PPCleaner

Google Ads Reviews Searches Wasted? Measure Before Blocking

Published November 21, 202611 min read
An ecommerce owner checking advertising invoices beside a laptop, calculator and packed orders.
The short answer

Don't block "reviews" and "complaints" as a group. Searches like "brightpath solar reviews" or "is brightpath solar legit" often come from people who already picked you and are checking before they buy. Searches like "brightpath solar cancel contract", "brightpath solar employee reviews" or "competitor customer service number" almost never lead to a sale.

Sort the terms by who is searching, not by the word "reviews". Then judge the research-stage groups by what they assist, not only by last-click conversions. Block the clear cases with exact or phrase negatives. Keep a broad negative like reviews out of your account, because it also cuts off your own brand's review searches.

Pay for most brand review searches, block most complaint ones

Searches that pair your brand with "reviews", "legit" or "scam" usually come from people near the end of their decision, so they are usually worth paying for. Searches that pair your brand with "cancel", "refund", "login" or "customer service" come from people who already bought, or never will. Those are usually waste.

The word "reviews" tells you very little by itself. "brightpath solar reviews" and "brightpath solar employee reviews" share the word, but one comes from a homeowner comparing quotes and the other from a job seeker reading Glassdoor. "best solar companies reviews" is a third case: someone who hasn't picked anyone yet.

So the useful question is not "should I block reviews?" It is: for each kind of review or complaint search, does paying for the click lead to a sale, now or later? You can measure that. Most accounts never do.

Why "add reviews as a negative" is incomplete advice

Most wasted-spend guides put "reviews", "complaints" and "free" on the same starter list of negatives. The reasoning is that researchers don't buy. That holds for "free". For "reviews", it only holds if you judge by last-click conversions over a short window, and that is the wrong yardstick for this kind of search.

Look at what happens in practice. A homeowner gets three solar quotes. Before signing, they search "is brightpath solar legit". They click your ad, read your reviews page and leave. Four days later they search "brightpath solar" and sign. On last-click attribution, the brand click gets the conversion and the "legit" click gets nothing. Look at the search terms report on that basis and the "legit" term shows spend and zero conversions. It looks like waste.

Block it, and one of two things happens:

  • The homeowner clicks the top organic result instead. That might be your own site. It might be a Reddit thread or a complaint aggregator that ranks for "[your brand] scam".
  • They click a competitor ad bidding on your brand plus "reviews". Competitors do this on purpose, because they know who is searching.

The standard advice misses three things. Review searches are one of the few research-stage queries where the searcher has already named you. A competitor can buy that position if you leave it. And the cost of being wrong doesn't show up in the search terms report: you lose conversions that later get credited to another term, or that never happen.

This doesn't mean you keep every review term. It means you judge them on assist value, which takes a different method from the usual "spend with no conversions" filter.

Sort review and complaint terms into six groups

Before you measure anything, put each term in a group. The group predicts the intent better than the keyword in it. Here are the six we use when we audit accounts, with typical examples for a home solar installer.

GroupExample search termsUsual verdict
Own brand, pre-purchase checkbrightpath solar reviews, is brightpath solar legit, brightpath solar scamKeep and measure
Own brand, complaint researchbrightpath solar complaints, brightpath solar bbbMeasure, often keep
Own brand, existing customerbrightpath solar cancel contract, brightpath refund, brightpath solar loginBlock
Own brand, not a customer at allbrightpath solar employee reviews, brightpath solar glassdoor, brightpath solar lawsuitBlock
Competitor reviewssunrun reviews, sunpower complaintsMeasure, often block
Generic category reviewsbest solar companies reviews, solar panel reviews, are solar panels worth it redditMeasure, often block

A few notes on the less obvious rows.

Complaint research is not the same as complaining. "brightpath solar complaints" is often typed by a prospect doing due diligence, not by an angry customer. Existing customers with a problem usually search for a phone number or "cancel". This group is the closest call. It is also where you have a reputation reason to show up: an ad that links to your reviews page is the one result on that page you control.

Platform names are mixed. "brightpath solar bbb" and "brightpath solar trustpilot" are pre-purchase checks, but the searcher wants the platform, not your site. Click-through on your ad is often low, and since you pay per click, low click-through costs you little. Judge these on spend, not impressions.

Competitor reviews depend on your position. If you are cheaper or rated higher than the competitor, an ad on "sunrun reviews" that says so can take a share of their late-stage buyers. If you're not, you pay to tell someone about a company they weren't considering. Here the measurement decides, not a rule.

Generic category reviews sit furthest from a sale. "solar panel reviews" is mostly people researching hardware, and the results are full of affiliate sites. These look like the question searches covered in Google Ads Question Searches Wasted? Test Before You Block. Same logic applies: test before you block, but expect to block more of them than of the brand groups.

If you run lead generation, the existing-customer and not-a-customer groups also produce junk leads, not just wasted clicks. A job seeker who lands on a quote form sometimes fills it in. The lead generation negatives guide covers that side.

