Should I Use Broad Match with Smart Bidding in Google Ads in 2026?

Jesse Heslinga
13 hours ago
11 min read

Google will tell you to use broad match with Smart Bidding. It's in the recommendations tab. It comes up in conversations with account managers. Their documentation calls it the "recommended setup" for 2026. And their own internal data backs it: campaigns running broad match with Smart Bidding show 67% more traffic and 34% lower costs in their benchmarks.

I run Google Ads for service businesses and I've tested this in a range of accounts. The honest answer is: it depends on your account's current state, and "67% more traffic at 34% lower cost" is not the same as "67% more customers at 34% lower CPL."

I run paid search, so I have a view from inside this. I've tried to give you the fair take below.

The Short Answer

Broad match + Smart Bidding works well when your account has clean conversion data, sufficient monthly volume (20+ conversions/month), and a strong quality signal (not just form fills)
Without those conditions, broad match tends to find cheap, low-quality traffic and Smart Bidding optimises for more of it
Phrase match is the most consistent performer for lead-gen accounts that are still building conversion history
The 2026 context: AI Max adds another layer of reach expansion on top of broad match, which makes conversion signal quality even more important as the foundation
The right setup depends on where your account is, not what Google recommends universally

What Broad Match and Smart Bidding Are Actually Doing

These are two separate features that Google recommends running together because each makes the other more useful.

Broad match means your keyword "emergency plumber" can show for searches like "plumber available now," "urgent pipe repair," "water leak help," and potentially "plumbing company near me." Google's algorithm decides which searches are close enough to match. In 2026, broad match matching is significantly wider than it was three years ago. The algorithm uses the entire context of your account (landing pages, other keywords, ad copy) to decide what to match.

Smart Bidding (Target CPA, Maximize Conversions, Target ROAS) uses machine learning to set bids in real time, for each individual auction. It considers hundreds of signals: device, time of day, location, search history, browser, and more. The goal is to bid higher when a click is more likely to convert and lower when it isn't.

Together, the idea is: broad match finds more relevant searches, and Smart Bidding decides which of those searches are worth bidding on. The algorithm does what a manual keyword list and manual bidding can't: it operates at scale, in real time, across a wider search surface.

The limitation is that Smart Bidding can only optimise for whatever you've defined as a conversion. If your conversion is a form fill, the system finds form fills. If your conversion is a booked job or a qualified lead imported from your CRM, the system finds those. The input to the algorithm is what determines the quality of the output.

Verdict: the combination can work, but "broad match + Smart Bidding" is a tool, and what you put into it determines what you get out.

Keep in Mind: When Google says broad match + Smart Bidding reduced CPA by X%, they're measuring cost per conversion, where a conversion is whatever the account defines. If the account defined "form fill" as a conversion, CPA went down on cheap form fills, not necessarily on customers.

What Google Says, and What the Context Is

Google's internal data shows that campaigns running broad match alongside Smart Bidding strategies see, on average, a 67% increase in traffic and a 34% reduction in costs compared to campaigns using the same Smart Bidding strategy without broad match.

These numbers come from Google's own analysis of Google Ads accounts. They measure cost per conversion (where conversion = whatever the account has set). They measure traffic volume. They don't measure lead quality, close rate, or revenue.

Independent analysis tells a different story for accounts that look at quality metrics:

Optmyzr's research across a large sample of accounts found that phrase match is the "consistent overperformer" for lead-gen, punching above its weight in conversion quality
Accounts that shifted 60% of their budget from broad match to exact and phrase match for high-intent keywords saw a 47% drop in cost per conversion, a 35% improvement in conversion rate, and 28% more qualified leads over three months (per analysis from Bigeye and ATTN Agency, 2025)

Neither set of data is wrong. They're measuring different things. Google is measuring volume and cost per conversion in accounts where a conversion is often a form fill. The independent data is measuring what happens when you tighten the match types and look at quality metrics.

Verdict: Google's data is real, but it's measuring the output of whatever you define as a conversion. If your conversion signal is strong, their data applies. If your conversion is a basic form fill, the efficiency gain is probably in form fill volume, not in customers.

When Broad Match + Smart Bidding Works for Lead Gen

There are conditions under which this combination performs well, and accounts that meet them should test it seriously.

You have 20+ conversions per month. Smart Bidding needs statistical volume to learn from. Below 20 conversions, the algorithm is operating on too little data and makes erratic decisions. Above 20, it has enough signal to generalise and start making reliable predictions.

Your conversion signal is high-quality. If you're sending offline conversion data back to Google (qualified leads or booked jobs from your CRM), Smart Bidding has a real-world signal to optimise toward. Broad match can then expand into new searches and the algorithm has the context to evaluate which of those searches match your customer profile. For how to set up this tracking, see how to set up offline conversion tracking in Google Ads.

