Search IS
How much of the search demand you actually turn up for — and how much is left.
What it means
Search IS is the share of the Search auctions your keywords were eligible for that your ads actually appeared in, shown as a percentage. It's the Search network only — Shopping and Display coverage sit outside it — and it counts auctions, not shoppers. It measures coverage, never quality.
Show the math
Formula and a worked example
Eligible impressions is Google's estimate of the auctions your keywords, targeting, schedule and approval status qualified for. Change the geo or the ad schedule and the denominator moves with it, so the percentage shifts even when your ads don't.
Search only. An account leaning on Performance Max can read low here while it's buying plenty of traffic elsewhere.
Worked example. Your keywords were eligible for 100,000 Search impressions in March and you appeared in 32,000, so Search IS is 32%. At a 5% click rate and a 1.6% conversion rate, the missing 68,000 impressions work out at roughly 3,400 clicks and 54 conversions you never had the chance to win.
Google splits the loss two ways: share lost to budget and share lost to rank. The first is fixed with money, the second with bids, relevance and landing pages. Confusing them is how budget gets poured into auctions you still won't win.
It answers the question
Is there more search demand to buy? A low Search IS means the auctions are running without you; a high one means you've taken most of what's there and growth has to come from somewhere else.
Why it matters
It's one of the few figures in the account that describes what you didn't do. Everything else measures the traffic you bought; this measures the traffic that ran without you, which makes it the honest test of whether more budget would find anything.
It also stops you scaling into a wall. Once coverage is already high, extra money mostly buys broader, colder matches and Cost/Conv.Cost/Conv.What each Google sale costs you. gets worse — which reads like a broken campaign when really the market ran out.
What good looks like
There's no standard percentage, and 100% is the wrong target — the last slice of any auction is the dearest and usually costs more than the sales are worth. Judge it against your own trend and your own economics: high coverage on the terms that convert, low coverage on the ones that don't. Compare within a category too, since a branded term can sit at 90% while a broad one is expensive at 20%.
How to improve it
| Lever | What you do | Expect | How long | Watch out for |
|---|---|---|---|---|
| Fast Feed the campaigns that run out of money | Raise daily budgets where the loss is to budget rather than to rank | Search IS up within days | 3–7 days | Cost rises immediately, and the impressions you gain are the ones you were outbid on before, so Cost/Conv. gets worse before it settles. |
| Fast Stop being eligible for what you don't want | Add negatives and tighten geo and schedule so the pool only holds auctions worth winning | Search IS up with no extra spend | 1 week | You improved the ratio by shrinking the pool, not by winning more. Clicks and conversions fall, and one careless negative can exclude a term you needed. |
| Slow Raise the bid where the sale still pays | Lift targets on campaigns where the cost per sale sits comfortably under contribution per order | Search IS up 10–20 points | 2–4 weeks | Avg. CPC rises across the whole campaign, not only on the auctions you gained, so sales get dearer even where you were already winning. |
| Slow Win on rank instead of on money | Improve ad relevance, assets and landing page experience so you qualify at a lower bid | Search IS up at a flat Avg. CPC | 1 quarter | Slow, and it takes copy and site work that never shows up in the ad account. Tighter ads also match fewer queries, which pulls the other way. |
Every lever costs something somewhere. The last column is the one to read twice.
Read it with
Coverage tells you the room you have. The price of a sale tells you whether to use it.
Winning more and paying less
You took more of the auction and each sale still got cheaper. Rare, and worth spending into while it lasts.
Efficient but small
Cheap sales and thin coverage. There's demand you're not turning up for, and the economics say you can afford it.
You bought the coverage
Expected when you scale. The question isn't whether it got dearer, it's whether it's still under the ceiling.
Losing ground and paying more
Someone else is taking auctions you used to win, and outbidding them is the most expensive answer available.
One counts the times you appeared, the other the times you were picked. A 60% Search IS next to a 20% Click Share means you're on the page and being scrolled past, which no budget fixes. When the two rise and fall together, coverage is the real constraint.
Together they tell you whether the budget is the ceiling. A high Cost at a low Search IS means you're spending hard and still missing most auctions, so the bids are too low or the pool too broad. A high Cost at a high Search IS means you've bought most of what's available, and the next euro goes into worse traffic.
Common misreads
You're missing 60% of the auctions your keywords were eligible for, and a great many of those searches were never going to buy anything. Impressions aren't customers.
Only if the loss is to budget. When it's lost to rank, extra money buys nothing — you get outbid at exactly the same rate with a bigger cheque.
It's the Search network only. Shopping and Display coverage sit elsewhere, so an account can look absent here and still be busy.
Also called
Search impression share · impression share · Search Impr. share
See yoursYour Search IS by campaign, next to the cost per sale that decides whether to spend into it.
Open Google Ads →