When fewer people search for your category every year, the account that keeps adding keywords is buying lower intent to hide a smaller market.
When the core search terms in your category are declining year over year, the fix is not broader keywords, higher bids or a new bid strategy. Confirm that demand really fell (rather than your share or your price), cap search at the size of the demand that is left, and move the difference into channels that create the next year's searches.
This is for marketing leaders at mid-market and enterprise B2B companies whose Google Ads program has been a dependable pipeline source and has started to shrink for reasons nobody can explain. It is a different problem from sizing demand before you pick a first channel, which we cover in our guide to B2B paid channel selection. This post is about an account that already runs search into a category that is getting smaller.
Three reasons non-brand search leads fall
Non-brand search leads drop for one of three reasons, and only one of them is a shrinking category. The other two are fixable inside the account, which is why the diagnosis has to come before any budget move.
→ Volume. Fewer people are typing your category terms into Google. The market for search attention got smaller.
→ Share. The searches are still happening, but you are showing up for fewer of them because competitors outbid or outrank you.
→ Price. You still show up, but each click costs more, so the same budget buys fewer visits. We wrote about that pattern separately in why rising Google Ads costs call for a better message rather than bigger bids.
A pattern we see often looks like this. Brand search leads hold steady for months while competitor and generic search leads slide, and the first internal reaction is to cut the budget on the campaigns that are underperforming. That is usually the worst move available, because the cut lands on the traffic that was still converting, and a share or price problem gets misread as a demand problem.
Ready for paid ads that pay off?
Book your free auditHow to tell if category search volume is really declining
The best demand sensor you own is your own account, not a keyword tool. Google defines impression share as "impressions / total eligible impressions" in its About impression share documentation. Rearrange that and you get eligible impressions, which is your impressions divided by your impression share. That number is an estimate of how many searches you could have appeared on, which is the closest thing to a market-size reading the platform gives you.
Here is how the arithmetic separates volume from share, with round illustrative numbers. Last September a core campaign earned 4,000 impressions at 80% impression share, which means roughly 5,000 eligible impressions. This September it earned 3,000.
→ If impression share is still 80%, eligible impressions fell to about 3,750. The market shrank by a quarter and you kept your share of it.
→ If impression share dropped to 60%, eligible impressions are still about 5,000. The market is flat and you lost share, which is a bidding, quality or budget problem you can work on.
Run the check in this order:
1. Pick the core category keywords that have been live and unchanged for at least 24 months, on the same location and language targeting.
2. Pull impressions and impression share for those keywords by month, and compare each month to the same month a year earlier so seasonality cancels out.
3. Calculate eligible impressions for each month and chart the year-over-year change.
4. Pull Search lost IS (budget) and Search lost IS (rank), which Google's impression share data guide defines as the share of time your ads did not show because of insufficient budget and because of poor Ad Rank. A climbing rank figure points at share, not demand.
5. Open the auction insights report and look at overlap rate and outranking share for the same period. Google's auction insights documentation notes the report shows nothing when your impression share is under 10%, so very thin campaigns will come back blank.
6. Check branded search volume alongside category volume. If both are falling, demand is shrinking. If brand holds while category eligible impressions fall, fewer new buyers are arriving even though the people who already know you are still searching.
One honest caveat on this method. Google says impression share "is based on an estimate" and that changes to your bids, quality or Google's own systems can change the set of auctions it counts. Read the trend over several months, not a single swing.
Why keyword tools are a weak second opinion
Keyword Planner's year-over-year change column is a useful sanity check, but its own forecast documentation says historical stats like average monthly searches are "only shown for exact matches" and count a keyword together with its close variants. The same page reminds you that search counts are "constantly fluctuating" with seasonality and current events. On a B2B term with a few hundred monthly searches, ordinary fluctuation can look like a trend in either direction.
