Most advertisers can tell you how many clicks they got last month. Very few can tell you how many they missed.
Impression share is the number that answers that, and it changes the question from "how did we do" to "how much of this market did we actually show up for." Those are different questions, and only one of them tells you whether there is room to grow.
Impression Share Is Market Share
The definition is simple. Impression share is the impressions you received divided by the impressions you were eligible for.
If your ad could have appeared 1,000 times for the searches you target and it appeared 400 times, your impression share is 40%. The other 60% happened. Somebody searched, an ad appeared, and it was not yours.
That is what makes it different from every other metric in the account. Clicks, conversions and cost all describe what you did. Impression share describes what you did relative to what was available, which is the only way to know whether a good month was actually a good month.
There Are Three Numbers, Not One
This is where most people stop too early, because the headline figure is the least useful of the three.
Search impression share is how often you appeared at all, anywhere on the page. Appearing at the bottom counts.
Search top impression share is how often you appeared in the top positions above the organic results. This is the one we watch, because that is where the clicks and the intent are.
Search absolute top impression share is how often you were the very first ad. Useful, and rarely the thing to optimize for.
A healthy account looks something like the steel manufacturer we manage: top impression share at 85%, absolute top at 63%, click share at 68%, with very little lost to budget and not much lost to rank. That is an account owning its market on the searches that matter.
An account at 40% overall might be at 15% in the top positions, which means it is technically present and practically invisible. Those impressions are real, they are just at the bottom of the page, below the map pack and the organic results, at a point in the scroll where most people have already made a decision.
That gap between the headline number and the top number is worth calculating on your own account before you do anything else. If they are close, your positions are healthy and the question is volume. If the top number is a fraction of the headline one, you are paying to be present in a place that does not get looked at, which is a bid and quality score problem rather than a budget one.
Why We Aim at 80%, Not 100%
The instinct is to push toward 100%, and that is usually a mistake.
Impression share is bought. Getting from 80% to 95% means bidding into auctions you were previously losing on purpose, and those are the auctions where somebody else values the click more than you do. The last slice of impression share is the most expensive slice, and it is rarely worth what it costs.
Around 80% top impression share on the keywords that genuinely matter is the target we work to. High enough that you are present for nearly every search worth having, low enough that you are not overpaying for the marginal ones.
And it applies to specific keywords, not the account. Account-wide impression share is an average across terms with wildly different value, which makes it nearly meaningless. The number is useful at the keyword or ad group level, on the terms you have already decided are worth owning.
Where It Went: Budget or Rank
When you are not getting the impressions, Google tells you why, and the two answers lead to completely different fixes.
Impression share lost to budget means you were eligible, you would have won, and the campaign had run out of money. That is the arithmetic behind a budget gone by noon. The fix is either more budget or a tighter net so the budget lasts.
Impression share lost to ad rank means you were in the auction and lost it. Either your bid was too low, or your quality score was, and since ad rank is the two multiplied together, the cheaper fix is usually the quality score side rather than the bid.
One practical note. Lost to rank is available right down to the keyword, but lost to budget is reported at campaign level only, which makes sense once you think about it, since the budget belongs to the campaign rather than to any one keyword. So read lost to budget on the campaign row first, and if it is near zero you know every gap further down is a rank problem.
Between them the two account for everything you did not get. A campaign losing 50% to budget and 5% to rank has a completely different problem from one losing 5% to budget and 50% to rank, and treating them the same is how people end up raising bids on a campaign that simply ran out of money.
Auction Insights: Who You Are Actually Against
The companion report is Auction Insights, and it answers the question every owner asks eventually: who is beating me?
It shows the other advertisers appearing in the same auctions, and for each one their impression share, their overlap rate with you, and how often they outranked you. You can run it at the campaign, ad group or keyword level, and running it at the keyword level on your most important term is far more revealing than running it on the whole account.
It is also the fastest way to diagnose a sudden drop. When a therapist we spoke to saw new client volume fall off a cliff, the first two things worth checking were whether conversion tracking had broken and whether a competitor had moved. Auction Insights answers the second in about a minute, and in her case the suspect was a competitor two buildings away.
Run it for the period before the drop and the period after, and compare. A new name near the top, or an existing one whose impression share climbed, tells you what happened without any guesswork.
Two columns are worth more than the rest. Overlap rate is how often the other advertiser appeared in an auction you also appeared in, which tells you who you genuinely compete with rather than who merely exists. And position above rate is how often they were placed higher when you both showed, which turns a vague sense of being beaten into a number.
Too Broad to See Your Own Competitors
Here is the failure that surprises people, and it connects this chapter back to structure.
Auction Insights needs enough concentrated data to report on. If your targeting is so loose that your impressions are spread across thousands of unrelated queries, Google does not have a coherent auction to describe, and the report comes back empty or useless.
We audited a rehabilitation equipment manufacturer whose search impression share was under 10%. Auction Insights showed them nothing at all. Not because they had no competitors, but because their targeting was so undefined that there was no consistent auction they were part of.
That is worth stating plainly: a very low impression share is not only lost traffic. It also removes your ability to see the competitive picture, which is the information you would need to fix it.
It Is a Diagnostic, Not a Goal
Three cautions before you go and chase the number.
Impression share says nothing about intent. It is purely about page position. You can hold 90% top impression share on a keyword nobody valuable is searching, which is an expensive way to be visible.
Low volume makes it unreliable. A keyword with a handful of eligible impressions produces a percentage built on almost nothing.
Raising it is not automatically good. Buying more impression share on a term that converts is growth. Buying it on a term that does not is just spending more.
The right sequence is to decide which searches are worth owning first, then use impression share to check whether you own them. Doing it the other way round, finding a low number and bidding it up, is how accounts get more expensive without getting better.
The Twenty Minute Version
- Add the columns. Search impression share, top impression share, absolute top, plus lost to rank. None are shown by default.
- Read lost to budget on the campaign row. It is only reported at campaign level, and it tells you whether anything below is worth diagnosing as a rank problem.
- Then work at the keyword level, not the account level, on the ten or so terms you have already decided matter.
- Read top impression share, not the headline figure. Present at the bottom of the page is close to not present.
- Aim at roughly 80% on those terms. Not 100%.
- Check where the rest went. Lost to budget is a budget or match type problem. Lost to rank is a bid or quality score problem.
- Run Auction Insights on your single most valuable keyword. If it returns nothing useful, your targeting is too broad to be in a coherent auction, and that is the finding.
- Compare two periods whenever performance moves. It is the fastest way to see whether a competitor changed something.
Impression share is the closest thing Google Ads gives you to a market share figure, and most accounts have never had the columns switched on, which means they have no idea whether they are winning a big share of a small market or a tiny share of a large one.
We check it on the keywords that matter in every audit, because it is the number that tells you whether an account has a performance problem or simply a presence one. Hand that over to us and we will run it. Or take five minutes now, switch on the columns, and look at the single keyword you care about most. That one row usually says more than the rest of the account put together.
