Google Ads for Insurance Brokers: How to Avoid Low-Quality Comparison Traffic

You open Google Ads on Monday morning and the account looks busy. Clicks are coming in. Search terms are piling up. Spend is moving exactly as Google said it would. But the leads are weak, the phones aren't ringing with the right kind of enquiries, and half the forms look like people who only wanted the absolute cheapest policy.

That pattern is common in UK insurance PPC. It isn't usually a tracking glitch or a sales problem. It's low-quality comparison traffic, and it drains budget faster than most brokers realise.

Google Ads for Insurance Brokers: How to Avoid Low-Quality Comparison Traffic comes down to one principle. Stop treating every insurance search as valuable. In this market, a click from someone hunting for a bargain on a comparison site is not the same as a click from someone looking for advice, specialist cover, or a broker they can speak to. If your account doesn't separate those intents, Google will happily spend your money on both.

The Hidden Cost of Clicks in Insurance PPC

A UK insurance account can look healthy right up to the point you check what came in. Spend is pacing. Clicks are up. Search terms look busy. Then the lead sheet tells the full story. You are paying for people who want the cheapest quote, a comparison site, a claims number, or a job.

That gap between activity and value is where insurance PPC goes wrong.

Broad intent sits at the centre of it. Terms like insurance, car insurance, business insurance, or even a product line such as landlord insurance can pull in every kind of searcher. Some want advice from a broker. Plenty do not. In UK accounts, I regularly see spend leak into searches for renewals, reviews, careers, complaint routes, login pages, and price-led comparisons. The ad gets the click, but the user was never a fit for the enquiry you want.

Why insurance clicks go bad so quickly

Google still uses Quality Score as a diagnostic based on expected CTR, ad relevance, and landing page experience, scored from 1 to 10, and poor-fit traffic usually drags those signals in the wrong direction, as outlined in this analysis of low Quality Scores in local Google Ads.

Insurance is especially exposed because the UK search market has trained users to shop with comparison language. Compare the Market, Go.Compare, MoneySuperMarket and Confused.com have shaped behaviour for years. So a broker who bids too broadly does not just enter more auctions. They enter the wrong auctions, against players built for price-first clicks and huge volumes.

I see the same trade-off in account after account. Looser targeting can make top-line traffic look stronger for a week or two. It also pulls bidding models, search term expansion, and budget allocation towards users who were never likely to become good policyholders.

Low lead quality in insurance PPC usually comes from poor intent control, not a lack of clicks.

The cost isn't just bad leads

Weak traffic creates three problems at once. Conversion rate drops. Sales teams lose trust in the account. Google learns from mixed signals and starts chasing more of the same.

That is why account hygiene matters before scale. A broker cannot judge Google Ads cost per click on price alone. A cheap click from someone hunting for a comparison result is often wasted spend. A higher CPC from someone searching for specialist cover, local advice, or a broker they can speak to is usually worth far more.

The FCA angle matters here as well. Insurance ads and landing pages need to be clear about what is being offered and who it is for. If the traffic arriving on the page expected an instant price table and finds a broker-led advice journey instead, bounce rate is only part of the problem. The message match was wrong from the start.

What to check in your own account

A few patterns show up fast when comparison traffic is draining budget:

• Search terms with shopping intent: queries include compare, cheapest, best price, quote online, money supermarket, go compare, confused, or free.
• Traffic looking for something else entirely: searches around jobs, salary, training, claims, contact number, login, or reviews.
• Weak post-click behaviour: users land, skim, and leave because the ad entered a comparison-style auction but the landing page is built for a broker conversation.
• Poor impression share in the searches you want: Lost IS (Budget) means spend is being swallowed too early. Lost IS (Rank) often points to weak relevance caused by noisy traffic and broad targeting.

If clicks are rising but worthwhile enquiries are flat, the issue is usually not scale. It is filtration.

Building a Fortress with Negative Keywords and Placements

A UK insurance account can burn through a week of budget on searches that were never going to become saleable enquiries. I see it most often in broker campaigns that target broad product terms, then also pick up comparison shoppers, job seekers, claims queries, and people trying to reach an existing insurer. Until that waste is blocked properly, bid strategy has very little room to work.

Negative keywords are the control point. In insurance, they need to be stricter than in many other sectors because the search terms are messy and the compliance risk is higher. If the ad suggests advice-led broker support but the query shows clear price-comparison intent, the click is usually poor value and the message match is weak from the outset.

