Cost Per Click Control: Search Term Hygiene, Match Types, and Bid Rules That Cut Wasted Spend

A rising cost per click is rarely one problem. It can come from more competition, a looser keyword set, an ad that earns fewer clicks, or a bidding system chasing volume. Teams often react by cutting bids across the board, which lowers the price and the results together.

This article is about control, not definitions. If you want to understand how pay-per-click advertising works from the ground up, start with our guide to pay-per-click campaigns. Here we cover the routines that keep the clicks you buy relevant: reading search terms, choosing match types, building negative lists, setting bid rules, and judging the result by cost per qualified lead instead of cost per click.

Separate Price Problems From Relevance Problems

Before changing anything, decide which kind of problem you have.

A price problem means you pay more for the same kind of click. Signs include a higher average cost per click on stable keywords, lower impression share lost to rank, or new competitors appearing in auction insights.

A relevance problem means you pay a normal price for the wrong clicks. Signs include search terms unrelated to the offer, a falling conversion rate, rising rejection reasons from sales, or a growing share of clicks from queries you would never have chosen.

The fixes differ. Price problems call for better ad quality, sharper landing pages, and bid decisions by segment. Relevance problems call for tighter targeting and negative keywords. Mixing them up wastes weeks.

Build a Weekly Search Term Routine

The search terms report shows the actual queries that triggered your ads. It is the most direct view of where money goes, and it takes about thirty minutes a week once you have a routine.

1. Sort by cost, highest first, over the last 30 days. Review the top terms before anything else, since they hold most of the spend.
2. Mark each term as relevant, irrelevant, or unclear. Relevant terms with good outcomes may deserve their own keyword. Irrelevant terms become negatives.
3. Look for patterns, not only single terms. A cluster of queries containing words such as "free", "jobs", "course", or "diy" usually points to a whole class of unwanted intent.
4. Check terms with clicks and no conversions, but be careful about volume. A term with three clicks and no leads is not proof of anything. Use a rule such as acting only when spend passes a multiple of your target cost per lead.
5. Record what you changed, so you can see the effect in the next review.

Illustrative example. A home services campaign spends $4,000 in a month. The report shows $600 of that went to queries containing "salary" and "training", from people looking for careers and courses. Adding those words as negatives frees up 15% of spend for the intent that actually produces enquiries. No bid change was needed.

Choose Match Types on Purpose

Match types set how closely a search must resemble your keyword. Platform definitions change, so check the current help pages, but the broad logic holds.

• Exact match: ads show for searches with the same meaning as the keyword, including close variants. It gives the most control and the least volume.
• Phrase match: ads show for searches that include the meaning of the keyword, with other words allowed around it. It is a reasonable middle ground for core terms.
• Broad match: ads can show for a wide range of related searches. It can find new demand, but without strong conversion data and tight negatives it can also spend quickly on loose matches.

A common approach is to start with exact and phrase on your most valuable terms, review the search terms report, and then test broad match in a separate campaign with a limited budget and conversion goals tied to qualified leads. Keep it separate so you can see what it does, and so it cannot drain the budget of campaigns that already work.

Whichever type you use, remember that automated bidding combined with broad targeting depends heavily on the quality of the conversion signal you provide. If the platform only sees form fills, it may favor cheap, low-intent searches. Passing back qualified-lead events, covered in our lead quality guide, gives it a better target.

Organize Negative Keywords in Layers

Negatives are the main protection against irrelevant clicks, and they work best when organized.

• Account-level lists: terms that never apply, such as jobs, free, torrent, or your own staff-related words.
• Campaign-level lists: terms that apply to one service but not another, for example excluding "repair" from a campaign selling new installations.
• Ad group-level negatives: used to steer queries toward the most suitable ad and landing page, so two groups do not compete for the same search.

