Don't Buy Clicks You Can't Answer: Ad Scheduling for Service Businesses

Most service businesses set a daily ad budget once and let it run flat across every hour of every day. The ads keep serving at 9pm on a Sunday, at 6am on a Tuesday, and all through the two hours your one office person is at lunch. The clicks cost the same as the ones at 10am on a Wednesday. The difference is that nobody picks up the phone.

Your ad schedule is really a staffing question

Paid search for a service business is not a shopping cart. Almost nothing closes on the page. The click buys you a phone call, a form, or a chat message, and the value of that click depends entirely on what happens in the next few minutes. If the answer is voicemail, you paid full price for a lead that is now shopping your competitor.

So the first question is not "when do people search for roofers?" It is "when can we actually respond?" Those two schedules almost never line up on their own. Search demand for most home services runs early in the morning and again after dinner, which are exactly the hours when a small office is unstaffed. Med spa inquiries spike in the evening, when someone is on their phone in bed and the front desk closed at six.

Write out your real coverage before you touch the ad platform. Not your posted hours — your actual ones. Who answers between 12:00 and 1:00? What happens to a call at 5:45 on a Friday? Is Saturday morning a real shift or does it depend on who volunteers? Most owners discover two or three coverage holes they had never named, and those holes are where the budget is leaking.

Read the hourly data before you cut anything

Google Ads and Meta both break performance down by day of week and hour of day. Pull that report for a period long enough to be meaningful — a full quarter if your volume is modest — and look at three columns together: spend, conversions, and cost per conversion. Do it per campaign, not per account, because a brand campaign and a generic-service campaign behave nothing alike.

What you are looking for is a block of hours that takes a real share of spend and returns very little. That is the candidate. But before you cut it, check two things. First, whether the tracking is honest: if your conversion setup only counts form fills and most of your evening inquiries come by phone, the evening will look dead when it is actually your best window. Fix the measurement before you act on it. Second, whether the volume is large enough to mean anything. A single Sunday with two clicks and no calls is noise, not a signal.

Also separate two very different problems that look identical in the report. Hours where nobody is searching with intent are a demand problem — cutting them saves money and costs you nothing. Hours where people are searching and you simply fail to answer are a response problem, and cutting them throws away demand you could have won. The fix for the first is a schedule change. The fix for the second is coverage.

Three ways to close the gap, in order of effort

The cheapest fix is to stop buying the hours you have decided not to cover. In Google Ads this is an ad schedule at the campaign level; in Meta it requires a lifetime budget on the ad set. Turning off a genuinely dead block redistributes that money into hours that convert, which usually shows up as a lower blended cost per booked job rather than more leads.

The middle option is to bid differently instead of going dark. If an hour produces leads but expensive ones, a downward bid adjustment keeps you present at a price that makes sense. This is the right move for the hours you are unsure about, and for anything seasonal where today's dead window becomes next month's busy one.

The most valuable fix is to extend coverage so you can keep buying the demand. That does not have to mean hiring. It can be a missed-call text-back so every unanswered call gets a reply within seconds, an AI first responder that answers the common questions and books straight into the calendar, or an after-hours routing rule that sends emergency calls to a phone somebody actually carries. If the hour produces real inquiries, buying coverage almost always beats turning off the ads — you are paying to keep demand rather than paying to avoid it.

One practical caution: if your account runs on automated bidding with limited conversion volume, aggressive hour-by-hour restrictions can starve the algorithm of the data it needs. Cut in blocks, change one thing at a time, and give it two or three weeks before you judge the result.

What to automate, and what stays your call

Automate the mechanics. The hourly performance pull should land in your inbox on a schedule instead of being something you remember to check. Missed-call text-back should be instant and unconditional. After-hours acknowledgement — a message that confirms the inquiry arrived and says when a human will follow up — should never depend on someone being awake. Bid adjustments inside a range you have already approved can run on rules.

Keep the judgement. Deciding that Saturday morning is worth staffing is a business decision about where you want to grow, and no report will make it for you. Deciding whether an expensive hour is genuinely low-intent or just badly measured takes someone who knows the phones. And deciding which calls a bot is allowed to handle before a human takes over — the emergency, the angry customer, the six-figure commercial inquiry — is a judgement about your brand, not a setting.

The underlying rule is simple enough to check tomorrow. Look at your last month of spend, hour by hour, and ask which of those hours you were genuinely able to answer. Every hour where the answer is no is either a schedule to cut or a gap to cover. Both are cheaper than what you are doing now.