Start with campaign-wide comparison blockers

Build the first layer around the terms that signal aggregator behaviour. In UK accounts, that usually means variations of compare, cheapest, best price, cheap, free quote, and brand-led comparison searches such as Compare the Market, Go.Compare, Confused.com, and MoneySuperMarket.

The aim is not to block every price-sensitive search. Plenty of legitimate prospects still care about price. The aim is to stop paying for auctions where the user wants an instant comparison table and is unlikely to engage with a broker-led journey.

A practical starting list usually includes:

• Comparison language: compare, comparison, compare quotes, compare the market, best price, cheapest
• Aggregator brands: money supermarket, moneysupermarket, go compare, gocompare, confused, comparethemarket
• Freebie intent: free policy, free quote, no obligation free
• Research-only modifiers: reviews, review, ratings, forum
• Non-new-business service intent: claims, claim form, customer service, login, contact number
• Employment and training: jobs, careers, salary, training, courses, apprenticeship

Use match types properly. Broad negatives can save time, but in insurance they can also block good volume if they are added carelessly. Phrase and exact negatives usually give more control, especially in specialist lines such as landlords, fleet, HMO, or professional indemnity where the query set is narrower and easier to shape.

Build shared negative lists at account level

Shared lists stop the same bad traffic leaking into every campaign. That matters in broker accounts with separate campaigns for motor, home, commercial, renewals, and brand protection, because the same junk terms show up repeatedly.

I usually split shared negatives into distinct lists rather than one giant bucket:

• Comparison and aggregator terms
• Jobs and recruitment
• Claims and customer service
• Research and reviews
• Low-intent freebie terms
• Competitor brands you do not want to pay for

That structure makes maintenance easier. It also reduces mistakes. If a broker offers claims support or runs a hiring campaign, that list can be removed from the relevant campaign without dismantling everything else.

For a cleaner build process, the filtering work needs to happen before campaigns scale. Good keyword research for PPC should separate target searches from exclusion themes early, not after a month of wasted spend.

Example negative keyword list for insurance brokers

No list is universal. A broker targeting “temporary car insurance” may still want some price-led language if the landing page and offer are built for it. A high-net-worth or specialist commercial broker usually needs a much harder line because comparison traffic almost never turns into quality business there.

Mine the Search Terms report every week

Weekly search term reviews are where this gets sharpened. Monthly is too slow in active insurance accounts, especially if broad match, Smart Bidding, or PMax are in play.

The useful pattern is to sort queries into three groups.

1. Add as negatives now
   Clear mismatches such as jobs, complaints, training, claim helpline, login, and comparison brands.

2. Monitor closely
   Queries that look relevant on the surface but produce weak leads, low engagement, or poor call quality. Terms like best insurer, reviews, or cheap specialist cover often sit here.

3. Promote into their own build
   Long-tail searches with clear broker intent, such as local commercial cover, unusual risk types, or searches that mention advice, specialist support, or a named policy type.

That discipline matters even more after the recent changes to search matching and PMax controls. Google has improved visibility in places, but it still does not protect insurance budgets for you. In 2025 and 2026 setups, negative lists and search term governance are still doing a lot of the heavy lifting.

Don’t ignore placements if Display or PMax is active

Placements are the second wall. They matter less in pure Search, but they matter a lot once Display inventory, YouTube, Discover, apps, or PMax enter the account.

Insurance advertisers often leak spend into mobile apps, low-quality parked content, and made-for-advertising sites where clicks are cheap and useless. Those clicks can make lead numbers look healthy while sales teams complain that nothing is quote-ready.

Use account-level placement exclusions where possible. Exclude obvious poor-fit app categories. Review website placements for low-engagement inventory. If PMax is running, check placement reports and use brand suitability and account-level exclusions to keep the campaign away from inventory that has no realistic path to an enquiry. That is even more important for FCA-sensitive products, where ad context matters as well as click quality.

A good setup is stricter than many advertisers expect. That is usually the point. In UK insurance PPC, profitable traffic rarely comes from trying to be everywhere. It comes from being selective enough to avoid the comparison swamp.

Refining Your Targeting Beyond Keywords

A broker in Leeds can target the same keyword as a broker in Croydon and get completely different lead quality. The search term alone does not explain the gap. Postcode mix, device habits, time of day, household profile, and whether the user behaves like a quote collector or a genuine buyer all shape what happens after the click.