Measure the assist before you decide

To measure a review term, you need three things: a conversion window long enough to catch delayed buyers, keywords isolated so attribution can credit them, and, when you have the volume, a block test. Do them in that order. Each step is more work than the one before, and many accounts get an answer at step one or two.

First, pull the terms. In the search terms report, filter on "contains" for each of: review, reviews, complaint, complaints, legit, scam, bbb, trustpilot, reddit, yelp, glassdoor, cancel, refund. Use 90 days, so low-volume terms have time to add up. If the filter setup is new to you, these search terms report filters show where to find it.

Check your conversion window

The click-through conversion window is set per conversion action, from 1 to 90 days. If yours is 7 days and your sales cycle is three weeks, a review-stage click that leads to a signature on day 20 will never show a conversion, under any attribution model.

Check the window against your real time to purchase. For a $40 ecommerce order, 7 or 30 days may be fine. For solar, roofing, B2B software or anything with a quote stage, a short window makes review searches look worse than they are. Changing the window only affects new conversions, so you need a few weeks of data before you judge again.

Isolate the terms so attribution can see them

This is the step that changes decisions. Most review searches reach your account by matching your brand keyword. The conversion credit they get, if any, is mixed into the brand keyword's numbers, and you can't separate it.

Pull them out:

  1. Create an ad group in your brand campaign called something like "Brand – reviews".
  2. Add exact-match keywords for the terms worth testing: [brightpath solar reviews], [is brightpath solar legit], [brightpath solar complaints].
  3. Add the same terms as exact-match negatives in the main brand ad group, so the traffic goes to the new one.
  4. Write an ad for this intent. It should mention your rating, the number of reviews and a link to your reviews page. A generic brand ad wastes the click.

With data-driven attribution, Google can now give credit to these keywords even when the final click was on something else. You'll see fractional conversions, like 0.4 or 1.7, in the Conversions column for the reviews ad group. Under last-click those fractions are zero, and that is exactly why review terms look like waste in so many accounts.

This step has a cost. Splitting traffic means fewer conversions per ad group, and in a small brand campaign that can leave Smart Bidding short of data. If your brand campaign gets a handful of conversions a month, skip the split and go straight to the decision rule with whatever data you have. The Smart Bidding and junk queries article explains why bidding data quality matters here.

Run a block test where volume allows

Attribution models estimate. A block test measures. If you have enough volume, block one group (say, competitor reviews) for four to six weeks and watch three numbers:

  • total conversions across the account, not only in the brand campaign
  • impressions and clicks on your plain brand keyword (if review searchers stop reaching you, a drop shows up here first)
  • cost per conversion across the account

If total conversions hold and spend drops, the block was right. If brand conversions fall with no other cause, the review clicks were helping.

Be honest about the limit: in a small account, a four-week test can't separate a real drop from normal week-to-week noise. If you get 20 conversions a month, a swing of three or four means nothing. Small accounts should rely on the attribution step and on judgement, and keep the test for the expensive groups.

The decision rule

Once you have data, compare cost per assisted conversion with your target. A worked example:

  • The "Brand – reviews" ad group spent $600 over 90 days.
  • Last-click conversions: 0.
  • Data-driven conversions: 2.5.
  • Cost per data-driven conversion: $600 ÷ 2.5 = $240.
  • Target cost per acquisition: $200.

At $240 against a $200 target, this group is marginal, not waste. Before blocking, fix what's cheap to fix. A dedicated ad and a landing page built around reviews often lift the conversion rate of these clicks. Run it another 90 days. If the cost is still well above target, block the weakest terms in the group and keep the strongest. On last-click alone you would have blocked all of it on day one.

Now take the same $600 on "sunrun reviews" with 0 last-click and 0.2 data-driven conversions. That's $3,000 per conversion. Block it.

Block the clear cases with the right match type

Block existing-customer and not-a-customer terms with exact or phrase negatives, never with a broad negative on "reviews". A broad reviews negative blocks your own pre-purchase brand searches along with the junk, and broad negatives behave unpredictably across match types.

Here is a safe set for the solar example:

  • Phrase: "employee reviews", "glassdoor", "indeed", "careers". Job seekers, whatever brand they type.
  • Phrase: "cancel contract", "customer service number", "login". Existing customers. Use your own wording here, taken from your real terms.
  • Exact: [brightpath solar refund], [brightpath solar lawsuit]. Specific terms where a phrase negative might reach too far.
  • Phrase, competitor by competitor, once the test says so: "sunrun reviews", "sunpower complaints".

Watch out for phrase negatives that look narrow but aren't. "complaints" as a phrase negative blocks "brightpath solar complaints", which may be one of your better research terms. "scam" blocks "is brightpath solar a scam", which is a pre-purchase check. When in doubt, use exact. Exact match negatives are slower to maintain, but they never block something you didn't see. For more patterns you can copy, see phrase match negative keyword examples.

Put the job-seeker and existing-customer negatives in a shared list and apply it to every Search and Standard Shopping campaign. Those terms are waste in every campaign, not only in brand.

Review searches in Performance Max and Shopping

Review searches don't stay in your brand Search campaign. Performance Max and Shopping pick them up too, and each has different controls.

