A small paid search program usually gets its verdict from one big deal or one quiet quarter. Neither is evidence. Here is the review we run before anyone keeps or cuts the budget.
Google Ads is worth it for a B2B company when its non-brand campaigns produce enough qualified opportunities, over at least one full sales cycle, to cover their spend at your win rate and deal size, without leaning on a single outlier deal. If you cannot answer that from your own data yet, the honest verdict is not keep or cut. It is that the program has not been read properly.
This post is for marketing leaders and founders who run a modest search budget, see a handful of leads a month, and now have to defend or kill that line item in next year's plan.
Why small search programs get judged on the wrong evidence
Annual planning is when a small paid program is most at risk. Someone in the room remembers the big customer that came from Google last spring. Someone else points out that the last three months produced almost nothing. Both are telling the truth, and neither is telling you whether the channel works.
At low volume, the numbers swing hard. One enterprise deal can make a $4,000-a-month program look like the best investment the company made all year. One slow quarter can make the same program look like a leak.
We see this pattern in almost every audit of a small account. The decision gets made on the most memorable data point, and the most memorable data point is almost always an outlier.
The fix is a short, structured review that answers four questions in order. Skip one and the verdict is a guess.
Ready for paid ads that pay off?
Book your free auditStep 1: Take brand out of the numbers first
Brand search leads come from people who already know your name. Most of them would have found you through the organic listing or a direct visit. If brand and non-brand sit together in the report, brand quietly carries the program and makes a weak non-brand effort look healthy.
Split every lead, opportunity and dollar into brand and non-brand before you look at anything else. The keep-or-cut question is really about non-brand, because that is the part of the program reaching buyers who did not already know you.
Brand is a separate, cheaper decision. A capped brand campaign is often worth keeping even when non-brand gets cut, especially when competitors bid on your name. We covered that split in more detail in branded vs non-branded search.
Step 2: Count qualified pipeline across a full sales cycle
Judge the program on qualified opportunities created by clicks, grouped by the month of the click, not the month the deal closed. A B2B lead that clicked in August might not become an opportunity until November. If you count by close date, the most recent months will always look empty, and that is exactly when the budget review happens.
Google's own reporting works this way. Its documentation on conversion lag notes that "Because Google Ads reports conversions by query date, you may not always see the most updated conversion numbers." Your CRM should be read the same way.
So take your average sales cycle and leave the youngest slice of clicks out of the verdict. If deals take roughly 90 days to go from first touch to qualified opportunity, a review in October should judge clicks from January through June or July. The months after that are still maturing.
Then count opportunities your sales team actually accepted, not form fills. A small program that produces 30 leads and 2 real conversations is a different program from one that produces 12 leads and 6 real conversations.
Step 3: Run the verdict twice, with and without the biggest deal
One outlier deal should neither save nor sink a small program. At low volume, closed revenue is mostly luck about which deal happened to land inside the window. Qualified opportunities are a steadier signal.
Here is the math we use, with illustrative numbers. Take the non-brand qualified opportunities from mature clicks, multiply by your normal win rate, then by your normal first-year contract value.
→ $48,000 of non-brand spend over the period
→ 8 non-brand qualified opportunities from mature clicks
→ A normal win rate of 20%
→ A normal first-year contract value of $40,000
→ Expected first-year revenue of 8 × 20% × $40,000 = $64,000
That program returns about 1.3 times its spend in year one on expected value, before renewals. Whether that clears your bar depends on your margins and retention, but it is a real answer.
Now look at actual closed revenue and remove the largest single deal. If the program only passes because of that one deal, it is a program that got lucky, not one that works. If it fails on closed revenue but passes on expected value, the most likely explanation is timing, and you probably have deals still in the pipe.
The rule is simple. A program has to pass on expected value from its opportunities. One huge deal is a bonus, not the business case.
Step 4: Decide whether you tested the channel or the setup
A lot of small programs have never actually been asked to produce leads. The most common example is an account running Maximize Clicks across many campaigns. Google describes that strategy plainly: it "sets your bids to help get as many clicks as possible within your budget." Clicks are what it buys, so clicks are what you got.
The second common example is a conversion list with dozens of actions, many of them page views, scroll events or tools that were retired years ago. When everything counts as a conversion, the account cannot tell a demo request from a careers page visit.
If either is true, the review has not tested Google Ads. It has tested a setup that was never pointed at pipeline. Cutting the channel on that evidence is like firing a salesperson who was assigned the wrong territory.
In that case the verdict is a third option, fix and re-read. Consolidate into a few high-intent campaigns, bid toward one primary action that sales actually wants, and set a re-read date one full sales cycle out. Our guide to cleaning the conversion action list walks through the second half of that.
Before you blame the channel, check the auction
Sometimes the setup is fine and lead volume still drops because a competitor started spending more on the same terms. That is a different problem from a channel that does not work, and it calls for a different response.
Google's auction insights report shows it directly. Overlap rate "is how often another advertiser’s ad received an impression when your ad also received an impression." If a new name shows up with a high overlap rate and your costs per click climbed in the same months, the drop has an explanation.
Rising costs alone are not a reason to cut. The question is whether your cost per qualified opportunity is still below what an opportunity is worth to you. If it is, you are paying more for something that still pays back. If it is not, the answer is usually sharper keywords and a sharper message rather than higher bids, which we laid out in how to respond to rising Google Ads costs.
The three verdicts and what each one means
Every review should end in one of three outcomes. Write the outcome down, along with the evidence behind it.
Keep. Non-brand passes on expected value from mature clicks, it still passes without the biggest deal, and the setup bids toward a real qualified action. Fund it, and look at whether it is limited by budget before you add anything new.
Fix and re-read. The setup was never pointed at pipeline, or there are too few mature months to judge. Do not cut it and do not grow it. Make the setup changes in one go, agree the pass threshold in writing today, and put the re-read on the calendar one sales cycle out.
Cut non-brand. The setup was sound, a full cycle has matured, and the program fails on expected value with or without the outlier. Turn off non-brand, keep a capped brand campaign if competitors are on your name, and write one sentence about what would make you reopen it, such as a new product line with its own search demand.
The middle verdict is the one most teams skip, and it is the one that matters most. In our experience, a meaningful share of programs that get cut were really programs that were never set up to be judged.
How this differs from deciding whether to start
If you have never run paid search, this review does not apply yet. The right question there is whether you are ready to advertise at all, which comes down to deal value, a defined qualified conversion and enough volume to learn. That is a different decision, and we wrote a separate five-check readiness test for it.
This review is for the program that already exists, has some history, and needs an honest verdict before the budget meeting.
What to bring to the budget meeting
Bring one page. On it, put non-brand spend for the mature period, non-brand qualified opportunities by click month, the expected-value calculation, the same calculation without the largest deal, a one-line note on bid strategy and primary conversion action, and the verdict.
That page will survive the room in a way that "we got a big customer from Google once" will not. It also forces the useful argument, which is about the evidence rather than whoever has the most memorable story.
If you want a second set of eyes on the review, we run this kind of read on existing accounts as part of our B2B Google Ads work, and we will tell you plainly when the answer is to cut.