Your account has history. A campaign that's been running for 6-12 months has learning data. The algorithm knows which types of searches have converted and which haven't. Broad match in an account with history performs differently from broad match in a brand-new account.

Your CPL budget allows for some learning inefficiency. The first 4-6 weeks of a broad match test can be messy as the algorithm explores the expanded search territory. Accounts with tight budget constraints may find the exploration phase expensive.

Verdict: if you meet all four conditions, running a controlled test of broad match with Smart Bidding is worth it. Set up a separate campaign, give it 8-12 weeks, and compare CPL and lead quality to your existing setup.

Pro Tip: Don't switch an entire existing campaign to broad match. Create a new campaign alongside your existing one, run both, and compare after 60-90 days. If the broad match campaign wins on quality metrics (not just volume), you can shift more budget to it. Switching everything at once throws away your baseline and makes the comparison impossible.

When Broad Match Doesn't Work, and Why

The failure mode is consistent across accounts I've seen it happen in. You switch to broad match, impressions go up sharply, clicks go up, conversions go up. CPL drops. You think it's working. Then your sales team starts complaining about lead quality, or you check your offline data and see that the close rate on leads from the campaign has dropped.

This happens when:

Conversion volume is too low. Without enough data, Smart Bidding reaches into broad match territory without reliable signals. It finds cheap clicks because cheap clicks show up more often, and the algorithm interprets "shown up often" as "likely to convert."
Your only conversion is a form fill. Broad match finds the widest possible audience. Many of them will fill in a form. A form fill is easy. A qualified lead who books and pays is harder. If the algorithm doesn't know the difference, it optimises for the easy thing.
The account is new. A new campaign has no history for the algorithm to draw from. Broad match in a new campaign is essentially giving the algorithm an empty slate and asking it to figure out your ideal customer from scratch. Phrase and exact match provide guardrails while that history builds.
The industry has significant irrelevant traffic. Some verticals attract a lot of non-buyer searches: research intent, students, job-seekers, DIY. Broad match in these verticals without extensive negative keywords can pull in substantial irrelevant traffic before the algorithm learns to filter it.

The pattern I see in accounts with the cost per lead rising problem is often a combination of broad match and weak conversion data that has allowed the algorithm to optimise for something other than what the business actually needs.

Verdict: if your conversion data isn't clean and your volume isn't sufficient, broad match amplifies bad signals rather than finding better ones.

The Match Type Setup That Performs Most Consistently

For lead-gen accounts that aren't yet meeting the conditions for broad match success, the most reliable starting structure is phrase match as the core match type.

Phrase match means your keyword "emergency plumber" shows for searches that contain that phrase or close variations of it, in that order, with other words allowed before or after. It's more targeted than broad match but wider than exact match.

Why phrase match works well as a default:

It provides more control over what searches trigger your ads compared to broad match
It still allows Smart Bidding to work across relevant variations
It generates a cleaner Search Terms Report, which makes negative keyword management easier
It tends to produce stronger conversion rates in accounts without large conversion history

A sensible match type structure for most lead-gen accounts:

Core high-intent keywords on phrase match. Your main service + location, urgent-intent modifiers, high-converting queries you've identified from the Search Terms Report.
Exact match for your proven top performers. If "emergency plumber [city]" has a strong track record in your account, lock it in as exact match to protect that traffic.
Broad match for expansion testing. Add broad match keywords for terms where you want to discover new search territory, but keep these in a separate campaign or ad group so you can track their performance independently.

As your conversion volume grows and your offline tracking improves, you can shift more budget toward the broad match expansion campaign if it's performing.

Broad Match in 2026: What AI Max Changes

In 2026, Google introduced AI Max, an optional layer you can add to Search campaigns. It combines broad match keyword matching with AI-generated ad copy variations, automatic URL expansion (sending users to the most relevant page on your site rather than your chosen URL), and audience signal expansion.

AI Max essentially takes broad match further. It's Google's most expansive Search campaign setup.

The same conditions that govern broad match performance govern AI Max performance, but more so. Clean conversion data, strong volume, and a reliable quality signal matter even more when the system has this much reach.

For most lead-gen accounts, I'd treat AI Max the same way I'd treat broad match: test it in a controlled campaign with strong conversion tracking in place, not as a default setting to turn on account-wide. The Performance Max for lead gen considerations are a useful reference here; many of the same caveats about automation and signal quality apply.

Verdict: AI Max is broad match with the dial turned further. The conditions for it working are the same, just with higher stakes.