Google Trends is worse for niche B2B terms. Its data FAQ explains that results are normalized and scaled 0 to 100, that terms searched by very few people show up as zero, and that the statistical noise it adds for privacy "is most noticeable on queries with low or no search interest". For a category phrase with double-digit monthly volume, the Trends line can look like a heartbeat monitor that means nothing.
Why broadening keywords makes a shrinking category worse
When category volume falls, the instinctive fix is to go wider, and I think it is the wrong one. Teams add adjacent terms, loosen match types, or hand the account a broader automated campaign so the lead count recovers. It usually does recover, on the dashboard.
What actually happens is that the bidding system finds volume wherever it can, and in a shrinking category the only volume left to find is lower intent. The form fills come back. The qualified pipeline does not, because the people who are no longer searching for your category did not move to a slightly different phrase. They stopped searching.
The cost shows up two quarters later, when sales starts asking why lead quality fell off a cliff while the paid report says volume is fine. By then the account has spent months teaching its bid strategy to chase the wrong people.
Cap search at the size of the demand that is left
Do not cut the core terms. Cut the budget that the core terms can no longer absorb. The buyers who still search for your exact category are the most valuable traffic you will buy all year. A shrinking category makes each of them more valuable, not less.
The practical move is to set search budget to what full coverage of the remaining demand costs, and no more:
→ Keep core category and brand terms at high impression share, with Search lost IS (budget) close to zero on those campaigns.
→ Let the total search budget fall to match. If eligible impressions dropped 25% and your cost per click held, the money that coverage needs dropped by roughly the same amount.
→ Resist refilling the gap with adjacent terms unless they have their own history of producing sales-accepted leads.
This is also where impression share becomes a planning metric rather than a vanity one. You are no longer asking how to buy more search. You are asking what it costs to own all of the search that still exists.
Move the difference into creating demand
If fewer people search for your category, somebody has to give them a reason to start. The money you freed from search should go to channels that reach buyers before they know the category name, such as LinkedIn, video, thought leadership and problem-aware content promoted to a defined account list.
The mistake is to judge that spend by the scoreboard you used for search. A demand creation channel will lose a last-click lead count against search every single month, and if it is judged that way it gets cut in the first budget review. Give it its own measures, agreed before launch, the way we recommend in our post on measuring B2B brand campaigns.
There is a useful loop here. The clearest sign that demand creation is working is that branded and category searches start rising again inside your own Google Ads account. Search stops being the engine and becomes the meter that tells you whether the engine is running.
When the right call is to pause non-brand search
Sometimes the honest recommendation is to stop. If eligible impressions on your core terms are now so low that full coverage cannot produce one qualified conversion a month, non-brand search is no longer a channel. It is a small tax.
We have told clients to pause in exactly this situation, and it is not a popular conversation. Keep a lean brand campaign so competitors cannot take your name for free, put the rest into demand creation, and re-run the eligible impressions check every quarter. If the category comes back, you will see it in the numbers before anyone else does.
A small search budget in a narrow niche can still be worth running when one qualified lead is worth many multiples of the monthly spend. The pause test is not about budget size. It is about whether the searches exist at all.
What to tell leadership
Frame a shrinking category as a market finding, not a performance failure. The paid team did not lose the searches. The market stopped making them. That framing matters, because a leadership team that believes search is underperforming will keep asking for the old lead count, and the only way to hit it is to buy the wrong traffic.
Bring three numbers to that conversation. The first is the year-over-year change in eligible impressions on core terms. The second is what it costs to keep full coverage of the demand that remains. The third is the budget you propose to move into demand creation, along with the measures you will judge it on.
This week, pull 24 months of impressions and impression share on your oldest core keywords and calculate eligible impressions by month. If the line is flat, your problem is share or price, and that is good news because it is fixable inside the account. If the line is falling, stop buying keywords and start planning where the next year's searches will come from. If you want a second set of eyes on that read, that is the kind of work our B2B paid ads team does every week.