In UK insurance accounts, targeting gets cleaner when keywords, audiences, geography, and first-party data all point in the same direction. That matters even more now that PMax signal controls have improved in 2025 and 2026, but still need firm boundaries if you want to keep comparison traffic out.

Use audiences as filters, not as a reach tool

The safest setup is observation first. Add in-market and custom segments, then review them against quoted cases, policy starts, and sales feedback before changing bids or feeding them into PMax audience signals.

I usually split audience work into three buckets:

• Intent audiences: in-market insurance categories and custom segments built from specialist policy language
• Risk audiences: segments that produce form fills but low contact rates, weak call quality, or poor close rates
• Exclusions or downgrades: audiences that repeatedly overlap with bargain-hunting or broad comparison behaviour

That sounds basic. It saves money.

If you're already using audience targeting signals in PPC campaigns, insurance is one of the few sectors where restraint usually beats expansion. Broader audience settings can help volume, but they also make it easier for Google to chase cheap clicks that look active and convert badly.

One source that pulls several of these tactics together is this Google Ads strategy for insurance agencies, which notes that UK geo-fencing, audience exclusions tied to comparison behaviour, postcode-level bid adjustments, and tightly controlled DSA can all help filter weaker traffic. The exact gains will vary by policy line and by broker, but the direction is right. Better targeting comes from narrowing who sees the ad, where they see it, and which searches you allow exploratory campaign types to cover.

Geo-targeting should follow where deals actually close

County-level targeting is often too blunt for brokers. I have seen accounts where two adjacent postcodes produce completely different results because one area is full of owner-occupied family homes and the other is dominated by short-term renters, students, or businesses outside the broker's sweet spot.

Use location reports against real outcomes, not just CPA. A cheaper lead from outer London may look fine in-platform and still waste the sales team's week if quote requests never progress. A more expensive enquiry from Surrey or Cheshire may be worth more if it turns into multi-policy business or a cleaner renewal book.

Good geo decisions usually come from questions like these:

That last point gets missed. If the ad implies local advice or specialist handling, the landing page and follow-up process need to support it. FCA-sensitive categories leave little room for vague positioning.

Dynamic Search Ads still have a place, but only inside a fenced area

DSA can help insurance brokers find useful long-tail intent, especially for specialist commercial lines, non-standard property, or niche add-ons your keyword structure has not covered yet. It can also waste spend quickly if the site is broad, outdated, or full of thin pages.

The safest approach is to keep DSA on a short leash:

• target only approved URL groups or tightly defined landing pages
• exclude blog content, careers pages, claims pages, and generic service pages
• review search terms and landing page matches every week
• mine converting queries into exact or phrase campaigns, then cut them out of DSA where appropriate

I treat DSA as a harvesting tool. It should find missed intent and feed the main account structure. It should not become the account structure.

That distinction matters more after the latest automation updates. Google is better at finding inventory than it is at judging whether a motor trade lead, a landlord enquiry, or an SME package request matches the kind of client your brokerage wants.

Use your own data to tell Google who a good lead is

Keywords and audiences get you part of the way. CRM feedback finishes the job.

Import offline conversions where possible. Split out quoted, bound, and retained business if your setup allows it. Feed Enhanced Conversions and first-party lists back into Search and PMax so the system can see the difference between a soft form fill and a policy that makes commercial sense.

In broker accounts with enough volume, that is usually where targeting starts to improve properly. Google stops optimising around people who like filling in forms and starts getting closer to people who buy cover through a broker.

Aligning Ad Copy and Landing Pages for High Intent

Ad copy does more than improve click-through rate. In insurance, it qualifies the click before it happens.

If your ads sound like a comparison site, don't be surprised when comparison shoppers click them. If your landing page opens with “cheap quotes” language, you're telling bargain hunters they’re in the right place.

Stop writing ads for everyone

The UK insurance market rewards relevance, not generic promises. That matters even more after Google’s March 2026 intent scoring update, which penalises low-engagement ads. The same UK broker compliance analysis notes that many insurance brokers are sitting on below-average Quality Scores, while A/B tests showed location-insertion and call extensions can improve scores and help filter comparison traffic at the same time.

That gives brokers a clear direction. Ads should sound like a broker, not a price engine.

