Impression Share Lost to Rank vs Budget: What Each One Means
The two lost IS columns point at completely different problems, and treating them the same is how you waste a day pushing the wrong lever.
The short version
- Lost IS (budget) is a funding problem; lost IS (rank) is a quality or bid problem. Never fix one with the other's lever.
- Act on budget loss above roughly 10 percent, but only uncap campaigns already inside their CPA or ROAS target.
- For rank loss, fix ad relevance and prune queries before you raise bids, since bids are the only lever that costs more per click.
- When both columns are high, pruning irrelevant search terms is the single move that helps both at once.
- Read these metrics at the ad group level: campaign averages hide the split and point you at the wrong fix.
Lost IS (budget) means your ads stopped showing because you ran out of money. Lost IS (rank) means your ads were eligible but lost the auction on Ad Rank. Budget is a funding problem you fix by raising the cap or trimming waste; rank is a quality or bid problem you fix by improving relevance, raising bids, or cutting the queries dragging you down. Pull both columns before you touch anything, because the fix for one makes the other worse.
What each column actually measures
Search impression share is the impressions you got divided by the impressions you were eligible for. The two "lost" columns split the gap into the two reasons you missed those eligible impressions. Lost IS (budget) is the share of eligible impressions you forfeited because your campaign was limited by budget at the moment the auction ran. Lost IS (rank) is the share you lost because your Ad Rank was too low to clear the threshold or beat competitors for a slot.
The numbers plus your actual IS should sum to 100 percent, which is the first sanity check. If search IS is 42 percent, lost to rank is 48 percent, and lost to budget is 10 percent, those add up and you know most of your missed volume is a ranking problem, not a money problem. Find both in the Google Ads UI by adding columns in any campaign or ad group view: Columns > Modify columns > Competitive metrics, then select Search impr. share, Search lost IS (budget), and Search lost IS (rank).
The distinction matters because the two have opposite remedies. Budget loss says the auction wanted to show you more but you couldn't pay. Rank loss says you were in the auction and came up short. You cannot solve a rank problem by raising the budget, and you cannot solve a budget problem by raising bids (that spends your cap faster and often increases budget loss).
The mechanism: why the auction splits it this way
Ad Rank is recalculated on every single query using your bid, your ad quality (expected CTR, ad relevance, landing page experience), the context of the search, and the expected impact of extensions and formats. When your Ad Rank clears the threshold for the query and ranks high enough among competitors, you're eligible to show. Lost IS (rank) aggregates all the auctions where that didn't happen.
Budget enters the picture after eligibility. Google paces your spend across the day. When the system judges you'll exhaust your daily budget, it starts skipping auctions you were otherwise eligible to enter. Those skipped-for-money auctions are what lands in lost IS (budget). This is why a campaign can show "Limited by budget" and have high lost IS (budget) even when its bids and quality are perfectly competitive.
The practical consequence: these two columns can trade off against each other. If you're budget limited and you raise bids to fix a rank problem, you win more auctions per click but burn the cap sooner, so impressions skipped for budget go up. If you're rank limited and you cut budget to control cost, you'll see lost IS (budget) climb while the underlying rank problem sits untouched. Always diagnose which constraint is binding before you pull a lever, because the levers interact.
Decision rules: budget loss
Treat lost IS (budget) above roughly 10 percent as a live signal worth acting on, and above 20 percent as a campaign that is actively leaving qualified volume on the table. Below 10 percent, pacing noise and intraday fluctuation make it not worth chasing.
Before you raise the budget, check whether the campaign is profitable at its current target. If ROAS or CPA is comfortably inside target and lost IS (budget) is high, raising the cap is close to free money: you're buying more of something that already works. Raise it in steps of 20 to 30 percent and watch for two days, because adding budget changes pacing and can briefly pull in lower-intent auctions.
