Google Ads Auction Insights: What Competitor Data Can Tell You
Read Google Ads Auction Insights without guessing competitor budgets, bids, or results. Learn which visibility shifts deserve a contractor's attention.
By Gavin Sevastian · Updated August 12, 2026 · 10 min read
A competitor starts appearing above you. Your agency sends an Auction Insights screenshot with one row circled, and the proposed answer is more budget.
Stop there. The screenshot has not proved the recommendation.
Google Ads Auction Insights can show how often advertisers appeared in the same eligible auctions and how their visibility compared. It cannot show a competitor's budget, bids, keywords, cost per lead, jobs, or profit. If the recommendation depends on one of those invisible numbers, the report does not support it.
I use Auction Insights as context for a real business question: did the competitive environment change enough to help explain what happened to your own useful leads? It is not a scoreboard and it is not an instruction to chase whoever sits above you.
Auction Insights is a shared-auction report
Google's current Auction Insights documentation says the report compares your performance with other advertisers participating in the same auctions. For Search campaigns, it can be viewed for eligible keywords, ad groups, or campaigns that meet Google's activity threshold.
The phrase “same auctions” carries the whole limitation. You are not seeing every contractor in the city. You are seeing domains that overlapped with the slice of eligible auctions represented by your selected account scope, date range, device mix, targeting, and campaign activity.
It is not a complete competitor list
A roofer can compete with you in the market and never appear in the selected report because you entered different auctions, used different service terms, targeted different territory, or did not meet at the same time. Another domain can appear frequently even if homeowners would not describe it as your closest commercial rival.
Google also says the report does not show when impression share is below 10%. Absence is therefore not proof that an advertiser never entered an auction. It may mean the visible threshold was not reached for that report.
Read the six Search metrics separately
The Search version currently provides six statistics: impression share, overlap rate, position above rate, top of page rate, absolute top of page rate, and outranking share. Similar names answer different questions.
Impression share is your visibility against eligibility
Google defines impression share as impressions received divided by the estimated impressions you were eligible to receive. Eligibility reflects factors such as targeting, approval status, bids, and quality. The broader impression-share documentation explains that this is an estimate based on the auctions in which the system considers the ad competitive.
Your impression share is not market share. It does not say what percentage of local homeowners chose you, what percentage of all searches existed, or what percentage of jobs you won.
Overlap rate describes meetings
Overlap rate tells you how often another advertiser received an impression when your ad also received one. A high overlap rate means you met often in the visible auction sample. It does not say which business won the customer.
The denominator matters. A domain could overlap frequently with a narrow campaign while appearing rarely across your broader account. Always name which campaign, ad group, keyword group, date, and device context produced the report.
Position above rate describes order when both showed
Position above rate applies when your ad and the other advertiser both received an impression. It reports how often the other advertiser's ad showed in a higher position.
That sounds like a direct ranking contest, but the commercial result remains unknown. The higher ad may have received no click. The click may have become a poor inquiry. Your lower ad may have produced a booked job. Position is evidence about presentation, not revenue.
Top and absolute top are not the same
Top of page rate describes impressions shown anywhere above the unpaid search results. Absolute top of page rate describes the first ad position above those results. Neither is a promise of attention, a click, or a lead.
These rates can help explain a visibility change, especially when read beside your own impression share and outcome trend. They do not establish that the highest possible position is the most profitable place to buy every click.
Outranking share combines two conditions
Outranking share tells you how often your ad ranked higher than another advertiser or showed when theirs did not. It is not the inverse of position above rate because it includes auctions where only your ad appeared.
That mixed definition is why a single percentage should not become a “win rate.” It describes relative ad appearance across shared and unshared impressions within the report's eligible scope.
| Metric | Useful reading | Unsupported leap |
|---|---|---|
| Impression share | How often you showed within estimated eligible opportunity | Your share of customers or jobs |
| Overlap rate | How often another advertiser showed when you did | How much that advertiser spent |
| Position above rate | Who appeared higher when both ads showed | Who won the lead |
| Top of page rate | How often ads appeared above organic results | The most profitable position |
| Absolute top rate | How often an ad held the first paid position | Proof of the best bid |
| Outranking share | How often you ranked higher or showed alone | A competitive win rate |
A row does not reveal competitor economics
Auction Insights does not publish another advertiser's daily budget, actual bid, cost, keyword list, conversion tracking, cost per lead, close rate, job value, or margin. It does not show whether the other business is happy with the result.
Google's Ad Rank explanation names multiple auction-time factors, including bids, ad and landing-page quality, thresholds, competition, search context, and expected impact of assets. That makes “they are above us, so they bid more” an inference, not a report fact.
You cannot reverse-engineer their account from your report
A visibility increase can follow many changes: a different bid, better relevance, a new asset, wider coverage, more eligible budget, a changed landing page, altered schedule, or a different mix of auctions. Your report does not isolate which one occurred in somebody else's account.
The same caution applies to a competitor disappearing. They may have paused, exhausted a budget, narrowed the location, changed hours, lost eligibility, or simply fallen below the visible threshold in your selected slice. Auction Insights cannot choose among those explanations.
Directories and Google domains may appear
You may expect a list of local contractor names and see a lead marketplace, directory, manufacturer, franchise, or aggregator instead. That is not necessarily an error. Those advertisers can enter the same service auctions while selling a different customer path.
