Google Ads for B2B: Long Cycles and Small Numbers

A consumer account gets a thousand conversions a month and can test its way to an answer. A B2B account gets twenty, and half of those are students, competitors and people who misread what you sell. Every piece of optimisation advice written for the first account is actively harmful to the second.

The difference is not scale. It is that in B2B the feedback loop is longer than the attention span of everyone involved.

The measurement problem comes first

If your conversion is a form fill, you are optimising for the wrong event and you will keep doing so until it is changed. The form fill is the cheapest thing to produce and the least correlated with revenue.

What you need is the later event fed back into the account: a qualified opportunity, a proposal sent, a closed contract. That means the CRM has to talk to the ad platform, and the enquiry has to carry its origin all the way through. Without that, everything below is guesswork — which is why we build conversion tracking through to the CRM before we touch the campaigns.

Structure around the decision, not the product

B2B buyers do not search for your product. They search for the problem, then for the category, then for vendors, then for you by name. Four different states, four different intents, and one campaign covering all of them will let budget drift to the cheapest and least valuable.

• Problem-stage terms — high volume, low intent, cheap. Worth buying only if you have something worth reading at the end of the click and the patience to wait for the return.
• Category terms — the core. Expensive, and where most of the real pipeline starts.
• Vendor comparison — including competitor names. Low click-through rate, high value per click.
• Brand — cheap, converts well, and inflates every report it is mixed into. Keep it separate or you will never know what the rest is doing.

Small numbers change the statistics

With twenty conversions a month, a week that produces one instead of five is noise, not a trend. Accounts are wrecked routinely by reacting to that week. The practical rules:

• Judge on a rolling ninety days, not on the calendar month.
• Do not run an A/B test you cannot leave alone for six weeks.
• Change one thing at a time, because with this volume you will never untangle two.
• Expect automated bidding to underperform until conversion volume supports it, and say so before the engagement starts rather than in month three.

The same discipline applies to reporting, which is why the retainer includes a written weekly note rather than a dashboard link — a dashboard invites exactly the weekly overreaction that costs B2B accounts their results.

The cycle outlasts the attribution window

A nine-month decision seen through a thirty-day attribution window makes the channels that started the journey look worthless. They get cut first, and the pipeline collapses two quarters later with no apparent cause.

Extend the window, keep the first-touch record, and accept that this quarter’s spend is next year’s revenue. That is an argument to have with the board before the campaign launches, and it is part of what growth strategy exists to settle.

What good looks like

Not a low cost per lead. A stable cost per qualified opportunity, a sales team that calls the list, and a pipeline you can attribute back far enough to defend the budget. If you have the first without the second, you have bought activity.

Getting there usually starts with lead generation and a written proposal — not with an audit of what you are running now.