Following Up on Unsold Estimates Is the Cheapest Revenue a Door Company Has

Somewhere in your phone, your CRM, or a stack of paper quotes, there is a folder of estimates you wrote and never heard back on. Nobody canceled them. The customer did not say no. The job just went quiet, and it has been sitting there long enough that calling now feels awkward.

We build and sell a system that follows up on exactly these estimates, so we are not a neutral party in this argument. We think it is worth doing, and this article is the reasoning, with the sourced parts labeled and the invented parts labeled too. The short version: a sitting estimate is a customer who already asked you to come out, already heard your price, and already spent real time with you. Getting a fair share of those jobs back costs almost nothing next to what it costs to generate a brand-new lead. Most shops are leaving that revenue on the table not because the customer said no, but because nobody followed up in a way that gave them a reason to say yes.

The arithmetic nobody runs

Here is a worked example. The numbers are round on purpose, to make the math easy to follow, not a claim about what any particular shop does. Swap in your own figures and the logic still holds.

Say a five-truck door company writes 60 estimates in a month and closes 40 percent of them on the spot or within a few days. We picked that figure for the example, not from a study, since we could not find a credible, sourced close-rate benchmark for garage door estimates specifically. Use your own close rate if you track one. That leaves 36 estimates that did not close right away. If even a third of those are reachable, still interested, and simply never got a second touch, that is 12 jobs. At a modest job value, say a spring repair in the range consumer cost guides put it in ($119 to $540, depending which guide you check, since This Old House, Forbes Home, Angi, ConsumerAffairs, and HomeAdvisor all publish different numbers and none of them are a survey of real invoices), 12 recovered jobs is somewhere between roughly $1,400 and $6,500 in a month. Compare that to what it costs to generate 12 brand-new leads through paid search or local ads in a competitive market, and the estimate folder usually wins, because you already paid for those leads once.

This is not a claim that every shop will recover a third of its unsold estimates. It is a claim that the marginal cost of trying is close to zero, because the lead acquisition cost was already spent. A follow-up program does not need to work spectacularly to beat the next dollar of ad spend. It needs to work a little.

Run the same math against a bigger job. A door replacement or a two-car opener upgrade runs well into four figures, and a handful of those sitting quiet in the estimate folder is worth more than a month of ad spend by itself. The point is not the exact dollar figure, which depends entirely on your job mix and your prices. The point is the comparison: a new lead costs whatever your marketing costs to produce it, plus the time to quote it, plus the odds it closes at all. A stalled estimate has already absorbed all three of those costs. Whatever it takes to follow up on it, that follow-up is cheaper than the lead was.

There is a bigger backdrop here too. The overhead and garage door services market is projected above $16.0 billion in 2026, with more than 114 million doors in service nationally, and it is currently being consolidated by private equity at a real pace, not a hypothetical one: more than ten PE-backed platforms have formed since 2022, and Oak Hill Capital’s agreement to acquire Guild Garage Group, announced in March 2026, was reported at an enterprise value above $800 million. Those larger, professionalized competitors are not going to outspend a small independent on follow-up discipline. They are going to build it into their systems the same way any well-run operation does. An independent shop does not need to match a PE-backed platform’s ad budget to compete with it. It needs to stop losing the jobs it already generated.

Why a sitting estimate is not the same problem as a new lead

Most of what gets written about response speed is about brand-new leads, the phone call or web form that just came in. A 2011 Harvard Business Review audit of 2,241 US companies found that 23 percent never responded to a web lead at all, and the companies that did respond took an average of 42 hours to do it. A separate 2007 study by InsideSales.com and MIT Sloan researcher James Oldroyd, covering six B2B companies and more than 15,000 leads, found that the odds of qualifying a lead fall by a factor of 21 between a 5-minute callback and a 30-minute one, and by more than sixfold within the first hour. That study was commissioned by a company that sells lead-response software and covers a small number of companies, so treat the exact multiplier with some skepticism. The direction of the finding, that speed to first contact matters and decays fast, is consistent with the independently published HBR audit above.

