The Story
Quick test. Of those of you who ran Meta ads last year, did image ads or video ads book more jobs? Which angle brought back customers for a second job? Which hook produced the leads your office wished would stop calling?
If the honest answer is "no idea," you're in good company. Even the industry's best data can't answer it. LocaliQ's Facebook benchmarks (updated October 2025) show that home improvement advertisers running lead campaigns pay about $2.23 per click and $41.26 per lead, and that Facebook lead costs across industries rose 20% this year. What that report won't tell you, anywhere, is whether images or videos perform better. The biggest benchmark in local services doesn't even split them out.
The benchmark can't print that answer because it's different for every account, and the only place it exists is in data almost nobody keeps.
The stakes on creative are not small. The best evidence we have (NCSolutions' 2023 analysis of nearly 450 consumer-goods campaigns) puts 49% of a campaign's sales impact on the creative itself, more than targeting and reach combined. And Motion's 2026 study of $1.29 billion in Meta spend found only about 5% of creatives ever pull real budget. Most ads are duds; the game is knowing which few aren't, by your definition of "works," not Meta's. Meanwhile the platform keeps nudging you toward video, whether or not video books jobs for you specifically. Nobody can tell you. Except your own records, if they were set up to answer.
The fix starts somewhere almost embarrassingly cheap: the moment you type a name for the ad. Most owners type whatever's handy and move on. That throwaway decision is the difference between an ad account that can answer "which creative brings customers who spend the most?" and one that never will. More on that in a minute. First, the trap that makes it urgent.
The Tip of the Week
Cheap leads lie. Make every contact record testify.
Meta itself concedes that form-fills and qualified leads are different animals. It sells a setting called Conversion Leads, where your CRM reports back which leads actually qualified so Meta can go find more of those. By Meta's own numbers, that setup delivers on average 15% lower cost per quality lead than optimizing for raw form-fills. Which means the ad that wins on cost per lead is often not the ad that wins on leads worth having.
Now stretch that down the funnel. The ad that wins on qualified leads can lose on booked jobs. The ad that wins on booked jobs can lose on lifetime value, because the $500 trim customer and the $5K removal-plus-plant-health-care customer probably didn't click the same ad. No study exists on this for tree services. The answer lives in exactly one place: your CRM. And only if every contact record knows which ad produced it.
The requirement: pull the ad name into your CRM at the contact-record level. Not "source: Facebook." The actual ad. Meta can pass an ad's identity through to your website forms automatically, lead-form integrations can map the ad and campaign names into CRM fields, and call tracking can inherit the same tags for phone leads. Almost nobody wires this up. Owners who do see what their competitors can't.
Four warnings before you run at this:
Rename a live ad, and you split your data. The name your website route captures is frozen when the ad is published, but lead-form integrations pull the new name. One ad becomes two in your records. Name it right before launch, then leave it.
This only works with discrete ads. Advantage+ and dynamic formats auto-assemble many variants under one ad name, and the variant identity never reaches your CRM. For anything you want contact-level answers on, run deliberate, separately named ads.
"I just ask people where they heard of us" doesn't get you there. They say "Facebook." Nobody says "the crane video with the second hook." That's why the record has to carry the name. And some people see your ad, then Google you and call; no link ever tracks them, so accept a baseline floor.
At 30 leads a month, you will not get scientific answers per ad. If roughly a third of your leads book, proving a 15-point booking-rate gap takes about 160 leads per concept. So pool ads together by angle rather than judging single ads, read quarters rather than weeks, and treat this as revenue accounting: which angle brought the customers who booked, and what were they worth? Revenue gaps show up long before statistical proof does.
The name is where you decide, before launch, what you'll want to know later. Built right, the name carries the format, the angle, the hook, and the batch, in a form that can be split apart later, and every contact record inherits all of it. Built like most accounts ("Video 3 final FINAL"), it carries nothing. There's a real system to this, and with one, your reports answer questions Ads Manager can't. We're purposely not including the system here; badly copied naming schemes cause a good share of the messes we get called in to fix.
Action Steps (this week, no software required yet):
Pull your last 25 CRM contacts and count how many records could tell you which ad produced them. That number is your baseline. (20 min)
Open Ads Manager and read your ad names cold. Could a stranger tell the format, angle, and hook from the name alone? Score each yes or no. (15 min)
Write down the four things you'd want every future ad name to tell you about the lead it produced. Keep the list; it becomes your naming spec. (15 min)
Ask whoever runs your CRM one question: "If I put information in the ad name today, would it land on the contact record automatically?" The answer decides your next step. (15 min)
Stop renaming live ads. Starting today. (0 min, requires only restraint)
Tracking which ad brought the lead is half the game. What happens after it lands is the other half, and that's where this week's partner lives.
TOGETHER WITH LEXI
What a Few Tree Service Companies Are Doing Differently
Nobody builds a tree service company because they love chasing down quotes that never got a callback. Most owners are good at the work, the climbing, the removals, the crew.
Chasing leads and rebooking a canceled job isn't the job they signed up for. They're not bad at follow-up; they're just busy running a business.
Here's what a few owners are using instead: See How
Worth a Look
Meta shut down its old Offline Conversions API in May 2025. The current path is the Conversions API, where you can upload offline and CRM events (qualified, booked, revenue) up to 62 days after the conversion. Useful, but 62 days is shorter than your repeat-customer cycle. CAPI teaches Meta's machine; the CRM stamp teaches you.
Meta added a per-asset "creative breakdown" in Ads Manager in July 2025 for its flexible formats and AI-generated images. Helpful directionally, but it only reports; nothing from it lands on your contact records. Platform reporting and your own books are different tools.
If you run instant forms: those leads never touch your website, so they need the integration route for name capture, not the website route. Worth knowing before you wire anything.
If you run Local Services Ads, your phone habits are about to cost real money. Starting October 1, Google will bill a missed call during business hours as a valid lead once the caller stays on the line 20+ seconds, and follow-up calls can be billed too if they meet the lead criteria. One exception: if your phone tree requires a key press, the timer only starts after the press, and no press means no charge (per Search Engine Land). Whoever answers your phone just became a line item.
Bottom Line
Agencies love sending screenshots of metrics. You know what never appears? Which ad brought the customer who spent $14K over three years. We build this for clients the unglamorous way: the ad name lands on the contact record, and the booked-jobs report does the talking. That's how an owner learns something like "the testimonial angle books fewer jobs but brings our best plant-health-care customers." No screenshot of CTR has ever said anything that useful.
When your ad account spits out a $41 lead, do you know today whether that lead ever became a booked job?
If not, the gap isn't your ads. It's that your ads and your CRM have never been introduced. The introduction starts with a naming decision that costs nothing and pays for years.
Run the 25-contact test, then reply with your number out of 25. I'll tell you what I'd wire up first.
Whenever you're ready, here's how we can help:
Do it yourself. The AI-Proof Tree Service Playbook. Your Google profile, site, and reviews sorted in an afternoon. $47. Get the Playbook
Come on the show. We interview tree service owners on my podcast Stumped Again. Grab a spot here
Go after the commercial money. See what public and commercial contracts are open in your area. Reply "commercial", and we'll show you what’s available in your market.
Let us handle it. Rather have our team help? Let’s discuss your tree service marketing
It's Labor Day weekend. Whatever the backlog says, the crew has earned the Monday off, and so have you. The trees will still be there Tuesday.
See you next week!
- Jacob Hastings
How did you like today's issue?
The Backcut
