Impression Share and Auction Insights in Google Ads 2026: Where Your Reach Is Lost

“Traffic is down, let’s raise bids” is the most expensive conclusion in paid search, and it is usually reached by guessing. Sometimes bids are irrelevant: the campaign hits its daily cap and stops entering auctions after 2 p.m. Sometimes the opposite — budget goes unspent because the ad keeps losing on rank. These two conditions require opposite fixes, and one metric tells them apart: impression share.

You can tell them apart in two minutes if you know which columns to enable and how to read them. Here is the full breakdown: impression share, lost IS to budget and to rank, the Auction Insights report, and six real diagnostic scenarios with concrete actions.

What impression share is and how it is calculated

Impression share (IS) is the ratio of impressions you received to the estimated impressions you were eligible to receive under your current targeting.

IS = impressions received / eligible impressions

Google estimates the denominator itself, factoring in targeting, settings, approval status and quality. That matters: impression share is an estimated metric, not an exact count. It is suppressed at low volume and it cannot be summed across campaigns by simple addition.

The family has several members, and confusing them is costly:

Metric	What it shows	What an increase means
Search impression share	Share of available demand you captured	You are present in more auctions
Search lost IS (budget)	What you missed because money ran out	Budget is the bottleneck
Search lost IS (rank)	What you missed because you lost the auction	Ad Rank problem: bid, quality, assets
Search top IS	How often you appear above organic results	Positional visibility improved
Search absolute top IS	How often you are the very first ad	Maximum prominence, usually maximum price
Search exact match IS	Share on queries that exactly match your keywords	Coverage of your core semantics
Click share	Your share of all clicks available in your category	Not just visibility, but appeal

A quick sanity check: impressions received + lost to budget + lost to rank ≈ 100%. If your sum is far off, you are almost certainly mixing network types or looking at too short a window.

Enabling the right columns

These metrics are hidden by default. In the interface:

1. Open Campaigns (or Ad groups, or Keywords — the metrics exist at several levels).
2. Click Columns → Modify columns.
3. Expand the “Competitive metrics” block.
4. Add: search impression share, lost IS (budget), lost IS (rank), top IS, absolute top IS, click share.
5. Save the column set under a name — for example “Reach diagnostics” — so it is one click away.

A habit worth forming: save two column sets, “Reach diagnostics” and “Money diagnostics” (conversions, cost per conversion, value, ROAS). Toggling between them makes troubleshooting dramatically faster. If you report outside the interface, these same fields pull cleanly via the API into a weekly view — see media buyer dashboards and reporting.

The diagnostic tree

First rule: read lost IS to budget before lost IS to rank. They are different diseases, and treating them in the wrong order means polishing ad quality in a campaign that simply switches off at 2 p.m.

Lost IS (budget) above 10%

The campaign regularly hits its daily cap. Before adding money, check the economics: what are this campaign’s CPA and ROAS relative to target?

• Target beaten with room to spare. Raise budget in 20–30% steps every three to four days and watch whether efficiency holds. Doubling overnight restarts bid strategy learning.
• Target barely met. Reallocate first — pull budget from campaigns with heavy rank losses and weak economics.
• Target missed. Budget is not the issue. Budget-capping a losing campaign is actually protecting you.

How much you can genuinely afford to pay comes from arithmetic, not intuition — the method is in media buying unit economics.

Lost IS (rank) above 30%

You enter auctions and lose them. Ad Rank combines your bid, expected CTR, ad relevance, landing page experience, auction context, and expected asset impact. The bid is one of six inputs — and typically the most expensive one to move.

Work from cheap to expensive:

1. Assets and extensions. The fastest rank gain that costs nothing. An ad group with no callouts, sitelinks or structured snippets is giving away rank for free. Mechanics: ad extensions and assets in Google Ads.
2. Relevance. Does the keyword appear in a headline? Does the ad’s promise match what the page delivers?
3. Landing page. Speed, query match, no aggressive interstitials. A slow page hurts twice — lower rank and lower conversion rate.
4. Bid and strategy. Only after the first three. A 15–20% target CPA increase is a reasonable test step; doubling it is a way to find out how much you can lose in a week.

High impression share, few conversions

The most underrated signal in the account. If you already capture 80% of eligible impressions, there is no more room to grow sideways — the constraint is further down the funnel: ad, page, offer, price. Raising bids here buys nothing. The upside is that you can see your ceiling immediately instead of spending a month trying to buy traffic that does not exist.

Rank losses rising, CPC rising, impressions falling

The signature of a new entrant in your auctions. Confirm it in sixty seconds with Auction Insights.

