Marketing News & Analysis
Google Ads for B2B Marketers: Use Competitor URLs to Test Which Ads Bring In Qualified Leads
A practical B2B playbook for using public URLs as audience signals, testing customer testimonials against product ads, and measuring qualified pipeline.
A head of marketing sees a competitor’s ads after visiting their own company’s website and asks a reasonable question: could a rival use Google Ads to follow our leads? Seeing an ad does not establish why it was served. But there is a real tactic behind the question. Google Ads lets advertisers enter website URLs when building a custom segment, and publicly visible competitor pages can help an advertiser describe the type of audience it wants to reach.
The distinction matters. A competitor URL is an input to an audience model. It does not identify who visited the URL, completed a form, or became a sales opportunity. B2B teams can still test whether people who browse similar websites respond to a relevant message. The right question is whether those ads produce qualified conversations and pipeline at an acceptable cost.
What Google Ads does with a competitor URL
When you choose the URL option in a Google Ads custom segment, you enter websites your ideal customers might browse. Google describes the resulting audience as people who browse websites similar to the URLs you enter. The exact delivery depends on campaign settings, and entering a URL also does not buy placement on that website. A competitor’s pricing or product page is a way to describe the kind of browsing behavior you are interested in. The page cannot give you access to a competitor’s analytics, conversion events, or CRM. Google Ads guidance on custom segments
This is why the thank-you page shortcut is misleading. Even if you know the address of a competitor’s form confirmation page, adding it to a custom segment cannot turn it into a list of people who submitted that form. You have no reliable basis for saying that anyone in the resulting audience saw the competitor’s LinkedIn ad, visited the specific page, asked for a demo, or is ready to switch vendors. The audience may be relevant enough to test, but the inputs do not prove intent at the person level.
A prospect can see your ad because of a separate search, browsing behavior, an existing audience relationship, or other campaign signals. If someone on your team sees a competitor’s ad soon after visiting a site, use the observation as a prompt to inspect your own targeting and measurement. A single impression cannot establish that the visit triggered the ad.
Keep prospecting, remarketing, and placements separate
The same URL can appear in different parts of advertising work, so teams sometimes use the word retargeting for three different jobs. Write down which job you mean before building the campaign.
| Approach | What you supply | What it helps you do |
|---|---|---|
| URL-based custom segment | Relevant public URLs | Describe an audience using similar browsing behavior |
| First-party website remarketing | Eligible activity on your own tagged site | Reach people who previously interacted with your business |
| Placement targeting | Eligible ad inventory or placements | Choose where an ad may appear, subject to campaign settings |
A first-party website segment can use events you collect on your own site, such as a visit to your integration page or a completed form, subject to eligibility and applicable requirements. That is a different input and a different relationship with the prospect. If you have enough eligible visitors, create a separate remarketing campaign or ad group for them and give it its own reporting. Google Ads guidance on your data segments
It helps to label the campaign plainly. A name such as “US B2B prospecting / vendor URL signals / testimonial test” will age better than “Competitor leads retargeting.” The label tells the next person what you actually tested and makes it harder to blend modeled reach with first-party audiences when you review results.
Build a short list of URLs that describe your buyer
Start with one buying job. Imagine your company sells software that helps mid-market finance teams close their books. The buyer might research close management platforms, reconciliation workflows, ERP integrations, and implementation requirements. A general financial news site could attract millions of people who are interested in finance and very few people trying to buy close software. A relevant vendor’s product page or a narrowly focused category resource gives the team a clearer reason for inclusion.
Build the shortlist from pages you can inspect publicly. Use competitor ads to learn how the category talks about the problem and to identify public landing pages. Compare those pages with the pages your own best-fit buyers use before a demo. You can also include public pages from adjacent vendors if they solve a step in the same workflow. A page that helps finance teams compare integrations may belong in the test even if the site does not sell a direct substitute.
For each URL, write one sentence explaining the buyer job it represents. If you cannot do that without guessing, put the URL aside. Keep comparison and product pages together when they serve the same decision, and separate broad education if you want to test it as another audience later. When the niche is small, add adjacent pages that share the workflow before adding unrelated high-traffic sites. That gives you a better chance of learning which signals belong together.
Record the URLs, the buyer role, the use case, the campaign type, the geographic and language settings, and the date you made the selection. Save the initial list before launch. When you need to change the audience, you can then compare versions and connect any performance change with what actually changed.
Set up the campaign around one question
Start with a hypothesis the team can evaluate: for this modeled audience, a testimonial about implementation will generate more qualified opportunities per dollar than an ad focused on product features. Use an eligible campaign type that offers custom segments for the setup you are running. Google’s documentation covers the available campaign types and explains that Performance Max does not accept custom segments as direct audience targeting. Verify the options in your account before you commit the budget. Google Ads custom segment setup
Create a URL-based segment in Audience Manager, then apply it in the eligible campaign or ad group. Check the location, language, exclusions, and any audience expansion settings that the campaign offers. If you leave expansion enabled, note that choice in your test plan and keep it consistent across both variants. Otherwise, it becomes harder to tell whether a difference came from the ad or a change in whom the campaign could reach.
Decide which existing contacts should be handled separately. If the question is acquisition, eligible first-party visitors and existing customers may need to be excluded or reported independently, depending on the controls available. Exclusions and deduplication take work, but otherwise a testimonial variant might look unusually strong because it reached more people who already knew you.
