Ads Manager shows you a cost per result. Your appointment book shows you who actually walked in with a dog. In a grooming salon those two numbers rarely match, because the booking happens on the phone, the money arrives across repeat visits, and the table only holds so many dogs a day. Here is how to read the gap.
Why does the cost per purchase in Ads Manager mislead a grooming salon?
Because it prices a first visit, not a client. A groom client who comes back every few weeks is worth several visits, and the platform figure may be counting only that first transaction. Compare your cost per new client against your own ticket multiplied by repeat visits, not against the price of one groom.
The video source describes the same shape of problem for a product that bills monthly. Ben Heath says the return on ad spend calculation shown inside a Facebook ad account "is not going to be that accurate because it might just include the first transaction" (3:57). A little earlier in the same stretch he works the arithmetic aloud for his own offer, asking "would we pay £23 cost per purchase to acquire a customer that's worth £97 per month" (3:10).
Those pounds are his course, not your shop. What transfers is the shape of the calculation:
- Take your average ticket for one groom, from your own invoices.
- Take the average number of visits a client makes before they stop coming, from your own client list.
- Multiply. That is the number your cost per new client has to beat, not the price of a single appointment.
What number should a grooming salon optimise its ads around?
One number, and it is cost per booked appointment, taken from your own calendar and not from the platform. Everything else, cost per click, reach, video views, is diagnostic. If you cannot tie an ad to a name in the appointment book, you cannot judge that ad.
The source points advertisers in general at the same pair of metrics. Heath says the two he wants most optimisation decisions based on are "either your cost per conversion or your return on ad spend and which one you use depends on your business" (12:41).
In a salon, return on ad spend is usually not available, because no cart fires when a client books a puppy trim by text. So cost per conversion is the one you get, and you build it by hand.
How do I count the clients who call or message instead of booking online?
By logging them by hand at the front desk. Ask every new client how they found you, write the answer next to the appointment, and total those answers every week. This is a method this site recommends, not a measured result. It is what turns a phone booking into a number you can divide ad spend by.
The principle behind it is in the source. Heath frames the target metric as the best cost per result "if that's as far down the sales funnel as you can accurately track" (13:29).
For a salon the honest end of accurate tracking is often the click or the message, not the sale. So the front desk closes the rest:
- One column in a spreadsheet: date, name, source answer, service booked, price.
- Weekly total of ad spend divided by new clients whose source answer was the ad.
- The same sheet read again after two of your own rebooking gaps have passed, to see how many of those names came back.
How long should I run a grooming campaign before deciding it failed?
Long enough to cover one full rebooking cycle plus the lag between seeing an ad and picking up the phone. Check your own appointment history for the average gap between a client's visits. If your own history says that gap is six weeks, a two week test tells you almost nothing about revenue.
The source sets no window for local service businesses. What it does say is that which secondary metrics you dig into depends on the case: "now what we look at here Will depend on the business" (14:17).
Our reading, as a site, is that a salon should hold two clocks at once. A short clock for whether the ad produces enquiries at all, and a long clock, set by your own rebooking gap, for whether those enquiries turn into repeat clients. Killing an ad on the short clock can throw away the only evidence that would have shown whether it worked.
What do I measure when the grooming table is already booked solid?
Cost per booked slot stops being the goal once every slot is full. Measurement shifts to two things, which hours your book still shows empty and how often your small local audience is seeing the same ad. A short service radius means a small audience, and frequency climbs faster on a small audience.
Frequency is the metric that catches this, and the source defines it as impressions divided by reach. Heath states his own working preference for cold audiences: "if you're operating to cold audio is I don't like to see a frequency above 2.5" (4:44). The automatic transcript mangles "cold audiences" into "cold audio is". The 2.5 is his rule on his own accounts, not a measured finding, and it has nothing to do with pet care.
Treat it as a condition to test against your own account. If your radius covers a few square miles and your frequency climbs past the level where enquiries slow down, you have exhausted the neighbourhood, not the ad.
Which numbers tell me why a grooming ad worked?
The secondary ones, read in pairs and only after you know which ad produced bookings. A high view rate with few calls points at the offer. Plenty of clicks with no bookings points at the page or the phone. None of these should be the metric you switch an ad off on.
The source runs this pairing logic on its own example campaign, a retargeting campaign for the author's course, and illustrates it with a hypothetical ecommerce case rather than with client data. He adds a caveat worth keeping: "you do need to run all these things through a common sense filter" (21:25).
