There’s no fixed number that works for every account, but as a practical starting point, most well-structured ad groups perform best with somewhere between 5 and 20 closely related keywords. The right number depends far more on how tightly those keywords relate to each other and to your ad copy than on hitting any specific count.
This guide covers why keyword count alone is the wrong question to focus on, what actually determines whether you have too many or too few, and how to structure your account so keyword volume stops being a guessing game, whether you’re starting from scratch or cleaning up an account that’s already grown out of control.
Why “How Many Keywords” Is the Wrong First Question
Google Ads doesn’t reward accounts for having more keywords. It rewards accounts where each ad group’s keywords are tightly themed around a single concept, because that’s what makes it possible to write ad copy and landing pages that feel directly relevant to the search. A campaign with 300 loosely related keywords crammed into a few ad groups will almost always underperform one with 60 keywords split carefully across tightly themed groups.
The real question isn’t “how many keywords should I have,” it’s “how many distinct concepts does my product or service actually cover, and how many keywords does each of those concepts genuinely need.” Once you think about it that way, the right number becomes a natural output of your account structure rather than an arbitrary target.
A Worked Example
Say you sell shoes. A single “Shoes” ad group with 80 keywords covering running shoes, hiking boots, and dress shoes will struggle, no single ad or landing page can feel relevant to all three. Split that into three ad groups instead: “Running Shoes” (8-12 keywords covering variants like “running shoes,” “best running shoes,” “running shoes for men”), “Hiking Boots” (similar range), and “Dress Shoes” (similar range). Same total keyword count, roughly, but each group now supports ad copy and a landing page that actually matches what the searcher typed.
Starting From Zero: What to Do Before You Have Any Data
Most of the guidance in this article, pruning thresholds, timelines, negative keyword strategy, assumes you have some conversion history to work from. If you’re launching a brand-new account, you don’t have that yet, so the approach needs to shift slightly for the first few weeks.
Start narrower than you think you need to. Build your ad groups around your highest-confidence concepts first, the products or services you’re most certain people search for by name, rather than trying to cover every possible angle on day one. A smaller, well-themed starting list reaches meaningful data faster than a large one, simply because your budget isn’t split as many ways.
For negative keywords, don’t wait for search terms data to start. Add obvious exclusions upfront, based on what you already know doesn’t apply to your business (terms like “free,” “jobs,” “DIY,” or competitor names, if they’re not relevant to your offer). This isn’t a substitute for the reactive, search-terms-driven negative list you’ll build over time, it just prevents avoidable waste in the early weeks before that data exists.
Once you have roughly 15-20 clicks per keyword, you have enough of a baseline to start making real keep-or-cut decisions using the pruning threshold described later in this guide. Before that point, treat every keyword as still in its trial period rather than judging it.
What Determines the Right Number for Your Account
A few factors shape this more than any general rule of thumb does.
Match Type Changes the Math
Exact match keywords need fewer variants than they used to, since Google’s exact match now accounts for close variants, word order, and implied meaning automatically. This means accounts that once needed 30 exact match variations of a single phrase to cover every wording now often need far fewer, because the match type itself absorbs that variation.
Phrase match needs a bit more room for variation, since it’s less forgiving of intent-based matching than exact match, but still doesn’t require the sprawling keyword lists that were standard practice several years ago.
Ad Group Theme Tightness Matters More Than Count
An ad group should generally represent one specific idea, one product, one service variant, one clear searcher intent, not a broad category with dozens of loosely connected terms stuffed inside it. If you’re finding it hard to write a single ad headline that feels relevant to every keyword in a group, that’s usually a sign the group needs to be split, regardless of how many keywords are currently in it.
Budget and Search Volume Set a Practical Ceiling
An account with a small daily budget doesn’t benefit from hundreds of keywords, since most of them will never accumulate enough impressions or clicks to generate useful data. In these cases, a smaller, tightly focused keyword list actually collects meaningful data faster than a sprawling one competing for the same limited budget.
Are Single Keyword Ad Groups (SKAGs) Still Worth Using?
This comes up constantly in PPC discussions on this exact topic, so it’s worth addressing directly. SKAGs, building one ad group per individual keyword instead of grouping several related terms together, were a popular strategy for years, largely because older exact match was rigid and literal, so isolating each keyword gave advertisers tighter control over ad relevance.
That rationale has weakened significantly now that exact match automatically covers close variants and implied intent. A single exact match keyword today often captures what used to require three or four separate SKAGs, which means the manual overhead of building and maintaining hundreds of single-keyword ad groups usually isn’t worth it anymore. SKAGs still make sense in narrow cases, a handful of your highest-value, highest-volume terms where you want granular control over bids and copy, but building an entire account this way is generally more maintenance than it’s worth given how match types behave now.
Does Performance Max Change This Entirely?
For Performance Max campaigns, yes, this changes fundamentally. Performance Max doesn’t use a traditional keyword list at all. Instead, it relies on asset groups built around themes, audience signals, and creative assets, and Google’s system automatically determines which searches, placements, and audiences to target based on your stated goals.