Auction Insights: who you are actually bidding against

Open it at campaign, ad group, or keyword level: select the row → Auction insights. It lists domains that competed in the same auctions and six metrics for each.

Metric	Meaning	How to use it
Impression share	Share of auctions where the competitor appeared	Scale of their presence
Overlap rate	How often you both appeared together	Who your real competitor is, not the imagined one
Position above rate	How often they ranked above you when both showed	Direct measure of auction pressure
Top of page rate	How often they appear above organic	How aggressively they bid
Absolute top of page rate	How often they hold position one	Who is buying premium placement
Outranking share	How often you ranked above them	Your trend against a specific rival

Reading it properly:

• Look at trend, not snapshot. Compare the last two weeks against the two before. A single number says nothing; a change says everything.
• Segment it. The report segments by time, device and network — which reveals that a rival is pressuring you only on mobile, only on weekdays.
• Do not chase the leader blindly. A domain with 90% impression share may be a marketplace or aggregator with completely different economics. Your goal is not to beat them; it is to stop overpaying in auctions they structurally win.
• Check overlap rate. Below 10% overlap, that player barely affects you regardless of how large their share looks.

For the wider competitive picture beyond auction data, see PPC competitor analysis.

Six scenarios from real accounts

Scenario 1 — Lost IS (budget) 45%, ROAS 1.5× target

What is happening: a profitable campaign hitting a ceiling.
Action: raise budget +25% every three to four days until ROAS approaches target. Do not change the bid strategy simultaneously or you will not know what worked.

Scenario 2 — Lost IS (budget) 0%, budget underspent, impressions low

What is happening: demand is constrained by targeting, or ads are not entering auctions.
Action: check ad and keyword statuses (low search volume, disapprovals), widen geo and schedule, revisit match types. Where the sensible expansion line sits now: keyword match types.

Scenario 3 — IS dropped from 60% to 35% in a week, CPC up 20%

What is happening: the auction composition changed.
Action: open Auction Insights, compare against the prior period, identify the new or newly aggressive domain. Then decide on economics — chasing on bid only makes sense if CPA headroom exists.

Scenario 4 — Absolute top IS 15%, click share low

What is happening: you are showing at the bottom, where CTR is structurally lower.
Action: strengthen assets and relevance; if the strategy allows, use impression-share targets deliberately and on a limited set of campaigns. Full comparison: Google Ads bidding strategies.

Scenario 5 — Exact match IS 90%, overall IS 40%

What is happening: your core semantics are fully bought; growth can only come from adjacent queries.
Action: expand deliberately with a tight negative fence. Method: negative keywords and search terms.

Scenario 6 — Lost IS (rank) 60% despite a high bid

What is happening: a quality problem, not a money problem.
Action: walk the chain — assets, relevance, landing page. If rank still will not move, you are probably trying to buy an auction that is structurally not yours, and the budget belongs somewhere you can compete.

How much traffic you can actually buy: calculating the ceiling

Impression share is a planning tool as well as a diagnostic one. It answers a question usually settled by intuition: “if I double the budget, how many more conversions do I get?”

The arithmetic is straightforward. Say the last 30 days produced:

• 60,000 impressions at 40% impression share;
• 35% lost to budget, 25% lost to rank;
• 4% CTR, 3% click-to-conversion rate, $0.80 average CPC.

Available auction volume = 60,000 / 0.4 = 150,000 impressions. Of those, 35% is lost to budget — roughly 52,500 impressions. Capturing all of them at the same CTR and conversion rate would add about 2,100 clicks and roughly 63 conversions, at a cost of about $1,680.

Compare that implied CPA (around $27) with your allowable threshold. Below it, raise the budget — and you now know to what number. Above it, the money belongs in a different campaign.

Three caveats, without which the model lies to you:

1. CTR and conversion rate are lower on incremental volume. You capture the best demand first. Discount by at least 15–25%.
2. CPC is higher on incremental volume. The auctions you currently lose or skip are more expensive on average.
3. Rank losses are not bought this simply. You can model them the same way, but only after quality work — otherwise you just pay more for the same volume.

Even with those corrections, this beats “let’s throw in another couple of thousand and see”. More importantly, it is expressed in conversions and dollars — the language a client or a finance director actually speaks.

