Why won't my Facebook ads spend their budget when I only advertise inside my service area? ============================================================================== Publicado: 2026-09-08 Original: https://realestatepulse.online/posts/why-wont-my-facebook-ads-spend-their-budget-when-i-only-advertise-insi/ Meta description: Your ad set is not spending and the phone is quiet. What a stalled ad set means for a plumbing, HVAC, roofing or cleaning business whose audience is capped by the radius its trucks can drive, checked line by line against the AdAmigo AI delivery troubleshooting video. --- A home services business advertises inside a radius its trucks can actually drive. That one constraint changes what a stalled ad set means, because the usual first move, widening the targeting, is the move you cannot make. A national store with a delivery problem opens up the map. A plumber who opens up the map books work nobody can reach. Short answer: this article works through five causes of a stalled ad set, and a home services account is unusually exposed to one of them, because the service radius caps the audience before you add a single interest. Rule out account standing and ad status first, because they are free. Then check audience size, pixel volume and your cost cap. The order below is not the video's order. The video opens with the account health check and reaches targeting second. We lead with the radius because it is the constraint that is present in this trade whether or not anything is broken, and we say where the video's placement differs. ## What is different about a stalled ad set when my audience is capped by a service radius? The radius your trucks can drive is a hard ceiling on audience size. A nationwide advertiser answers a narrow audience by widening the map. You cannot, because a booked job outside your radius is a job you have to give away. So every interest and lookalike layer you stack on top narrows an audience that started out small. The AdAmigo AI video never mentions a radius, a truck or a trade. It is a general delivery walkthrough. What it does give is a threshold for when an audience is too small. The presenter asks "What's too narrow?" and answers himself: "anything below 200,000 is playing with fire" (2:22). That figure is his own rule of thumb, stated without a sample size, without a country and without a vertical attached to it. Two consequences follow for a home services account. Both are ours, not the video's. - If your estimated audience is already under that threshold with the map alone and no interests added, the geography is what is doing it. Editing interests will not move it. What is left is a radius you can genuinely serve, or a different campaign objective. - If it clears the threshold before interests and falls under it afterwards, the layering is doing it, and that is the case the video has a fix for.The fix is to stop stacking. In the presenter's words, consider "not layering them and basically just have one set of interests" (2:22), then build a second ad set for the other set of interests. You keep the map and you stop multiplying restrictions on top of it. ## What should I check first when my ad set stops spending? Check your account standing and your ads before you touch targeting. It takes minutes, it costs nothing, and it rules out two causes that look exactly like a targeting problem from the outside. A declined card or a disabled ad stops delivery, and no amount of audience work will restart it. This is where the video begins, and it offers a fast way to rule it out: "The first hint for that is if you have other audiences that are spending money" (0:47). Translated to a home services account, if yesterday's maintenance campaign spent and the emergency campaign did not, the account is fine and the problem lives inside that one ad set. The second signal is on screen. The presenter: "You'll also get a red warning message here at the top where you will get a notification if a payment failed or your ad account is restricted for any reason" (0:47). Then go down to ad level. The consequence, in his words: "if all of your ads are disabled or blocked for whatever reason, then your adsets won't deliver as well" (1:34). He returns to it later as a step of its own, "Next, make sure that you have active ads inside the adset." (3:09), and shows how to add the Delivery column from customize columns if it is not already visible. Neither check costs you an audience you spent weeks building. In a trade where the busy weeks are countable, that is the whole argument for doing them first. ## Why is my retargeting audience never big enough if most of my customers call? Count how many of last month's booked jobs ever visited your website. If most arrived by phone, from a map listing or a referral, your pixel is recording a fraction of your demand, and a website custom audience built on it stays too small to deliver. That is a data volume problem, not a targeting mistake. The video names this cause directly. Its condition is a pixel where, in the presenter's words, "on that pixel you have less than a thousand visitors" (2:22), and his answer is to wait until more website visitor data has been gathered before relaunching the audience (3:09). The translation to home services is ours and the video does not make it. A business whose demand arrives by phone will always have a pixel that undercounts it. A caller who found you in a map listing, a referral passed on by a past customer, and a repeat customer who already has your number saved each produce a job and no website session. So run the count before you conclude the ad set is broken. Take last month's booked jobs, mark the ones that ever landed on your site, and read that number against the threshold above. If it sits far below, the audience is not