This means the “how many keywords” question simply doesn’t apply the same way to Performance Max. If a meaningful share of your budget runs through PMax, your levers are your asset group themes, your audience signals, and your creative quality, not a keyword count. For accounts running a mix of Search and Performance Max, treat the keyword guidance in this article as specific to your Search campaigns, and manage Performance Max as a genuinely separate discipline.
Signs You Have Too Many Keywords
A handful of patterns tend to show up in accounts that have genuinely overexpanded their keyword lists:
- A large share of keywords showing zero or near-zero impressions over a meaningful stretch of time, meaning they’re not actually competing in any real auctions
- Search terms report entries that consistently don’t match the intent of the keyword that triggered them, a sign of overly broad or poorly grouped keywords
- Difficulty writing ad copy that feels genuinely relevant to every keyword in a given ad group
- Budget spreading so thin across keywords that none of them accumulate enough clicks or conversions to draw reliable conclusions
What Too Many Keywords Actually Costs You
Beyond wasted management time, an overloaded keyword list dilutes your data. Google’s systems and your own reporting both work better with concentrated signal, a smaller list of keywords that consistently gets real traffic gives you clearer performance data than a sprawling list where most terms barely register. It also tends to lower relevance at the ad group level, since it’s harder to keep ad copy tightly matched to every term when the list has grown too broad.
When to Actually Prune a Keyword
“Significant spend without conversions” is vague on its own, so here’s a more concrete way to think about it, once your account has enough history to calculate a typical cost-per-conversion. A reasonable checkpoint is once a keyword has accumulated clicks equal to roughly 3 to 5 times your typical cost-per-conversion without producing a single conversion, that’s a fair point to pause it and reassess, rather than acting on the first few unconverted clicks, which can easily just be normal variance. For newer keywords with limited data, waiting for at least 15-20 clicks before making any keep-or-cut decision avoids overreacting to a small, unrepresentative sample. This threshold assumes an account with some established conversion history behind it, a brand-new account won’t have a reliable cost-per-conversion figure to measure against yet, which is why the earlier “trial period” approach matters for the first few weeks.
Signs You Have Too Few Keywords
The opposite problem is less common but still worth checking for, particularly in newer accounts or narrow niches.
- Consistently low impression volume across the account despite a reasonable budget, suggesting the keyword list isn’t capturing enough of the relevant search demand
- A search terms report showing frequent, relevant searches that aren’t matched by any keyword currently in the account
- Ad groups built around only one or two keywords when the underlying concept clearly has more natural variations worth targeting
It’s worth distinguishing this from a single keyword that individually gets little to no traffic while the rest of the account performs normally, that’s usually not a “too few keywords” problem, it’s more often a sign that specific term simply has low search volume or that your bid isn’t competitive enough to win those auctions, which is a bid or budget issue rather than a structural one.
A Practical Starting Framework
Rather than picking an arbitrary number, this approach tends to produce more reliable results:
- Map out the distinct products, services, or intents your business actually needs to cover in search.
- Create one ad group per distinct concept, not per keyword, so the group stays tightly themed.
- Populate each ad group with 5 to 20 keywords that are genuine variations of that same concept, using primarily exact and phrase match.
- Let the campaign run until you have enough data, generally a few weeks of consistent impressions, before judging whether any given keyword deserves to stay.
- Review the search terms report regularly and add genuinely relevant new terms as negative keyword and expansion decisions become clearer, rather than front-loading the account with keywords you’re guessing might work.
This keeps the keyword count as a byproduct of good structure, not a number you’re trying to hit or avoid. Once your structure is in place, the next lever worth understanding is how negative keywords interact with this same count.
Negative Keywords Do Half the Work
A well-managed negative keyword list is often what allows a smaller positive keyword list to perform well in the first place. Instead of adding more keywords to try to capture every possible search variation, a tighter positive list combined with a solid negative keyword list, built from your search terms report over time, usually produces cleaner data and better relevance than simply expanding the keyword count. Think of negative keywords as the filter that lets a smaller, more concentrated keyword list do its job properly, rather than a separate maintenance task disconnected from the keyword-count question.
Does This Work Differently With Smart Bidding?
Keyword count also interacts directly with Smart Bidding, and it’s worth addressing before finalizing your structure. Smart Bidding strategies rely on conversion data to optimize bids, and a keyword list that’s too fragmented, dozens of near-duplicate keywords each collecting a handful of conversions, actually works against Smart Bidding rather than helping it. Consolidating similar keywords, particularly given how broadly modern exact match already covers variants, tends to concentrate conversion data in a way that helps automated bidding strategies learn faster and more reliably.
If your account runs primarily on Target CPA, Target ROAS, or Maximize Conversions, this is a real reason to lean toward fewer, more consolidated keywords rather than a large, granular list, since fragmented data usually slows down the algorithm’s ability to find a stable pattern. Google’s own guidance on how Smart Bidding uses conversion data covers this mechanism in more depth if you want to understand exactly how the learning process works.