Keep the first test narrow. Use one buyer job, one modeled audience, one offer, and two creative approaches. Where the campaign offers a supported experiment, use it to set up a cleaner comparison. If you run two ads together and let the platform favor one early, treat the result as a directional comparison. Changes in exposure can matter even when the audience settings look identical.
Test a product message against a real customer testimonial
A product-led ad gives the buyer something concrete to inspect. For the finance example, it might show how a reviewer sees outstanding reconciliations, which systems the software connects to, or what an audit trail looks like in the product. Start with a single capability that addresses the buyer job behind the URLs. A tour of every feature makes it difficult to know what drove the response.
A testimonial should answer a question someone on the buying committee would actually ask. It could speak to rollout, the first workflow the team moved into the product, or how the customer handled a dependency their peers will recognize. Use an approved quote or case study and keep the scope clear. If the customer describes a specific result, show the relevant context and avoid turning it into a promise for every buyer. The best proof gives the prospect something to check on the landing page or discuss in a demo.
Use the same offer for both ads. If the product ad links to a feature page and the testimonial links to a low-friction checklist, any performance gap may come from the offer rather than the creative. Send both to a relevant page that can answer the questions raised in both versions. Keep format, placements where possible, geography, dates, and call to action comparable. Give each variant its own tracking identifier so the CRM can preserve the distinction.
Write for someone exploring a problem. A modeled audience includes people with different levels of knowledge. Copy that assumes a prospect has submitted a competitor’s form can feel out of place and has no basis in the targeting data. A plain question about deployment, integration, or team workflow gives both an early researcher and an active evaluator a reason to pay attention.
Judge the test on qualified pipeline
Decide with sales what counts as qualified before spending. For some teams, that means a meeting with a target account, a relevant role, and an active business need. For others, the first reliable milestone is a sales-accepted opportunity. Pick the earliest stage that is consistently recorded and meaningfully related to revenue. Define how you will treat duplicate forms, disqualified accounts, existing opportunities, and leads from outside the target segment.
Track impressions and clicks to diagnose delivery. Then compare form submissions, qualified opportunities, spend per qualified opportunity, and pipeline value by variant. Cost per form is easy to read and can be useful as a guardrail. It is less useful as a decision metric when a large share of forms never make it to sales. Pipeline value should use consistent rules for stage and amount, especially if a small number of deals dominate the total.
Here is an illustrative example. The product-led ad spends $4,000, generates 40 forms, and yields four qualified opportunities. The testimonial ad spends the same amount, generates 25 forms, and yields eight qualified opportunities. The product-led ad wins on cost per form at $100 versus $160. The testimonial wins on cost per qualified opportunity at $500 versus $1,000. Eight opportunities are still a small sample, so the right next step might be another test cycle rather than a large budget shift. The example shows why a team should agree on the decision metric before looking at results.
Decide how long you will allow for leads to mature. If your median time from inquiry to qualification is two weeks, comparing both ads at the end of their first week will miss much of the outcome you care about. Use a consistent lookback window and compare cohorts that had the same time to progress. Separate early indicators from the later pipeline readout in the report so stakeholders know which decisions the data can support.
If you have a reliable CRM and permission to use the required data, consider feeding eligible qualified or converted lead events back into Google Ads. Google’s enhanced conversions for leads documentation explains how eligible first-party information can help match offline outcomes back to ad interactions. Check that conversion actions, timestamps, and deduplication are working before using the data to guide bidding. Maintain a CRM report as an independent check on the platform numbers. Google Ads guidance on enhanced conversions for leads
Decide what to change after the test
If both ads draw traffic but almost no qualified leads, start with the audience and offer. Review whether your URLs describe the actual buying job and whether the landing page answers the questions raised by the ads. If the visitor is evaluating implementation and the page only asks for a demo, it may need a short explanation of how deployment works. Before adding budget, inspect which CRM disqualification reasons are appearing most often.
If the testimonial wins, test which objection it answered. You might compare approved proof about implementation with approved proof about workflow adoption, while holding the rest of the creative and offer steady. If the product-led message wins, look at which capability resonated and consider adding proof directly below it on the landing page. Carry the learning forward one change at a time, so you can tell whether it holds beyond the first comparison.
If your first-party campaigns and modeled prospecting show different results, keep both views. Your own visitors already have a relationship with the business, while URL-based audiences are useful for testing which messages can earn attention in the category. Report them with different expectations for conversion rate and time to qualification. A sound prospecting test should help you decide where to spend the next dollar, even when it cannot tell you what a competitor’s buyers did.
Frequently asked questions
Can I use a competitor’s thank-you page to target their leads?
Adding that URL cannot identify the competitor’s visitors or form submitters. Use relevant public pages as inputs to a modeled audience, and assess the audience against qualified outcomes you can observe in your own systems.
Does this work with LinkedIn lead forms?
A public landing page associated with a LinkedIn campaign can help describe a buyer category. It does not expose LinkedIn form submissions, contact details, or a list of people who saw the ad. Consider the LinkedIn campaign useful context for your research into messaging and offers.
What should I compare if I have very few opportunities?
Track qualified meetings or sales-accepted leads as interim measures if they are consistent across both variants. Keep opportunity and pipeline reporting in the plan and give cohorts time to mature. Treat an early difference as a signal to investigate, especially if just one account accounts for most of the apparent lift.
When should I use first-party remarketing instead?
Use it when the goal is to reach eligible people who interacted with your own business, such as visitors to a key product page, and you have the applicable data and controls. Use a modeled URL segment when you are exploring potential new buyers with relevant browsing behavior. Keep the campaign naming and reporting clear enough to tell the two apart.