An unsold estimate decays too, but on a longer curve, because the customer already spent real time with you. They let you into the garage, watched you look at the springs or the opener, and heard a number. That is a different relationship than someone who filled out a form and has heard nothing back. The decay is not measured in minutes for an estimate, it is measured in weeks, and the same principle that makes speed to lead worth the effort on a brand-new inquiry (covered in more detail in our guide to speed-to-lead for garage door companies) applies here too, just on a slower clock. The mistake is treating a two-week-old estimate the same way you would treat a new lead that just came in, which usually means either calling too aggressively too soon, or forgetting about it entirely once the initial excitement of the appointment wears off.

The three real reasons an estimate goes quiet

Almost every stalled estimate falls into one of three buckets, and they need different messages.

Price. The number was higher than they expected, or higher than a competitor’s number, or just higher than what they had budgeted for a garage door repair, which most homeowners do not plan for. The right message here is not a discount by default. It is often a financing option that changes the monthly number without changing your price, which is its own topic covered in our guide to point-of-sale financing for garage door jobs.

Timing. They are not ready, the money is tied up in something else this month, or the job is not urgent enough yet. These customers usually do come back, on their own schedule, and the job of a follow-up sequence is to be the name they remember when they are ready, not to rush them.

A second opinion. They got, or are getting, another estimate. This is the one owners take most personally and handle worst, usually by either going silent (ceding the job) or by badmouthing the competitor (which reads poorly and rarely works). The better move is to make your own value plain: what is actually included, how the price was built, and what a lower number from someone else might be leaving out. A written, consistent price structure makes this conversation much easier to have without sounding defensive, which is the subject of our guide to building a flat-rate price book.

You will not always know which of the three you are dealing with. That is fine. The cadence below is built to surface the reason rather than assume it.

A cadence that will not annoy anyone

This is a starting structure, not a rule from a study. Adjust the spacing to your own market and job mix.

• Day 2. A short, low-pressure text or call. Confirm they received the estimate, ask if they have questions, and stop there. This is not a sales call, it is a check that the estimate landed and made sense.
• Day 7. A more substantial touch. This is where you address the likely objection directly: a financing option if the job was pricey, a specific answer to a common concern, or simply a reminder of what is included in the price that a lower competing bid might not include.
• Day 21. The last scheduled touch, and it should not be “just checking in.” See the next section for what actually belongs here.
• Then stop, unless the customer replies, books, buys from someone else, tells you no, or asks to be left alone. Every one of those outcomes ends the sequence. A sequence with no stop rule is the fastest way to turn a warm customer into someone who blocks your number.

Writing the stop rules down before anything sends matters more than the exact day count. A stop rule you define after a customer complains is a stop rule you needed a week earlier.

What to say on day 21, since “just checking in” is not it

By the third touch, the customer has already heard from you twice. A repeat of the same message reads as nagging. What usually works better is something specific to the door itself or the season: a note that a part quoted at a certain price has a manufacturer lead time that is starting to matter, a mention that the current weather or the coming season is the reason a spring or opener issue tends to get worse rather than better on its own, or a straightforward statement that the estimate is still valid for a defined window and then it will need a fresh look, since material costs do shift.

We are intentionally not citing a specific seasonal-surge statistic here. Versions of “service calls jump X percent in the summer” circulate widely in garage door marketing content, and we could not trace any of the specific numbers to a source we would stand behind. The general pattern, that heat stresses springs and openers and storms drive emergency repairs, is plausible and consistent with what shops report, but treat any precise percentage you see elsewhere with real skepticism.

Financing closes some of these, not all of them

Point-of-sale financing is a legitimate tool for the price objection specifically. It turns a large opener or door replacement into a monthly payment a customer can say yes to today instead of waiting to save up, and offering it as part of the day-7 or day-21 message can move a genuinely price-stalled estimate. It does not do anything for the timing or second-opinion buckets, and it does not fix a follow-up that never happened in the first place. A financing option nobody ever mentions to the customer closes zero jobs. The sequence has to exist before the tool inside it matters.