The failure mode: raising budget on a campaign that's only breaking even. The impressions you were skipping are not a random sample of your good impressions. Pacing tends to skip auctions later in the day and in lower-value contexts, so the marginal impression you buy by uncapping is often worse than your average. If you uncap a borderline campaign, expect CPA to drift up, not hold flat. When budget is tight, the better first move is usually to free money by pruning waste (search terms, placements, poor geos) so your existing cap covers more of the auctions that convert.
Decision rules: rank loss
Lost IS (rank) is normal and you will never zero it out, because no advertiser wins every auction. Worry about it when it's high on your highest-intent, best-converting terms specifically, not as a campaign-wide average. A 50 percent lost IS (rank) on a broad prospecting campaign is fine; the same number on your branded or bottom-funnel exact-match terms is a problem costing you conversions.
You have three levers, in order of preference. First, improve ad quality: tighten ad relevance so the query, the ad copy, and the landing page line up, which raises Ad Rank without raising cost per click. Second, prune the queries dragging the ad group down. A tight, relevant set of search terms concentrates both budget and Ad Rank on auctions you can win; every irrelevant query you let through pulls spend and dilutes the average quality signal. Third, and last, raise bids, because it's the only lever that costs you more per click directly.
The trap: raising bids to chase rank while you're budget constrained. You'll win higher positions on fewer auctions and your lost IS (budget) will spike, so total impressions may not move at all. Fix the budget constraint first (or confirm it's not binding), then address rank. If both columns are high at once, that's your signal to prune hard: pruning is the one move that helps both, because it removes the low-value auctions eating your cap and lifts the quality signals that drive rank.
Reading the two columns together
Run the diagnosis as a quick decision tree. High budget loss, low rank loss: you're competitive but underfunded. Prune waste first, then raise the cap if the campaign is inside target. High rank loss, low budget loss: you have room to spend but aren't winning. Work ad relevance and query tightness before bids. Both high: you're losing auctions and running out of money, the worst case, and pruning irrelevant search terms is the highest-leverage first move because it attacks both at once.
Pull this at the ad group level, not just the campaign level. Campaign averages hide the split. One ad group can be 90 percent budget constrained while another in the same campaign is purely rank limited, and the campaign number will show you a muddy middle that points you nowhere. The search terms report (Campaigns > Insights and reports > Search terms) is where you find the irrelevant queries to prune once you know pruning is the right call.
Honest limits: Google does not tell you which specific auctions you lost or why a given query lost on rank versus budget. These columns are aggregates estimated across the auctions you were eligible for, and the eligibility denominator is Google's estimate, not a counted number. Treat the figures as direction and magnitude, not precision. A move from 40 to 30 percent lost IS (rank) is a real improvement; a move from 40 to 38 is inside the noise.
Questions people also ask
Why don't my lost IS numbers add up to 100 percent with my search IS?
They should, within rounding, as long as you're looking at the same date range and segment. If they don't, check that you haven't mixed a filtered view or a period where the campaign's settings changed mid-range. Changes to location targeting or ad schedule can shift the eligible-impression denominator across the window.
Can a campaign show high lost IS (budget) without saying 'Limited by budget'?
Yes. The status label is a campaign-level summary tied to recent pacing, while the lost IS (budget) column is computed over your selected date range. A campaign that was budget constrained earlier in the range can still carry high lost IS (budget) even if today it's no longer flagged as limited.
Should I raise bids or raise budget first when both lost IS columns are high?
Neither, first. Prune your search terms and tighten ad group relevance, because that's the only move that improves both rank and budget efficiency at once. Raising bids while budget constrained just spikes your lost IS (budget), and raising budget while rank constrained spends more on auctions you're still losing.
Is high lost IS (rank) always bad?
No. No advertiser wins every auction, so some rank loss is normal and healthy. It only matters on your highest-intent, best-converting terms. High rank loss on broad prospecting queries is often fine and sometimes desirable, since those are the auctions you'd lose money winning.