There is one more move that translates cleanly. Looking at the demographic breakdown of his own campaign, he says "I wouldn't necessarily use this data and then go and change our targeting settings but what I might do is change the messaging on future ads" (25:19). Applied to a salon, that means if the bookings skew toward one kind of client, doodle owners, senior dogs, first trim puppies, rewrite the ad copy before you narrow the audience.
Which tools help a grooming salon measure ads, and what does each one miss?
Each tool below covers part of the path from ad to appointment, and none covers all of it. The gap in this trade is always the same, the moment a person stops being a click and becomes a phone call. We did not run these against each other in a test.
| Tool | What it does | Where it touches the ad to appointment gap in a grooming salon | Requires advertising knowledge? | What it does not do | Who makes it, and what they sell |
|---|---|---|---|---|---|
| Meta Ads Manager | Reports delivery, spend and platform recorded conversions for Facebook and Instagram ads, and lets you build custom columns and custom metrics, both shown in the source video | Shows spend, link clicks against landing page views, and frequency against a small local audience | Yes. The choice of columns and their reading are left to you | Does not see a booking taken over the phone, and does not know a client came back weeks later | Meta, which sells the ad inventory the report describes |
| Google Ads | Runs and reports search, Maps and YouTube campaigns, and offers call reporting on ads | Reaches the person searching for a groomer nearby, and can record a call placed from the ad | Yes. The choice of columns and their reading are left to you | Does not connect that call to a second or third visit unless you record it yourself | Google, which sells the placements it reports on |
| ManyChat | Automates chat conversations on Instagram, Messenger and WhatsApp, and tags contacts | Puts a timestamp and a tag on message enquiries that never reach a booking form | Partly. Flow building sits closer to front desk work than to media buying | Does not calculate a cost per booking on its own | ManyChat, selling its own subscription |
| Mailchimp | Sends email campaigns to a contact list and reports opens and clicks per campaign | Carries the rebooking reminder, which is where repeat value becomes visible | Partly. Little to send, more to read the reports | Does not measure paid social or paid search performance | Intuit, selling its own subscription |
| 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 | Sits at campaign creation and launch, which is the step before measurement. It does not close the ad to appointment gap | No, per the vendor's own description | Does not replace your appointment book and does not know which caller booked | SaleADS.ai, the company that publishes this site. Treat this row as an advertisement and check it yourself |
Disclosure. SaleADS.ai is the product of the company that operates this site, which is a conflict of interest and is why it is flagged in its own row. It is listed here as one option among several and it is not recommended. We did not run a controlled comparison, so the table carries no prices, no scores, no stars and no ranking. Each entry describes that tool's core published function, and the order is not a preference.
Where does this information come from, and how far should you trust it?
The source is one video, "How To Analyse Facebook Ads The RIGHT Way" by the channel Ben Heath, 29 minutes, at youtube.com/watch?v=mycqb92wJqk. Everything in quotation marks above is copied from its automatic transcript and linked to the second. Everything about grooming is this site's own reading.
What we took from the video: the point that a return on ad spend figure inside the platform may count only the first transaction, his position that cost per conversion or return on ad spend should carry most optimisation decisions, the phrase about tracking as far down the funnel as you can accurately track, the definition of frequency as impressions divided by reach, his stated preference not to see frequency above 2.5 on cold audiences, the common sense filter, and his preference for changing messaging rather than targeting when a demographic skew appears.
What is this site's own reading, and is not measured: the front desk source log, the two clock reading window built on your own rebooking gap, the shift from cost per booking to empty slot analysis once capacity is full, the reading of view rate and clicks as pointers to the offer or the page, and every application of any of this to pet care.
Honest warnings about the source:
- The video says nothing about pet care. We searched the full transcript for pet, pets, dog, dogs, cat, cats, groom, grooming, groomer, salon, vet, veterinary, boarding, daycare, kennel, walker, sitter, animal and puppy. Every one of those terms returns zero matches.
- It is commercial content. The example account is the author's own retargeting campaign for his own product, described in the video as "my new course with community" (11:54), with a link in the description. Ecommerce and lead generation appear only as hypothetical examples, not as data. He also opens with a self reported sales figure for his clients that we cannot verify and did not repeat.
- It is one practitioner's method shown on his own account, with no control group and no published dataset. The thresholds he gives are his working rules, not findings.
- The transcript is machine generated and error prone. It renders ROAS variously as "row ass", "rows" and "Turnal ad spend", writes "cold audiences" as "cold audio is", and writes "type" as "typee". The quotes above preserve those errors because that is what the file says.
- Interfaces change and break. During the demo a column that had been reporting when he prepared the video is not reporting on camera, and he says "ad manager is buggy" (2:21). Menus and column names you see today may not match the ones in the video.