Regional Considerations for Lebanon and UAE Advertisers
Advertisers running smaller daily budgets, which is common for many businesses in Lebanon given the market’s generally lower advertising costs compared to the UAE, tend to see better results from a tighter, more focused keyword list. A smaller budget spread across too many keywords rarely accumulates enough data on any single term to make confident decisions.
In the UAE’s more competitive and costlier auction environment, campaigns are usually working with larger budgets to remain competitive to begin with, which does support a slightly broader keyword list, but the same underlying rule still applies: each keyword still needs enough of that larger budget behind it to generate real data, so a bigger budget should translate into better-supported ad groups, not simply a proportionally larger keyword list with the same thin coverage per term. In both markets, the mistake to avoid is the same: adding keywords faster than your budget can actually support them with meaningful traffic.
Common Mistakes When Managing Keyword Volume
A few patterns show up repeatedly in accounts struggling with keyword count:
- Treating “more keywords” as inherently safer, when an unfocused list dilutes both ad relevance and conversion data
- Never pruning keywords that have accumulated significant spend without conversions, letting dead weight sit in the account indefinitely
- Building ad groups around loosely related themes instead of one tightly defined concept per group
- Ignoring how modern exact match already covers close variants, leading to redundant keyword lists that were only necessary years ago
For the most current guidance on how match types behave, Google’s own documentation on keyword matching options is worth reviewing directly, since match type behavior has changed meaningfully in recent years and this is an area where outdated advice still circulates widely.
Restructuring accounts like this is a regular part of our Google Ads management service, we’ve inherited plenty of accounts built on the older assumption that more keywords always means more coverage. The typical fix looks like consolidating near-duplicate exact match terms that modern matching already covers, splitting overly broad ad groups into tighter concept-based groups, and building out a negative keyword list from months of accumulated search terms data that had never actually been reviewed. Most of the performance gain comes from that consolidation and cleanup, not from adding anything new.
Frequently Asked Questions
How many keywords should I use per ad group in Google Ads?
Most well-structured ad groups perform best with 5 to 20 closely related keywords. The right number depends on how many genuine variations exist for that specific concept, not a fixed target to hit.
How many keywords is too many for a Google Ads account?
There’s no universal ceiling, but a strong signal you’ve gone too far is a large share of keywords with zero or near-zero impressions, or difficulty writing ad copy that feels relevant to every term in a group.
Does having more keywords improve my Google Ads performance?
Not by itself. Performance depends more on how tightly your keywords match your ad copy and landing pages than on the total count. A smaller, well-targeted list generally outperforms a sprawling, loosely related one.
Should I use fewer keywords if I’m running Smart Bidding?
Often yes. Smart Bidding relies on concentrated conversion data, and a fragmented keyword list can slow down how quickly the algorithm identifies reliable patterns. Consolidating close variants tends to help automated bidding strategies more than a large, granular list does.
Are Single Keyword Ad Groups (SKAGs) still worth using?
Not as a whole-account strategy in most cases. Modern exact match already covers close variants automatically, which removes most of the original reason for building single-keyword ad groups. SKAGs can still make sense for a small number of your highest-value terms, but building an entire account this way is generally more maintenance than it’s worth now.
Does Performance Max still use keywords the same way?
No. Performance Max relies on asset groups, themes, and audience signals rather than a traditional keyword list, so the “how many keywords” question doesn’t apply to it the way it does to standard Search campaigns.
How many keywords should I start with if my account has no conversion history yet?
Start narrower than you think you need, build around your highest-confidence concepts first, and treat every keyword as still in its trial period until it has at least 15-20 clicks. The pruning thresholds and timelines that apply to established accounts aren’t reliable yet without that baseline data.
Is it true that modern exact match needs fewer keyword variations than it used to?
Yes. Google’s exact match now covers close variants, different word orders, and implied meaning automatically, which means accounts often need far fewer manually added variations than they did several years ago.
Making the Decision
The right number of keywords isn’t a fixed figure, it’s a byproduct of how tightly your ad groups are themed and how much budget and search volume your account actually has to support them. A smaller, well-structured list that gets real traffic, backed by a solid negative keyword list, will consistently outperform a large, unfocused one, regardless of what number either lands on.
Start with the Adchievers homepage to see the full range of paid media services we offer, or if you’d rather have someone audit your existing keyword structure and tell you specifically where it’s too broad or too thin, contact Adchievers and we’ll take a look at your account before you spend more time guessing.
Written by Elias Saliba | Founder, Adchievers
Elias is a performance marketing specialist with over ten years of experience running paid campaigns across Google, Meta, TikTok, LinkedIn, Snapchat, and X for 250+ clients in 28+ countries, including agencies and direct advertisers across Lebanon and the UAE. He has personally overseen more than $30M in managed ad spend and writes from direct, real-world campaign experience.