Is Performance Max Good For Lead Generation?
- Jesse Heslinga
- 20 hours ago
- 8 min read
Updated: 5 hours ago
Performance Max is the campaign type Google has pushed hard the last 5 years. Still, there is a lot of disagreement about the use of Pmax for lead-gen businesses. As a frequent Reddit reader of /PPC, I see topics about this every other week. Some people swear by it, and plenty of others turned it on and swore never again. Both experiences are real, which is exactly why the question keeps getting asked.
This is because there is no one size fits all answer. Performance Max can be great for lead generation, but out of the box it's built to optimise for volume, not for customers. Whether it becomes your best campaign or your worst comes down to one thing (what you teach it a "good lead" is). This post covers what PMax is genuinely good at, where it goes wrong for lead gen, when to use it instead of plain Search, and how to run it without using up all your budget.
The Short Answer
Performance Max can work well for lead gen, but only if you feed it clean signal about which leads actually become customers. Left on the default settings, it optimises for cheap form fills and floods you with spam leads.
It's strongest once your Search campaigns are already working and maxed out (impression share roughly 70%+) and you have real conversion volume (50+ a month) for its AI to learn from.
For most lead-gen accounts the right split is Search first (70 to 80% of budget), PMax second (20 to 30%) for incremental reach, not the other way round.
The single biggest lever for PMax lead quality is offline conversion tracking: import your closed deals so Google optimises toward buyers, not form-fillers.
The honest answer: it depends on one thing
Performance Max optimises toward the goal you give it. If that goal is "form submitted" or "call started," Google goes and finds the people cheapest to do that action, which is not the same as finding people who become customers. I like to describe it as the "feedback loop of doom": it fills your pipeline with unqualified leads, then doubles down on whatever produced them.
So the real question isn't "is PMax good or bad." You have to look deeper, and tell it what a good lead actually looks like. Feed it clean data about which leads closed, and PMax can genuinely find buyers you'd miss with Search alone. Feed it nothing but raw form fills, and it will confidently spend your budget on junk.
Verdict: PMax isn't good or bad for lead gen on its own. It's an amplifier. It scales whatever signal you give it, so the quality of your setup decides the quality of your leads.
Keep in Mind: this is the same trap behind most junk-lead problems. If you count every form fill as a win, you're training Google to find more of the cheapest, worst leads, whether you're running PMax or not.
What Performance Max is actually good at
Give it credit where it's due. PMax does things a Search campaign can't:
One campaign across every Google surface. Search, Display, YouTube, Gmail, Discover and Maps, all from a single campaign, without building and managing each separately.
Finding incremental demand. Once Search is capturing the people actively looking for you, PMax can find buyers across YouTube and Display who weren't searching yet. That's reach you can't get from Search alone.
Real-time optimisation at scale. With enough conversion data, its AI adjusts bids and placements continuously toward what's working.
Verdict: PMax shines as a reach multiplier on top of a working account, not as the first thing you switch on in a brand-new one.
Where it goes wrong for lead gen
The failure mode is almost always the same, and it's why PMax has a bad reputation with lead-gen owners:
It counts the wrong conversion. Optimising to raw form fills is the root cause. (See the answer above.)
Display and YouTube pull lower-intent traffic. Those placements convert people who weren't really looking to buy, so lead quality drops versus pure Search.
Final URL expansion is on by default. Google can send clicks to any page on your site it thinks fits, not just your landing page.
You can’t steer where spend goes. Without discipline, a single PMax campaign is a black box that can quietly drift budget toward junk placements.
Verdict: none of these are reasons to avoid PMax. They're reasons to never run it on defaults for lead gen. Every one has a fix, covered below.
“For a client in the art auction space, PMax was generating plenty of leads but about half of them were junk. In this case we didn't rely on offline conversion, but we looked at the pages the PMax runs traffic to. It was clear it was steering to a page with ‘Free Taxation’, which brought in the junk leads. We fed the PMax a page feed, and excluded that page, so the campaign was optimising for the higher quality paid consultations page.”
Performance Max vs Search: which for lead gen?
Most lead-gen accounts should earn their way into PMax, not start there. A rough guide:
Situation | Better choice | Why |
Brand-new account, no conversion history | Search | PMax has no signal to learn from; you'll pay to train it |
Search works but is capped (impression share ~70%+) | Add PMax | You've maxed the people searching; PMax finds incremental demand |
Tight budget (under ~€1.5k to €2k/mo) | Search | Not enough volume to feed PMax's AI cleanly |
Strong offline conversion tracking + 50+ conv/mo | PMax works well | Google can optimise toward real buyers, not form fills |
You need tight control over exact queries | Search | Search + negatives gives precision PMax can't match |
A common healthy split for lead gen: 70 to 80% of budget in Search, 20 to 30% in PMax for incremental reach and testing. Search captures the high-intent people actively looking; PMax extends beyond them.
Verdict: if your Search campaigns aren't dialled in yet, fix those first. PMax multiplies a working account; it won't rescue a broken one.
Pro Tip: don't judge PMax against Search on cost per lead. Judge it on cost per qualified lead or per booked job. PMax often looks cheaper on raw leads and worse on real ones, which is exactly the trap.