Performance Max. PMax often matches brand-plus-reviews searches, especially if brand exclusions aren't set. You can see these terms in the campaign_search_term_view report, which covers Search, Shopping and Performance Max. You can block them, but a Performance Max campaign takes only campaign-level negatives, up to 10,000, and doesn't accept shared lists. Your job-seeker and existing-customer list has to be added to each PMax campaign separately, and kept in sync by hand when the shared list changes.

There's a second issue. If PMax is taking your pre-purchase review searches, they're not reaching the "Brand – reviews" ad group you built to measure them, so your measurement is incomplete. Decide where these searches should land. Performance Max brand cannibalization covers how to spot the overlap, and the brand and non-brand split methods cover how to route it.

Standard Shopping. Product review searches like "dyson v15 reviews" or "weber spirit e-325 review" trigger Shopping ads more often than advertisers expect. The person is close to buying the exact product you list, and a Shopping ad with a price and a star rating is a good answer to that search. Don't block product review terms in Shopping without looking at the data. Block them only where they're plainly about something else, such as "dyson v15 reviews reddit" when that term has spent with no conversions over 90 days, or reviews of a model you don't stock. Standard Shopping accepts negatives and shared lists the same way Search does.

Display, Video and App report no search terms at all, so none of this applies there.

Where PPCleaner fits, and where it doesn't

PPCleaner reads the search terms your account paid for in Search, Shopping and Performance Max every night. It blocks the ones that were never going to convert. Job-seeker searches, existing customers looking for a login or a cancel page, and reviews of products you don't sell fall squarely in that group, and those are the blocks it makes. It never blocks a term that converted in the last 90 days. Every block can be undone in one click for 90 days.

The limit, stated plainly: PPCleaner sees conversions as your Google Ads account reports them. If your account uses last-click attribution and a short conversion window, a review term that only ever assisted shows zero conversions, in PPCleaner and in Google Ads. The tool can't credit a sale that your own conversion setup didn't record. That's why the first two measurement steps above are worth doing whatever tool you use. With data-driven attribution and a window that fits your sales cycle, an assisting review term carries fractional conversions, and the 90-day rule protects it.

PPCleaner doesn't run the block test or build the "Brand – reviews" ad group for you. It doesn't touch keywords, ads, bids or budgets. Measuring assist value is your decision, and this article gives you the method. Clearing out the plain junk around it every night, so it doesn't eat your budget while you test, is what PPCleaner does.

If you review terms by hand, add this category to your regular pass. The 20-minute search terms analysis has a place for it. Review terms are usually a small slice of spend, but on a brand campaign that slice falls on the people closest to buying.

negative keywordssearch termsattributionbrand campaignswasted spend

Frequently asked

Should I add "reviews" as a negative keyword in Google Ads?

Not as a broad or phrase negative on the single word. It blocks your own brand review searches, like "yourbrand reviews", which often come from people about to buy. Block specific groups instead: phrase negatives such as "employee reviews" and "glassdoor" for job seekers, and exact negatives for terms you have checked.

Why do review searches show spend and no conversions in my search terms report?

Often because the conversion happened later, on a different search. The buyer checks reviews, leaves, then comes back days later through a plain brand search that takes the credit under last-click attribution. A conversion window shorter than your sales cycle hides them as well. Data-driven attribution and a longer window show some of this assist value.

Are "complaints" searches always from unhappy customers?

No. "yourbrand complaints" is often typed by prospects checking you out before they buy. Existing customers with a problem more often search for a phone number, a login or "cancel". Measure complaint searches before blocking them. Showing an ad that links to your reviews page also puts a result you control at the top of the page.

Should I bid on competitor review searches like "competitor reviews"?

Only if you can show you are a better choice and the data backs it up. Isolate those terms in their own ad group, run them for 60 to 90 days with an ad that compares you directly, and compare cost per data-driven conversion with your target. If it is far above target, block them competitor by competitor with phrase or exact negatives.

How do I block review searches in Performance Max?

Add them as campaign-level negative keywords. A Performance Max campaign accepts up to 10,000 of them but does not accept shared negative keyword lists, so you have to add your job-seeker and existing-customer negatives to each PMax campaign separately and keep them in sync yourself.

About the author

Alain Abulafya

Founder of PPCleaner · Google Ads consultant & media buyer

Alain has worked in digital marketing for more than 15 years, over 10 of them managing and optimizing Google Ads campaigns. He has worked with more than 100 advertisers in e-commerce and lead generation, and he is a Google Partner. He built PPCleaner to automate the search term reviews he used to do by hand for his clients.

Stop the leak

Clean your search terms. Save time and money.

Every night, PPCleaner reviews the searches your campaigns actually paid for, groups them by intent, and excludes the ones that never convert. No spend threshold to wait for, nothing touched but negatives — and one click undoes any of it.

Start 7 days free

Search, Shopping and Performance Max · Cancel in one click

Related articles

Reviews and Complaints Searches in Google Ads: Pay or Block? — PPCleaner