Google also notes that google.com can appear in some circumstances involving Google-hosted domains, Google Business Profiles, or a competitor destination hosted on a Google property. Do not turn that row into a story about Google secretly competing for your customer without reading the documented explanation.
Different business models can share one auction
A directory may value the click across several contractors. A franchise may value a lead across a wider region. A local owner may need the job to fit one crew and one schedule. The fact that all three entered the auction does not make their economics comparable.
This is why blindly matching a high-visibility advertiser can hurt a local business. You may be chasing somebody whose sale, territory, funding, or lead model is completely different from yours.
Scope changes the story
An account-level view, one campaign, a set of keywords, one week, and one device can produce different competitive pictures. None is automatically the true one. Each answers a different question.
If roofing replacement leads declined, inspect a scope that represents that service and period. A whole-account report blending brand, repair, replacement, and another territory can hide the relevant movement. Narrow it so the question and report belong together, then keep enough activity that one odd day does not drive the conclusion.
Search Partners have a reporting limit
Google states that Search Partners do not distinguish top positions from other positions and are not included in Auction Insights data. If your campaign runs across the wider Search Network, the competitive report and the network report do not describe exactly the same surface.
The Search Partners guide covers that boundary. Do not fill the missing site detail with a guessed competitor explanation.
Performance Max adds another scope boundary
Google currently makes Auction Insights available for Search, Shopping, and Performance Max campaigns. For Performance Max, the report covers Search and Shopping ad impressions on the Search Network and is segmented accordingly. It is not a competitor view across every PMax channel.
That matters because Performance Max can serve across a broader mix of inventory. The Auction Insights row may help explain competitive visibility on its Search-facing slice, but it cannot explain YouTube, Display, Discover, Gmail, Maps, or every conversion attributed to the campaign.
Keep the campaign-type question separate
If you are deciding whether broader automation belongs beside Search, use the Performance Max versus Search guide. Auction Insights can add competitive context. It cannot settle which campaign type produced incremental qualified work.
Pair competitor visibility with your own outcomes
A competitor trend deserves attention when something useful changed for your business at the same time. Put qualified calls, relevant forms, estimates, booked work, service mix, geography, and spend beside the report.
If impression share fell while qualified demand and booked work held, the competitive movement may not require action. If useful leads fell, cost rose, and the selected report shows a sustained visibility shift in the same service and territory, investigate further. The report has helped narrow the question. It still has not prescribed the fix.
Platform movement and business movement are different
Your ad can lose absolute top position without losing useful customers. It can gain impression share by entering more marginal auctions. It can outrank a directory more often while producing fewer booked jobs.
The Google Ads reporting guide for contractors explains why the media event and business outcome must remain distinct. Auction Insights belongs on the media side of that bridge.
Use ignore, investigate, or change
Ignore a one-off movement that has no matching business consequence. A competitor appearing above you once is not an emergency.
Investigate when the shift is sustained, relevant to the service and area, and accompanied by a change in your own outcomes. Check your eligibility, budget coverage, query mix, ads, assets, page, and conversion evidence before writing a competitor story.
Change something only when the evidence identifies a lever you control and a consequence worth addressing. That might be a budget decision, but it might also be a service-message mismatch, weak landing page, limited schedule, poor measurement, or traffic that should never have been bought.
Public ad research is a complement, not proof
Google's Ads Transparency Center can show ads from verified advertisers across Google products. It may help you understand public messaging. It still does not expose the competitor's bids, targeting, full campaign structure, conversions, or profit.
Use public ads to see what customers may encounter. Use Auction Insights to see relative presence in shared eligible auctions. Use your own business records to decide whether either observation matters.
Ask the report to answer one question
Do not request “a competitor analysis” and accept a colourful table. Ask which service, campaign, territory, period, device, and outcome the report is meant to explain.
The best conclusion may be that a competitor became more visible and nothing worth changing happened to your business. It may be that your own eligibility or rank weakened. It may be that demand changed while the competitors did not. The report earns its place when it separates those possibilities rather than turning every row into a threat.
Auction Insights tells you who shared part of the auction and how the ads appeared relative to one another. Your calls, estimates, booked work, and margins tell you whether you have a competitive problem worth solving.
Common questions
It is a report comparing your ad visibility with other advertisers that participated in the same eligible auctions. For Search, it includes impression share, overlap rate, position above rate, top and absolute top rates, and outranking share.
No. It does not reveal another advertiser's budget, bids, spend, keywords, conversions, lead cost, jobs, or profit. A visibility row cannot support those claims.
No. Impression share is your impressions divided by estimated eligible impressions within Google Ads. It is not your share of local searches, customers, revenue, or completed jobs.
You may not have entered enough of the same auctions in the selected scope, or the advertiser may be below Google's visible threshold. Missing from one report does not prove the business never advertised.
Google says this can occur with Google-hosted domains, Google Business Profiles, or competitor landing pages hosted on a Google property. Read the documented context before treating it as an ordinary competitor row.
Google says Search Partners do not distinguish top positions from other positions and are not included in Auction Insights data. Network reporting and the auction report therefore have different boundaries.
Yes, where the campaign meets Google's requirements. The PMax report covers Search and Shopping ad impressions on the Search Network, not every channel where Performance Max can serve.
Not automatically. Check whether the shift is sustained and whether qualified calls, estimates, or booked work changed in the same service and territory. Then identify which controllable issue the evidence supports changing.