Where this has to live so it survives a busy week

The single biggest reason follow-up sequences die is that they depend on a person remembering to run them, and that person also answers phones, writes estimates, and occasionally installs doors. A list in someone’s head, or a spreadsheet that only gets opened when things are slow, will not survive the first genuinely busy stretch, and garage door work has a lot of genuinely busy stretches.

The fix is structural, not motivational: the follow-up has to be something the system generates and tracks, not something an owner has to remember to check. That means every unsold estimate gets a next action and an owner automatically, overdue touches show up somewhere visible instead of quietly not happening, and the sequence stops itself on the rules above rather than relying on someone to notice a reply came in. We go through what that setup actually looks like, tool by tool, in our guide to setting up an automated lead follow-up sequence, and the underlying argument for why systems outperform memory and good intentions is laid out more fully in why leads don’t convert, systems do.

The part that compounds: a closed follow-up produces a review

A job that gets recovered through follow-up is a completed job like any other, which means it is also a legitimate opportunity to ask for a review once the work is done. Google states plainly that local ranking is based on relevance, distance, and prominence, and that prominence is influenced by how many reviews a business has and how positive they are. That is not a claim about how many leads a review produces, it is Google’s own description of one of the three factors behind whether your business shows up when a homeowner searches for a garage door company nearby. Every follow-up job you close is also, indirectly, a small contribution to that.

How to know if it is working

Three numbers, reviewed monthly, tell you whether the sequence is actually doing anything:

1. Unsold estimate count. How many estimates are sitting in the pipeline with no decision, at any given time. If this number only grows, the sequence is not running or is not reaching people.
2. Touch completion rate. Of the estimates that should have received a day-2, day-7, or day-21 touch, how many actually got it. This is the number that tells you whether the system is working, separate from whether the follow-up is effective.
3. Reactivation close rate. Of the estimates that got the full sequence, what share eventually closed. This is your own shop’s number, not an industry benchmark, because we could not find a credible, sourced industry-wide figure for this and would not want you comparing yourself to an invented one.

Track these three for two or three months before judging the program. A single slow month tells you little, and estimate volume can swing enough month to month in this trade that a single busy or slow stretch will make the reactivation rate look better or worse than it really is. What you are watching for is a trend across a full season, not a verdict from four weeks of data.

Write the numbers down somewhere you will actually see them again, not just in the moment you calculate them. A monthly note in whatever you use to run the business, even a plain spreadsheet, is enough. The point is not precision. The point is having a real answer the next time someone asks whether the follow-up sequence is worth the trouble, instead of a guess based on how the last two weeks felt.

Where this does not help

Follow-up recovers estimates that stalled for a reason unrelated to the estimate itself: bad timing, a competing quote, sticker shock that financing could have solved, or simple forgetfulness. It does not rescue an estimate that was wrong. If the price was genuinely out of line with the market, if the scope missed something the customer actually needed, or if the customer already hired someone else and the job is done, no amount of well-timed follow-up changes the outcome. It also does not replace demand generation. A shop writing too few estimates in the first place has a lead-volume problem, and reworking the follow-up sequence on 20 estimates a month will not fix that; that is a different conversation, about how many quotes are coming in to begin with.

What to do this week, without buying anything

You do not need new software to start. Pull your estimates from the last 60 to 90 days and sort out the ones with no clear yes or no. For each one, write down which of the three reasons it probably stalled: price, timing, or a second opinion. Draft one message for each reason, the kind you would actually want to receive, not a generic “checking in.” Then work the list by hand this week: a short text or call for each one, following the day-2, day-7, day-21 shape above, with your own stop rule decided in advance. You will learn two things fast: how many of these estimates were genuinely still live, and how much manual effort it takes to run this by memory. That second number is usually the argument for building it into a system before the next busy season buries the list again.