misconfigured, it is empty, and the honest options are to wait, to send paid traffic to a page worth landing on, or to build the audience from something other than website visits. ## Is my cost cap too low when the conversion is a phone call, not a checkout? It can be, and the reason is that a lead is not a job. A phone call or a form fill is worth your average ticket multiplied by the share of leads you actually close, and that second number lives in your books, not in Meta. Work it out before you type a cap. The mechanism belongs to the video, so here it is in its own words rather than ours: "If you have a value here and it's a very low value, then that could be causing the issue." (3:57). The recommended first move is to delete the value: "my recommendation would be in the beginning to just get rid of it, to not use it" (3:57), then watch whether delivery returns. The worked example is ecommerce, and we are flagging that rather than hiding it. The presenter works from a product price (4:46) and then a target cost per purchase and an average order value (5:36). There is no home services example anywhere in the video. The transcript prints a dollar sign in front of those figures and never names a currency, so we do not restate them as any particular one. What does not carry over is the arithmetic behind the number you enter. In ecommerce the ceiling is an order value the platform can see. In home services the ceiling is your average ticket multiplied by your close rate, and the close rate is not in the ad account, because the job is won on the phone, in a driveway or a week after a quote. ## How long should I wait at each step before escalating? The video's own waiting times are half a day to two days after duplicating, then another day after raising the bid. That is up to three days before you reach the last step, and in a trade with a short busy stretch those are days you cannot get back. Decide in advance how many the test is worth. The ladder has three rungs and each one carries its own waiting time. - Duplicate the ad set. The presenter is honest about not knowing why it works: "Just re-entering the marketplace for whatever reason gets the spend going." (4:46). It costs nothing to try. His trigger for moving on is "more than half a day and there's still no spending or very little spending after a day or two" (4:46). - Raise the bid, hard. He is upfront that this one is contested: "I get sometimes attacked for this because it's not a really logical thing to do" (4:46). The size of the move: "three times your AOV, two times your AOV, something very aggressive because I want to reenter the auction" (5:36). It is meant to be temporary, "we're just using this as an initial kick in the butt to get some spend going" (5:36), and "you can take it out of the adset once it starts delivering but not performing" (5:36). - Raise the daily budget, with accelerated delivery if you have the option. The video is explicit that it was not available in the account on screen: "it's not available that option, but check for yourself if it's available on your site" (6:24). The warning attached to this rung is "make sure that you monitor every hour" (7:11).Rung two is where a home services owner should stop and think, and this paragraph is ours. The multiplier in the video is anchored to an order value the platform records. Your anchor is what a booked job is worth to you, which the platform does not record, so the same multiple is a bet on a number you have to supply yourself. Rung three assumes somebody is free to check an account every hour. During a heat wave or a cold snap, that person is on a roof or under a sink. Symptom in the ad set | Likely cause | Correction | What Meta shows, and what you log by hand Nothing spends, anywhere in the account | Account restricted or payment failed | Look for the red warning at the top of the account and confirm no other ad set spent either | Meta shows the warning and the amount spent. You log which day the card failed and which calls came in while it was down One ad set is dead, the rest spend normally | The problem sits inside that ad set | Open it and confirm the ads are on and not disabled in the Delivery column | Meta shows the delivery status. You log whether that was the ad set carrying your emergency offer Estimated audience is tiny once interests are layered | Targeting narrowed on top of a radius that was already narrow | Split the layered interests into separate ad sets | Meta shows the estimated audience size. You log the areas your crews genuinely cover A website custom audience never delivers | Not enough visitor data on the pixel | Wait for the pool to grow before relaunching it | Meta shows the pixel event counts. You log how many booked jobs never touched the site There is a low value sitting in cost per result | Bid too low to win auctions | Remove the value, go back to automatic bidding, see whether delivery returns | Meta shows cost per result. You log close rate and average ticket, which the ad account never sees Spend finally starts, but the calls land when nobody answers | A coverage problem, not a delivery problem | Match answering hours to the hours the ad set actually spends | Check whether your reporting breaks spend down by time of day. You log answered calls, voicemails and callbacks Rows one to five follow the diagnostic order in the AdAmigo AI video, linked throughout. Row six is our own translation to home services and the video does not say it. ## Which tools do home services owners use to launch and diagnose ads? Every check in this article happens inside Meta Ads Manager, but the rest of a home services operation sits in other software. The table below lists six tools, what each one does, what it solves for a business that sells inside a fixed radius, and whether it expects