Good filtering language usually leans into:

• Broker expertise: specialist broker advice, customized cover, policy review
• Human support: speak to an adviser, UK-based team, discuss your cover
• Specific cover needs: landlord cover, fleet policies, tradesman liability, non-standard home insurance
• Local credibility: location insertions where they effectively help relevance

Poor filtering language usually leans into:

• cheapest
• best price
• instant quote with no context
• vague “save money” claims

That last point also matters for compliance. FCA-sensitive sectors can't afford lazy superlatives that invite disapprovals or attract the wrong click in the first place.

Your landing page should continue the filter

A strong ad can still be wasted by a weak landing page. If the user clicks expecting specialist help and lands on a thin, generic quote form, intent drops immediately.

I prefer landing pages that prove three things fast:

1. Who you help
   Name the policy type, sector, or customer profile clearly.

2. Why a broker is useful
   Explain the advisory angle. Market access, customized cover, unusual risk placement, renewal review, claims support.

3. What happens next
   Set the next step properly. Consultation request, callback, personalized quotation, document review.

A good landing page doesn't just persuade. It screens out people who only want a bargain and keeps the ones who want advice.

This walkthrough is worth watching if you're tightening the conversion journey from ad to lead form.

Message match is what protects Quality Score

When ad copy, keyword intent, and landing page all line up, Google sees a cleaner relevance signal and users see a clearer offer. That’s good for Quality Score, but it’s also good for lead quality because the message filters before the click and after it.

A broker selling advice-led cover should sound advice-led at every stage. If the campaign promises personal service but the page feels like a stripped-down quote aggregator, the account is working against itself.

Navigating Smart Bidding and Performance Max Safely

A familiar insurance PPC pattern goes like this. Search campaigns start producing decent enquiries, Google recommends Target CPA and Performance Max, the account follows the recommendation, and within a few weeks the lead numbers look healthier while sales quality drops. Call handlers start hearing, "I'm just comparing prices," far more often.

That shift is predictable. Automation follows the signals you give it. If the account is still feeding Google a mixed bag of quote requests, weak calls, and comparison-led form fills, Smart Bidding usually buys more of the same.

Why automation misfires in broker accounts

Insurance is a poor fit for early automation because the easiest conversion is rarely the best conversion. Google can spot a form completion. It cannot reliably tell the difference between a serious fleet prospect and someone checking five brokers in one lunch break unless your tracking setup makes that distinction clear.

I usually keep tighter control at the start. Manual CPC or Enhanced CPC is often the better choice while the account is still being cleaned up, lead tracking is being fixed, and search campaigns are proving which queries bring in business rather than noise.

That slower start saves money later.

Performance Max needs tighter boundaries in insurance

Performance Max can work, but it needs a narrower job description than Google suggests. In UK insurance accounts, I see it perform best when it supports search activity rather than replacing it, especially for remarketing, renewals, branded reinforcement, and known customer lists.

Used too early, it broadens reach into placements and audience pools that look active in-platform but produce weak commercial value. That matters even more for brokers working under FCA expectations around clear promotions, fair messaging, and suitable customer journeys. If Google starts assembling combinations of assets across lower-quality inventory, the problem is not just wasted spend. It can also create compliance headaches that do not show up in a standard campaign summary.

A safer rollout plan

I prefer to introduce automation in stages.

1. Prove intent on Search first

Get search campaigns into shape before handing more control to the machine. That means stable query quality, clear conversion actions, sensible negatives, and enough evidence that the leads turning up are worth the follow-up time.

2. Test Smart Bidding where signals are clean

Start with the campaigns that already produce the best enquiries. Keep budgets controlled, watch lead quality every week, and resist judging success on CPA alone. A cheaper cost per lead is meaningless if the close rate falls.

3. Keep PMax on a short leash

For brokers, that usually means using first-party audiences, customer match lists, remarketing segments, brand exclusions where appropriate, account-level negative keywords, and regular asset and placement reviews. Google has added more controls into PMax across 2025 and 2026, but "more" does not mean "enough". You still need active oversight.

If you want a current view on what control is available, this guide on Performance Max in 2026 and how to control Google's automated campaigns is worth reading alongside your own account data.

Guardrails that stop drift

The accounts that hold quality under automation usually have the same protections in place:

• Qualified conversion inputs: feed in calls, consultation requests, and lead stages that reflect sale likelihood, not just raw form volume
• Strict exclusions: use account-level negatives and placement exclusions to block comparison terms, job seekers, research traffic, and irrelevant content
• Conservative budgets: give new automated campaigns enough spend to learn, but not enough to waste a month of budget before problems are spotted
• Audience discipline: prioritise first-party data, existing customer lists, renewal audiences, and site visitors who have shown real intent
• Compliance checks: review asset combinations, final URLs, and auto-generated elements so the campaign stays consistent with FCA expectations and your own approval process

One hard lesson from UK broker accounts is that PMax often looks better in Google Ads than it looks in the CRM. The platform reports a conversion. The sales team reports a time-waster. In insurance, the CRM usually tells the truth faster.

