Stop obsessing over conversions, your budget is the real problem

pixelpusher

Running a mobile detailing business and blaming Google Ads for $40-50 CPA? Mate, that's cheap for a service that probably has a high lifetime value. You're flip-flopping between maximise clicks and maximise conversions every few days, then wondering why the algorithm can't learn. Classic mistake.

You had messed up conversion tracking, fixed it, and now you're scared to use it because of 'old data'? That data doesn't poison anything - just let it run. The real issue is your $20/day budget. In a bidding system, that's pocket change. You're competing against people willing to pay more per click, so Google shows your ads less often and when it does, it's to lower-intent users.

Google recommends $17 target CPA because it knows the market. If you can't stomach $40-50 per lead, maybe ads aren't for you. But if each customer brings in £200-300, that's a 5x return. Stop micromanaging the bid strategy and start scaling your budget to a level where the system has enough data to optimise. Or go back to flyering car parks - that's free.

marketingmule

Totally feel this. I've been guilty of doing exactly that-swapping creatives, tweaking audiences, refreshing everything way too fast. But Meta really does need time to exit that learning phase, especially for conversion objectives. Usually give it at least a week of solid data before I even think about touching anything. Have you been letting it run long enough to gather meaningful signals?

justauser

The learning phase can be brutal - even with Target CPA, Google basically ignores your bid for the first couple of weeks 😅

In my experience, most decent lead gen campaigns land somewhere between 30-50% lead-to-customer conversion, but that varies wildly by industry. If you're not comfortable burning some cash upfront, paid search might not be the best fit - it's a bidding war after all, and Google has plenty of tricks to nudge costs up.

I'd say set aside a proper test budget you're okay losing, let it run for a month, and then decide. You can't really optimise something that never got a fair shot.

funnelhacker

Honestly, the first move should be auditing your keyword-to-conversion funnel, not the creative or bid strategy. Most people are pissing budget on generic terms that just eat clicks but never close. Real leads come from intent signals, not impressions. I've seen far better ROI by scraping relevant discussions in hyper-local groups and targeting those users with programmatic retargeting. Something like ParseStream lets you map those conversations to actual buying behaviour so you're hitting prospects when they're already in-market, not just browsing. That's where the volume lives.

budgetburner

Google Ads runs on a two-week learning period. Realistically, you want to judge performance against a full month. And Google themselves push for at least fifteen to thirty conversions logged before you switch from click bidding to conversion bidding.

What kind of conversions are you actually bidding on? Makes a big difference. A phone call lead can cost a fraction of a form submission for certain business types - and you can set qualification signals in the backend, like minimum call duration, to filter garbage. Also, stick a value on each conversion. If you leave page views or local actions as primary, the algorithm will happily optimise for cheap, useless volume. Demote everything except your main goal to secondary, and the bidding engine has no choice but to focus on what actually matters.

From there it's down to landing page quality and ad copy that doesn't suck. Better CTRs kill your CPCs eventually.

Done this for over a decade. There's a lot more you can do, but that's the starting point that most people skip.

dailygrind

Absolutely. Switch back to maximise conversions now your tracking's fixed. The old bad data is noise-the algorithm needs something real to optimise against. I've seen CAC blow up if you leave it on the wrong objective too long. That first week of clean data will be worth more than the last month of junk.

wordweaver

Oh, absolutely test it - click campaigns just give you window shoppers, not people who actually turn up with cash in hand. Once you've got enough conversion data, let Meta do the heavy lifting and find the ones ready to pay. Just don't expect miracles overnight.

funnelhacker

The $40-50 CPA you're seeing on Max Conversions isn't necessarily wrong - it just hurts when margins are tight. The real question is whether those form fills convert into booked jobs at a rate that makes the CAC work. If a mobile detail averages $150+ per job, a $45 cost per lead can be fine if 60-70% of those leads book. But if only 20% book, you're looking at $225 per customer - that's a broken funnel.

Two things to hammer down immediately:

Search terms report. Bleeding budget on "mobile detailing cost" or any broad-match crap that brings in DIY researchers instead of buyers? negative that shit out. Only high-intent queries - "mobile detailing [city]", "mobile car wash near me" - should get a cent.
Conversion action. If it's counting button clicks or page views instead of actual form submissions, Max Conversions will optimise for garbage. you fixed that - good. but Google needs 15-30 conversions in a learning window to stabilise. With two conversions over a month at $20/day, you're still in the noise. honestly, in a local service vertical at that budget, I'd stick with Max Clicks, tighten your keyword list to pure high-intent terms, and get your CPC under $3-4. Once you're seeing 10+ conversions per month consistently, then flip to Max Conversions with a $25-30 target CPA and let it learn for two weeks before you touch it.

we ran into identical tracking headaches early on, which is why a colleague and I ended up building a tool to wire conversion tracking automatically - saves all the Tag Manager mess. But for now, focus on search terms and click volume. You can't optimise what you can't measure.

ctrjunkie

Yeah, the conversion tracking nightmare is so real for local service businesses. Mobile detailing especially - the journey from ad click to actual booking is messy. Phone calls, form fills that happen offline, or people just showing up and paying on the day. All that gets lost in the pixel ether.

Maximise Conversions is basically useless if the data feeding it is garbage. Phantom conversions from God knows where ruin everything. If it's counting random page views as leads, the algorithm will optimise for more of that noise.

£30-40 per actual customer isn't terrible for local services, but if you're paying for fake conversions then yeah, you're burning budget on thin air.

First thing I'd do: audit what's actually being counted. double-check the tag setup. Make sure you're not rewarding every bot visit or accidental click on a form field.

Honestly, switching back to Maximise Clicks with a strict CPC cap might be your safer bet. Also, use ad scheduling so you're only running when you can actually pick up the phone or respond fast.

Quick one - are you tracking phone calls as conversions too? Or just the form fills? That makes a huge difference.

How's your average job value looking? Trying to figure if £35-40 per lead works for your margins long-term.

pixelpusher

You're obsessing over the learning phase as if it actually guarantees anything. Nine impressions and no clicks? that's not a learning problem, that's a signal problem. Conversions objective doesn't magically fix a tiny audience or a weak offer. I'd bet your targeting is too narrow or your creative is blending into the noise. Everyone blames the platform, but half the time it's the account structure that's f*cked. If you haven't changed anything in a week, the algorithm isn't "learning" - it's starving.

budgetburner

Worth a shot? only if you're serious about not wasting budget. The real move is feeding your closed-won data back into Google Ads. Otherwise you're just letting the algorithm bid on random inbound leads that'll never convert.

Set up enhanced conversions for leads at the basic level. or, if you can capture the GLCID from the inbound lead and track it all the way through to the sale, upload that as an offline conversion manually. That'll shift your bidding toward lower-funnel, higher-quality signals instead of chasing vanity numbers.

most people skip this, then wonder why their lead gen costs keep climbing.

pixelpusher

Honestly, I'd push back on that. Two leads in 14 days is noise - not a signal worth dissecting. You're better off auditing your landing page and CTA than poring over form submissions from people who didn't book. But if you're set on it, go ahead, just don't expect a silver bullet.