Going deeper: levels and segments

A campaign is too coarse a unit to act on. Three levels of detail, walked in order:

Level	What it reveals	Decision it drives
Campaign	Overall picture, weekly priorities	Where to move budget between campaigns
Ad group	Which theme underperforms inside a campaign	Where to rewrite ads and strengthen assets
Keyword	Pinpoint losses on high-value terms	Where to change match type or bid
Segment: device	Mobile versus desktop	Adjustments and mobile page priority
Segment: time	Hours and days where money runs out	Ad scheduling instead of a bigger budget
Segment: geo	Regions with different competitive pressure	Splitting campaigns by geo, or bid adjustments

A frequent finding when segmenting by time: that 30% budget loss turns out to be concentrated in two or three evening hours. The fix is redistribution through ad scheduling, not more money — and it is free.

Limitations worth remembering

• It is an estimate. Google computes the denominator and you cannot audit it. Use IS for direction, not for precise math.
• It is suppressed at low volume. New campaigns and rare queries show a dash. That is normal; accumulate data.
• Networks are separate. Search and Display impression share are calculated independently and do not add up.
• Limited in automated campaign types. Performance Max and Demand Gen do not expose the familiar search impression share; diagnostics there rely on other signals — see our Performance Max guide.
• High impression share is not profit. 100% IS on an unprofitable query is a fully funded loss. The metric answers “where am I present”, not “where should I be”.

Building it into a weekly routine

Fifteen minutes a week, same order every time:

1. Load the “Reach diagnostics” column set, last 7 days versus previous 7.
2. Sort by lost IS (budget). Anything above 10% goes on the economics review list.
3. Sort by lost IS (rank). Anything above 30% with healthy economics goes on the rank work list.
4. Open Auction Insights for two or three priority campaigns and compare to last week.
5. Write down conclusions and make exactly one change per campaign. Not five — one, or next week you will not know what moved the number.

These metrics also belong in your regular account audit as the “reach and competition” section — the full checklist is in the Google Ads account audit checklist. Note too that as automated query matching expands, auction composition shifts faster than it used to — context in our breakdown of AI Max for Search campaigns. And if the conversion numbers in your account do not reconcile with real sales, all this reach diagnosis rests on sand — see server-side tagging with sGTM.

One more case worth naming: sometimes impression share is capped by neither budget nor rank, but by the account itself — spend limits, missing history, suspensions. No column configuration fixes that. Agency Google Ads accounts with real spend history address that constraint, and the wider PPC Rebels service overview covers what else sits around it.

FAQ: impression share and Auction Insights

What counts as a good impression share?

There is no universal number. For brand campaigns, 80–95% is normal — handing your own brand to competitors is expensive. For generic commercial terms, 40–60% while hitting your CPA target is healthy. Chasing 100% on non-brand terms is almost always unprofitable: the last few percent of reach cost disproportionately more.

Why do the three metrics not sum to exactly 100%?

Rounding, the estimated denominator, and mixed network types in the selection. A couple of percentage points is normal; a gap of tens of percent means Search and Display data are mixed.

Can I see competitor names?

Auction Insights shows participating domains. It is not an exhaustive list of every advertiser — Google surfaces those who overlapped with you often enough to be meaningful.

Is Auction Insights available for Performance Max?

Not in the familiar search impression share form. PMax diagnostics rely on other reports — asset groups, search themes and channel-level results.

Why do I see a dash instead of numbers?

Not enough data in the selected window. Extend to 14–30 days, or view the metric at campaign level rather than keyword level.

How can I raise impression share without raising bids?

Three levers, in order of effectiveness: add and improve assets, increase ad relevance to the query, and clean up and speed up the landing page. All three influence Ad Rank directly and cost nothing per click.

Should I use the Target Impression Share bid strategy?

Only where visibility matters more than efficiency — usually brand or brand-defense campaigns. On performance campaigns it buys position at any price and ignores conversions.

Which matters more: impression share or click share?

Click share is more honest: it accounts not just for being shown, but for being chosen. A wide gap between high impression share and low click share points straight at a weak ad or a poor position.

How do I spot a competitor bidding aggressively?

The tell is a rising position-above rate for them alongside your rising CPC and falling impression share on the same demand. If your conversion rate has not changed, you are simply paying more for the same result.

Does impression share affect Quality Score?

Not directly. There is an indirect link through CTR: impressions in lower positions earn lower CTR, and expected CTR is a Quality Score component. Sustained low placement gradually erodes quality.

How do I view impression share by time of day?

Segment the report by hour of day or day of week. It often turns out budget losses concentrate in specific hours — in which case the fix is ad scheduling, not a bigger budget.

Impression share fell but competitors look unchanged — now what?

Check your own side: disapproved ads, keyword statuses, landing page changes, paused ad groups, budget exhaustion on specific days. In most cases the cause is inside the account, not outside it.