How to run PMax without the junk
If you're going to run it, run it properly. In order of impact:
1. Feed it offline conversions. Import your closed deals (or qualified leads) from your CRM so Google learns which clicks became customers, not just which became form fills. This is the single biggest lever for lead quality, full stop.
2. Build a conversion hierarchy and bid on value. Separate conversion actions for each stage (form fill → qualified lead → booked/closed), then use value-based bidding so Google de-emphasises the low-value stuff automatically. A demo request and a newsletter signup should not be worth the same to the algorithm.
3. Turn off Final URL expansion. For lead gen, send clicks to the page you chose, not wherever Google guesses.
4. Structure asset groups by theme. Group them by the actual services you offer, not one generic catch-all. Use search themes to steer intent.
5. Use campaign-level negative keywords. You can now add up to 1,000 negatives at the campaign level to a PMax campaign. Use them: block job-seeker, DIY, educational and competitor terms, the same themed lists you'd use on Search.
6. Exclude bad placements and audiences. Run a placement report regularly and exclude the junk sites and YouTube channels eating spend. Add negative audiences (existing customers, wrong geos, job seekers).
7. Feed real audience signals. Give it your customer lists and qualified-lead lists as signals, not just "website visitors." Better seed, better direction.
8. Exclude your brand name. Searches on your brand should always be handled by a dedicated brand campaign, so you can steer spend on your own name. Exclude them from PMax.
Common mistake: turning PMax on with no offline conversion data, no exclusions and Final URL expansion left on, then blaming the campaign type when the leads are junk. The defaults are built for e-commerce volume, not lead-gen quality.
Common mistakes
Running PMax before Search works. It amplifies; it doesn't fix. Get Search converting first.
Optimising to raw form fills. Trains Google to find the cheapest, worst leads. Import outcomes instead.
Leaving it on defaults. Final URL expansion on, no exclusions, generic asset groups. That's the junk-lead recipe.
Judging it on cost per lead, not per qualified lead. The metric that looks good is often the one hiding the problem.
How I handle this for lead-gen businesses
I almost never start a lead-gen account on Performance Max. I get Search working first, high-intent keywords, clean tracking, tight negatives, because PMax only multiplies what's already there.
Traditional marketing teaches marketing in awareness stages. You start from the lowest awareness all the way to the bottom of the funnel, when potential customers are ready to buy. In Google Ads we want to work differently. We start from the bottom up.
Once Search is converting and capped out, I'll bring PMax in for incremental reach, but never on defaults. Offline conversion tracking goes in first so Google is optimising toward booked jobs, not form fills, then Final URL expansion off, themed asset groups, campaign negatives, placement exclusions, and real customer lists as signals. Run that way, PMax can genuinely become the best campaign in the account. Run on defaults, it's the fastest way to a pile of spam leads.
Ready to get more real leads out of Google Ads? If PMax is spending your budget but the leads aren't turning into work, it's usually the setup, not the channel. Worth a proper look.
Frequently Asked Questions
Is Performance Max good for lead generation
It can be, but not on defaults. PMax optimises toward the goal you set, so if that's raw form fills alone, it finds cheap form-fillers, not customers. Feed it offline conversion data about which leads actually closed, run it alongside a working Search campaign, and it can perform very well. Left unstructured, it's a common source of spam leads.
Why does Performance Max generate so many spam leads?
Because by default it optimises for the cheapest way to hit your conversion goal, and cheap leads are usually the worst ones. Its Display and YouTube placements also pull lower-intent traffic. The fix is importing closed-deal data so Google optimises toward buyers, turning off Final URL expansion, and adding negative keywords and placement exclusions.
Should I use Performance Max or Search campaigns for lead gen?
Start with Search. It gives you control over exact queries and negatives, which is what lead gen needs. Add Performance Max once Search is converting and capped (impression share around 70% or higher) and you have real conversion volume. A common split is 70 to 80% budget in Search, 20 to 30% in PMax for incremental reach.
How much conversion volume does Performance Max need?
Roughly 50 or more conversions a month gives its AI enough signal to optimise well, and you generally want at least 30 conversions in 30 days before trusting a target CPA. Below that, PMax doesn't have enough data to learn from, and you'll spend budget training it rather than getting efficient leads.
Can I add negative keywords to Performance Max?
Yes. Since early 2025 you can add up to 1,000 negative keywords at the campaign level in Performance Max. Use them like you would on Search: block job-seeker, DIY, educational and competitor terms. It's one of the most useful controls Google has added for keeping PMax lead quality up.
How do I stop Performance Max wasting budget on junk placements?
Turn off Final URL expansion, run a placement report regularly and exclude the low-quality sites and YouTube channels, add negative audiences (existing customers, wrong geos, job seekers), and structure asset groups by service theme instead of one generic group. Most importantly, import offline conversions so bidding chases quality, not volume.
Does Performance Max replace my Search campaigns?
No, and treating it that way is a common mistake. Search captures people actively searching for you; PMax finds incremental demand across YouTube, Display, Gmail and Maps. The strongest lead-gen accounts run both: Search as the core, PMax layered on top once Search is working and maxed out.
Comments