Keep a shared sheet of negatives with the date added, the reason, and the person who added it. Review it quarterly, because a negative added in haste can block valuable traffic. Check search volume and conversion history before adding single, broad words such as "cheap" or "near", which may carry real intent in your market.

Use Bids as a Rule Set, Not a Mood

Bid changes made on gut feeling are hard to evaluate and harder to repeat. A small rule set is better. Each rule has a trigger, an action, and a review date.

• If a keyword has spent more than twice the target cost per lead with no leads, reduce the bid or pause it.
• If a keyword produces qualified leads below target cost and has lost impression share to rank, raise the bid in small steps.
• If a device or location produces a much higher cost per qualified lead over enough conversions, apply a bid adjustment or split it into its own campaign.
• If a campaign hits its daily budget early every day on strong results, raise the budget before raising bids.

When you use automated bidding strategies, bid caps and target cost settings act as guardrails. They can limit how much the system pays per click or per conversion, though tight limits can also reduce volume. Treat them as controls to test, not defaults to set once. Remember too that automated systems need time to learn after changes, so judge them over a sensible period and avoid piling changes on top of each other.

Read Impression Share to Find the Ceiling

Impression share shows how often your ads appeared compared with how often they could have. The lost share is split between budget and rank.

If you are losing share to budget on profitable campaigns, you are leaving good demand unbought, and budget is the lever. If you are losing share to rank, your bid or ad quality is too low relative to competitors for those auctions. Raising bids may help, but improving ad relevance and landing page experience often lowers the price you need to pay.

Impression share also tells you where to stop. When share on your best terms is already high, more budget buys clicks of lower quality, and the marginal cost climbs, as explained in our post on ad budget allocation and scaling.

Improve Click Quality Through the Ad Itself

Your ad is a filter. Copy that states who the offer is for, where it applies, and roughly what it costs discourages clicks from people who would never buy. A lower click-through rate is not always bad if the people who do click convert at a higher rate.

Practical adjustments include naming the service area in the headline, stating a starting price or a minimum order when that suits your model, using the final URL that best matches the query, and adding sitelinks and callouts that answer common questions before the click. Test these using the methods in our A/B testing guide, and judge them by cost per qualified lead.

Judge Cost Per Click in Context

A higher cost per click is acceptable when the conversion rate and quality rise with it. A low cost per click is not a win if the clicks do not convert. The relationship is simple: cost per lead equals cost per click divided by conversion rate.

Illustrative example. Keyword A costs $2 per click and converts at 4%, so a lead costs $50. Keyword B costs $3.5 per click and converts at 10%, so a lead costs $35. If B also produces a higher qualified rate, it is much better than A despite the higher click price. Reports that rank keywords by click cost would point you the wrong way.

For the full set of measures and how they connect, see our guide to performance marketing KPIs. Before trusting any of these numbers, confirm that conversions are recorded correctly using the conversion tracking audit checklist.

Common Mistakes

• Cutting every bid because average cost per click went up
• Adding negatives from three clicks of data
• Leaving broad match in a campaign with no negatives and loose conversion goals
• Judging keywords by click-through rate instead of qualified leads
• Making several bid, budget, and ad changes in the same week
• Never reviewing the negative list after adding terms

Frequently Asked Questions

How often should I review search terms?

Weekly for active campaigns with meaningful spend. Smaller accounts can review every two weeks, as long as they look at the highest-cost terms.

Is a lower cost per click always better?

No. Cost per click matters only in relation to conversion rate and lead quality. Compare cost per qualified lead.

Should I avoid broad match completely?

Not necessarily. It can work with strong conversion tracking, tight negatives, and a controlled budget. Test it in a separate campaign.

How many negatives are too many?

There is no fixed number. Quality matters more than count. Each negative should have a reason, and you should review the list periodically.

Key Takeaways

Diagnose whether you have a price or relevance problem, review search terms every week, choose match types deliberately, organize negatives in layers, and manage bids with written rules. Judge every change by cost per qualified lead, not cost per click.