Common Mistake: Turning on AI Max or broad match across all campaigns simultaneously, then trying to diagnose performance issues. Run expansive match types in isolated campaigns first. If something goes wrong, you can turn it off without disrupting your whole account.

The Conditions That Determine the Answer

Condition	Broad match + Smart Bidding	Phrase match + Smart Bidding
New account, little history	Risky: algorithm has nothing to learn from	Better starting point
Under 20 conversions/month	Algorithm learns too slowly	More stable
Conversion = form fill only	Likely to find cheap form fills, not customers	Same risk, smaller scale
Offline conversion data in place	Effective: algorithm has real-world signals	Also works, less exploration
20+ qualified conversions/month	Works well: sufficient learning signal	Works well, less reach
High irrelevant traffic vertical	Needs extensive negatives first	Easier to manage
Account with 6+ months of history	Strong candidate for broad match test	Solid fallback

How I Handle This for Lead-Gen Businesses

My default starting setup for a new client is phrase match with Smart Bidding (Target CPA or Maximize Conversions with a target). I get the conversion tracking right first: both online form fills and offline qualified-lead imports from the CRM. Once the account has 60-90 days of clean data and we're at 20+ conversions per month, I'll test a broad match expansion campaign alongside the existing setup.

If the broad match campaign produces qualified leads at a comparable or better CPL after 8-12 weeks, I shift budget toward it. If it produces volume but lower quality, I close it and stay with phrase match.

The mistake I see most often is clients coming to me after their previous setup added broad match to everything because Google recommended it, without fixing the conversion tracking first. The campaigns look efficient on paper (low CPL) but the leads aren't converting into customers. The algorithm was doing exactly what it was told: optimising for form fills, not customers.

Not sure if your match type setup is right for your account? If you've been running Google Ads and something feels off, a second set of eyes is worth ten minutes. I'll look at your match types, your conversion tracking, and where the budget is actually going, and tell you straight whether something needs changing or you're in good shape.

Book a free account review

Frequently Asked Questions

Is broad match better or worse than phrase match for lead gen?

It depends on your account's conversion data and volume. Phrase match is more consistent for accounts that are still building history or that have below 20 conversions per month. Broad match can outperform phrase match when Smart Bidding has strong, quality-rich conversion signals to work from. The right answer for your account comes from testing both with clean tracking, not from a general recommendation.

How many conversions do I need before broad match with Smart Bidding works?

Google recommends at least 50 conversions per month for Target CPA to work optimally, but most accounts see stable learning from around 20. Below that, Smart Bidding's decision-making is too erratic for broad match's expanded reach to be reliable. Start with phrase match and build toward that threshold before testing broad.

Can I use broad match without Smart Bidding?

You can, but it's not recommended. Without Smart Bidding, you have no real-time bid adjustment to filter the broad match traffic by conversion likelihood. You'd be bidding the same amount on every search that matches your keyword regardless of whether it's likely to convert. Manual bidding with broad match tends to produce poor results for most accounts.

Google keeps recommending I switch to broad match. Should I follow the recommendation?

Treat Google's recommendations as suggestions, not instructions. They're generated automatically and don't account for your specific account conditions, conversion quality, or sales cycle. Review the recommendation in context: if your account meets the conditions (volume, conversion quality, history), it may be worth testing. If it doesn't, wait until it does.

What is AI Max and how does it relate to broad match?

AI Max is an optional feature for Search campaigns in 2026 that combines broad match keywords with AI-generated ad copy variations, automatic URL expansion, and expanded audience signals. It's a more aggressive version of the broad match + Smart Bidding combination. The same conditions apply: clean conversion data and sufficient volume are prerequisites for it to work well.

Should I use broad match for brand keywords?

Brand campaigns (bidding on your own business name) are a special case. The search volume is inherently more targeted because people are already looking for you. Exact match or phrase match for brand terms is almost always the right call. You don't need broad match to expand reach on searches for your own name.

If broad match is finding bad traffic, will adding more negative keywords fix it?

Partially. Negative keywords filter out the clearly irrelevant categories (job-seekers, DIY researchers, students) and should always be in place regardless of match type. But negative keywords can't fix a fundamental signal quality problem. If Smart Bidding doesn't know what a good conversion looks like because all conversions are form fills, negative keywords reduce the noise but don't tell the algorithm what to optimise toward. Fix the conversion signal first.

About the author

Jesse Heslinga | Google Partner | 7+ Years Google Ads | Lead-Gen Expert

I run Google Ads for lead-gen service businesses at Groove Media across clinics, home services, and professional services, working with clients directly, no account-manager layer. I build every account around one question: is this spend turning into real customers, not just cheap form fills?