Impression share
The proportion of available auctions an ad actually appeared in.
Impressions received divided by impressions the account was eligible for. The gap is reported split between share lost to budget and share lost to rank, which tells you whether the constraint is money or competitiveness.
Impression share = impressions ÷ eligible impressions
Key takeaways
- The split between share lost to budget and share lost to rank is the useful part, not the headline percentage.
- Lost to budget means the ads could win and the money ran out. Lost to rank means more money changes nothing.
- Eligible impressions are an estimate, so treat impression share as a reliable trend and a soft absolute.
- Chasing 100% buys the least relevant slice of the auction and raises average cost per result.
- The denominator moves when targeting changes, so a fall can be arithmetic rather than performance.
The split is the metric, not the percentage
Impression share on its own says you appeared in 43% of available auctions, which is a fact with no action attached. The two loss figures underneath it turn the same data into a decision.
Share lost to budget means your ads were competitive enough to win those auctions and the campaign had stopped spending by the time they happened. The constraint is money, and adding it will buy impressions at roughly the cost you are already paying.
Share lost to rank means you were in the auction and did not win it. Adding budget changes nothing, because budget was never what stopped you. The constraint is Ad Rank, which is bid and quality together.
The two ask for opposite responses, and conflating them is how an agency spends a budget increase on a problem the increase cannot solve.
What "eligible impressions" actually counts
The denominator is Google estimate of how many auctions your ads could have entered, built from auction data, your targeting, and quality signals. It is not a directly observed count of searches, and it cannot be, because Google is estimating demand you never participated in.
This makes impression share reliable as a trend and softer as an absolute. Moving from 40% to 55% over a month is meaningful. Arguing that 55% is objectively the right figure is arguing about an estimate.
It also means the denominator moves independently of anything you do. Widen targeting, add locations, or extend an ad schedule and the pool of eligible auctions grows, so impression share falls even though you appeared in more of them than before. A drop straight after a targeting change is usually arithmetic.
The cleanest budget argument an agency has
Share lost to budget quantifies demand the client is currently paying nothing to reach, in the same units as everything else in the account. It converts a request for more money from an opinion into an estimate: at this cost per conversion, this share represents roughly this many additional conversions available.
That estimate holds better at the margin than further out. The auctions you lose to budget are not uniformly as valuable as the ones you win, since budget pacing tends to spend earliest in the day and on the most available inventory. Expect the incremental conversions to cost somewhat more than the current average, and say so rather than promising a linear scale-up.
The corresponding honest move is to raise share lost to rank when it is high. It says the account has a relevance problem that money will not fix, and volunteering that before a budget conversation is what makes the budget conversation credible.
Why chasing 100% is usually wrong
The auctions you are not winning are not a random sample of the ones you are. They skew toward searches where your ad was less relevant, your bid was less competitive, or the intent was further from a purchase, which is precisely why you lost them.
Buying that last slice therefore costs more per conversion than the account average, and it does so at the exact point where every incremental conversion is already the most expensive one. Pushing impression share from 80% to 95% reliably raises blended CPA.
There are cases where it is right anyway. Defensive brand coverage, a product launch where presence matters more than efficiency, or a category with so little volume that missing an auction means missing the only customer that week. All of those are strategic decisions the client should make knowingly rather than efficiency targets an agency should default to.
What it means for an agency
Impression share lost to budget is the cleanest argument for a budget increase there is: it quantifies demand the client is currently paying nothing to reach. Bring the number, not the opinion.
Where it goes wrong
- Reading lost-to-rank as a bidding problem when it is a relevance problem, and raising bids into a poor Quality Score.
- Quoting impression share for a campaign whose targeting was recently widened. The denominator moved, so the fall is arithmetic, not performance.
- Promising that a budget increase converts share lost to budget into conversions at the current CPA. The marginal auctions cost more.
In short
Impression share is one number wrapped around an estimate, and its value is almost entirely in the two loss figures underneath it. Use lost to budget as evidence when asking for money, raise lost to rank yourself before somebody else does, and treat the last twenty per cent of any auction as something to buy deliberately rather than a target to hit.
Frequently asked questions
What is a good impression share?
There is no target figure, because the right one depends on whether the remaining auctions are worth winning. A brand campaign defending its own name might reasonably aim high; a broad generic campaign at 90% is probably buying low-intent traffic at a premium. Read the two loss figures instead, since those say whether the gap is worth closing and what it would take.
What is the difference between lost to budget and lost to rank?
Lost to budget means your ads were eligible and competitive but the campaign had run out of money before those auctions happened, so adding budget buys them. Lost to rank means you entered and did not win, which is a bid and quality problem that budget cannot solve. They call for opposite responses, and spending a budget increase on a rank problem is a common and expensive mistake.
Why did impression share fall when I did not change anything?
Check whether the denominator moved. Widening targeting, adding locations or extending an ad schedule all increase the pool of eligible auctions, so the percentage falls even though you appeared in more auctions than before. Otherwise, a new competitor entering the auction raises the bar and lowers your share without anything in your account changing.
Is impression share available on Meta?
Not in the same form. Impression share is a Google Ads concept resting on the idea of a countable set of eligible auctions for a keyword. On Meta the nearest equivalents are reach against the estimated audience size and delivery diagnostics such as auction overlap, which answer a similar question about whether you are constrained by budget or by competitiveness.
Related terms
Sources
Last reviewed
Keplent keeps track of these for you.
It holds what each client agreed to, watches what their campaigns actually did, and tells you in plain sentences what needs you this week.
Join the waitlist