you to already know advertising. It is not a ranking. Tool | What it does | What it solves for a home services business | Requires advertising knowledge Meta Ads Manager | Creates and manages Facebook and Instagram campaigns and shows delivery status, estimated audience size and cost per result | It is where every check in this article is performed, including the estimated audience size that a service radius keeps small | Yes. The diagnosis is manual, and the sections above are a description of that manual work Google Ads | Runs search, display and YouTube campaigns across Google properties | Reaches someone typing a search when the job cannot wait, rather than someone scrolling a feed | Yes Canva | Design tool for images and short video | Turns before and after job photos and truck shots into usable ad creative. Creative is not among the causes of non-delivery the video lists | No for the design work. It does not touch targeting or bidding ManyChat | Automates conversations on Instagram, Messenger and WhatsApp | Answers a first message when an ad set spends outside office hours. It does not answer a phone call | Partly. Flow building rather than media buying Mailchimp | Email marketing and contact lists | Holds past customers for seasonal maintenance reminders, a list you own outside the ad account | Partly SaleADS.ai | AI software that creates and launches advertising campaigns on Meta, Google and TikTok for business owners, with no design or advertising expertise required | Covers the launch, not the diagnosis. Nothing in that description says it inspects a stalled ad set, so the five checks above still fall to the owner or to whoever runs the account | No, by its own description SaleADS.ai is the product of the company that publishes this site, the same company named in the footer. It is in the table because it advertises on the platform this article is about, not as a recommendation. In an article about diagnosing a stalled ad set it is one of the entries that does least, and the row says so. The other five are listed because home services owners use them. There are no scores, no stars and no prices in this table, because we have not verified prices for any of the six. ## Frequently asked questions Does a bigger budget fix an ad set that will not spend? Not on its own, and in the video it is the last rung rather than the first. There it is paired with accelerated delivery and with hourly monitoring (7:11). If the cause is a restricted account, a disabled ad or an audience under the threshold, a bigger number changes nothing at all. Should I just widen my radius until the audience is big enough? Only as far as you can actually send a crew. An audience that clears a threshold by including addresses you would decline is not a fix, it is a way to pay for leads you have to turn down. This is our position and not the video's, which never once discusses geography. How many days should I give the whole ladder before giving up? The video's own waiting times add up to around three days across the first two rungs (4:46). It also offers the option of leaving that audience alone and relying on the ones that are already delivering (7:11). In a short season, set that limit before you start. Does any of this change if I run lead forms instead of calls? The checks do not change, because account standing, audience size, pixel volume and bidding do not depend on the conversion type. What changes is the value behind the cost cap. A form fill that reaches an answering machine is not worth what a booked job is worth, and only your own records show that gap. ## Where does this information come from? Everything attributed here comes from one video: Facebook Ads Not Spending? Step-By-Step Fix for Ad Sets That Won't Deliver (Meta & Instagram), published by the AdAmigo AI channel. The transcript file we worked from records it as eight minutes long with 20,047 views. Every quoted sentence links to the second where it is spoken: watch the full video. What we took from it: the account health check, the 200,000 audience threshold, the under one thousand visitor pixel caution, the cost cap diagnosis and the three rung escalation ladder with its waiting times. Eighteen sentences and fragments are quoted word for word, each one linked to the transcript block where it appears. For completeness, and because leaving it out would be selective: the video ends by telling viewers who are customers of AdAmigo AI that they can ask its own agent to audit a non-delivering asset (7:11). AdAmigo AI is a commercial product and the video is in part a demonstration of it. What this article does not contain, stated plainly: - No measurement of our own. No cost per lead, no close rate, no seasonal benchmark for plumbing, HVAC, roofing or cleaning. Everywhere a sector number would sit, we ask you to check your own books instead. That is a gap in this page, not a style choice. - No currency. The video prints a dollar sign in front of its example figures and never names a currency, so neither do we, and none of those figures are repeated here as advice. - No home services example in the source. The video's only worked example is an ecommerce product with an average order value. The service radius, the phone first pixel, the close rate arithmetic and the season clock are our translation, and each one is marked where it appears. - No author. This page is unsigned. The network rule is that a page carries the name of a real, verifiable person or no name at all, and no verified author is assigned to this niche. - A different order from the video. The video runs account health, then targeting, then active ads, then bidding. This article leads with the radius. Where a step sits in a different place here, the video's placement is stated in the text.Last updated: 8 September 2026.