The trade-off is simple

Automation saves bidding time. It also reduces visibility and gives Google more room to chase easy conversions.

That trade-off is acceptable once the account has clean signals and clear exclusions. Before that point, it is usually an expensive shortcut. The brokers who get good results from Smart Bidding do not hand over control and hope for the best. They set limits, review search and sales quality closely, and treat PMax as a supervised channel, not a default setting.

Measuring What Matters and Proactive Optimisation

Insurance accounts improve when lead quality becomes visible. If every enquiry is treated as equal, Google learns the wrong lessons and the account keeps drifting toward cheap, easy conversions.

That’s why measurement has to reflect commercial reality. A rushed quote form from someone shopping ten brokers at once is not the same as a consultation request from someone with a non-standard risk.

Track lead types separately

At minimum, I’d split conversion actions by intent and sales value. That usually means separate actions for:

• Quote request forms: useful, but often mixed quality
• Consultation or callback requests: usually stronger buying intent
• Phone calls: especially important for broker-led services
• Returning visitor enquiries: often stronger than first-click form fillers

The point isn't to create reporting clutter. It's to stop the platform from optimising toward the easiest form submission if that lead rarely turns into business.

A practical tracking setup should also include call tracking so phone leads don't disappear from the picture. Insurance buyers often want to speak to someone before committing, especially for commercial, specialist, or unusual cover.

Watch the metrics that expose quality issues

Insurance PPC doesn't need more vanity reporting. It needs a short list of metrics that show where quality is leaking.

I focus heavily on:

Those last two are where many brokers find the truth. The campaign with fewer leads often produces the better business.

A repeatable optimisation routine

A good optimisation cycle doesn't need to be elaborate. It needs to happen consistently.

Weekly checks

• Review Search Terms: add negatives, flag useful long-tail queries, cut recurring junk.
• Check lead quality by campaign: not just count, but sales fit.
• Review audience segments: look for weak or irrelevant traffic patterns.
• Inspect ad messaging: make sure new assets haven't drifted into vague, price-led language.
• Review any automated campaign surfaces: especially if PMax or broader networks are active.

Monthly checks

• Assess conversion actions: are they still weighted toward useful lead types?
• Review lost impression share trends: identify whether budget or rank is limiting the campaigns worth scaling.
• Check landing page alignment: message match slips over time, especially after site updates.
• Compare postcode and location patterns: shift spend toward areas producing stronger commercial outcomes.

Good optimisation is mostly routine. The expensive mistakes happen when nobody checks the account closely enough to notice what Google has started buying.

Keep sales feedback in the loop

Many PPC setups struggle due to a common disconnect: Marketing sees a conversion. Sales sees a poor-fit lead. If that feedback never goes back into the account, Google keeps chasing more of the same.

The fastest gains usually come from combining platform data with human judgement. Which forms were serious? Which calls were just comparison shoppers? Which policies closed? That feedback shapes keyword decisions, ad copy, audience exclusions, and eventually bidding strategy.

Conclusion From Wasted Spend to High-Value Clients

Low-quality comparison traffic doesn't disappear because you lower bids or rewrite a few headlines. Brokers fix it by building a tighter account from the ground up. Strong negatives. Smarter audience signals. Ad copy that sounds like a broker, not a comparison site. Landing pages that qualify the click. Automation used carefully, not blindly.

The bigger shift is strategic. Stop judging success by click volume and raw lead count. Judge it by whether the campaign attracts people who want advice, specialist cover, and a real conversation.

That’s how Google Ads for Insurance Brokers: How to Avoid Low-Quality Comparison Traffic becomes a profitable channel instead of a constant fight with wasted spend. The brokers who win in search aren't buying more traffic. They're filtering harder and optimising with more discipline.

If you want a second pair of eyes on your account, PPC Geeks offers UK-led Google Ads support for businesses that need tighter targeting, cleaner lead generation, and less wasted spend. They're a specialist PPC agency with deep experience across lead gen, ecommerce, remarketing, landing pages, and campaign optimisation, with a strong focus on transparent reporting and practical account control.