shows up daily means a follow-up inside a week. One that shows up on a heavy-use day means a follow-up after the next likely heavy-use period, often 10 to 14 days. One that shows up seasonally is not on the annoyance clock at all and should have been tagged demand.

A reasonable default when nothing better is known: 12 to 16 days. It is far enough out that the customer has hit the problem again and is past the moment where they felt talked into something, and near enough that your quote is still the reference point.

The failure at 2 or 3 days is specific and worth naming: the customer has not experienced anything new since they declined, so the only thing that has changed is that you asked twice. That is the message they receive, whatever the words say.

Step 3: Set the demand-clock interval from your own booking curve, not from the season

This is the step that produces the biggest single improvement and almost nobody does it, because it requires looking backward at your own data.

Take three years of bookings for the seasonal service in question and count them by week of the year. You are looking for the week where volume starts climbing, not the week where it peaks. The peak is when everyone else is already booked, and a reminder that arrives at peak is competing with your own capacity constraint as well as with the customer's inertia.

Send the reminder roughly 4 to 6 weeks before the climb begins. That window exists because the customer needs time to notice, decide, and get onto a calendar that is not yet full, and because a reminder that arrives too early gets filed as not-yet-relevant and never resurfaces.

Step 4: Schedule budget-clock follow-ups to the event, not to a cadence

When a customer defers a large item for money reasons, they almost always name a horizon: after the holidays, once the other project is done, next year's budget, when the tax refund comes. Write down the horizon they named, verbatim, and set the follow-up to about two weeks before it.

Do not send anything in between. A quarterly check-in on a customer who told you next spring achieves nothing except spending down their tolerance for hearing from you, so that when the actual moment arrives your message is the fourth one and gets treated like the first three.

Step 5: Fix the small timing details last

These matter far less than the clock, and shops obsess over them in the wrong order. Once the clock is right:

• Avoid the day your invoice is due or just came due. A follow-up that arrives beside a payment reminder gets read as a payment reminder.
• Avoid the first and last hour of a working day, when messages compete with the customer's own work.
• For anything requiring a decision with a household partner, avoid mid-week midday and prefer times when both are likely reachable.
• Never follow up during a widespread local outage, storm, or emergency affecting your service area, unless the message is about that emergency.

The worked example

The demand clock. A shop pulled three years of spring service bookings by week of the year and found volume began climbing around week 14 and peaked in weeks 16 through 18. In year one it had sent its reminder in week 16, at the peak. It sent 180 reminders and 22 customers booked, which is 22 of 180, or 12% of that send.

In year two it moved the send to week 10, four weeks ahead of the climb, to the same list size of 180. Sixty-one customers booked, which is 61 of 180, or 34% of that send. That is 61 against 22 on equal list sizes, a factor of about 2.8, from changing nothing but the week.

The obvious objection is that year two might simply have been a heavier spring. The shop checked total spring bookings from all sources across both years and found them within a few percent of each other, which does not prove the timing caused the lift but does rule out the simplest alternative. It also matched a mechanism the office could see directly: in year one, a meaningful share of the customers who responded to the week-16 reminder could not be scheduled inside three weeks because the calendar was already full, and some of those never rebooked.

The annoyance clock. The same shop looked at 96 quotes for deferred repairs that customers declined at the time of the visit, over one year. It had no policy, so the follow-up timing varied by whoever picked up the list. Forty-one of the 96 were followed up within 2 to 3 days, and 5 of those converted, a rate of 12% of that subgroup. Fifty-five were followed up between day 12 and day 16, and 17 converted, a rate of 31% of that subgroup. The two subgroups sum to the 96.

This one deserves more caution than the seasonal result, and the shop treated it that way. The assignment was not random: the fast follow-ups tended to be the ones the office did immediately because the quote was simple, and the slower ones tended to be larger items that sat in a pile. So the 31% group may be converting better because those jobs were more substantial and the customer was more invested, not because of the delay. What the shop could say honestly is that following up fast did not appear to help, which was the opposite of the assumption the office had been operating on, and that was enough to justify testing a deliberate 14-day default on the next year's quotes rather than continuing to treat speed as obviously good.

Note that this is the exact opposite of the rule for a live inbound lead, where speed is decisive and minutes matter. A stranger with an active problem and a declined quote from a shop that already visited are different situations, and importing the speed rule from one to the other is a common and expensive mistake.

What changes the answer

A customer who named a date. Every rule above is overridden by the customer telling you when to call. Their stated date beats your derived interval, always, and the record needs a field for it or the information is lost the moment the tech drives away.

Commercial accounts with procurement cycles. The budget clock is real and formal: fiscal-year boundaries, capital request deadlines, board approval calendars. Ask directly what the cycle is, and follow up to that calendar. Missing a capital request deadline by two weeks costs you a full year.

Trades with true emergency demand. If most of your work is unplanned failure, the demand clock barely exists and effort belongs on the annoyance clock and on being findable. Confirm this with your own booking curve rather than assuming it; many shops that describe themselves as emergency-driven have a clearly visible seasonal shape in their records.

A very short list. With fewer than about 60 sends, a difference between two timings is not measurable and you should follow the mechanism rather than the numbers. Send ahead of the climb because the reasoning is sound, not because a 20-record sample said so.

How to verify you got this right

Check that the reminder preceded the climb, not the peak. Plot last year's sends against last year's bookings by week. If the send week sits inside the high-volume weeks, the timing is wrong regardless of how the response looked.

Check the capacity side of a good response. A reminder that produces bookings you cannot schedule within a reasonable window is only partly a success, and the customers you turned away are more likely to be gone next year. Count how many responders were scheduled inside three weeks.

Check that clock tags are being set at the visit. Pull ten deferred items and see whether the tag was filled in by the tech or by the office afterward. Office-assigned tags are guesses, and they will drift toward whatever interval the office finds convenient.

Check that stated customer dates are being honoured. Find five records where the customer named a horizon and see whether the follow-up landed near it. This is the single highest-conversion follow-up a shop has available and it is the one most often lost between the truck and the office.

References

• Trade-standard practice for quote follow-up and seasonal service scheduling
• U.S. Federal Trade Commission guidance on commercial email and text contact frequency and opt-out handling
• See related: The Quote Follow-Up Checklist
• See related: The Follow-Up Cadence After a Big-Ticket Quote
• See related: How to Write a Win-Back Message That Gets Answered