B2B PPC waste: how to find the third of the budget doing nothing

Ask most performance marketers to fix a flat account and they will reach for tighter targeting, fresher creative, smarter bidding. Sometimes that is the right call. More often, in accounts I inherit, a meaningful slice of the budget is buying activity that would not be missed if it disappeared tomorrow.

The reason it persists is not incompetence. It is that nobody is rewarded for removing a line item, and every line item has a plausible story attached to it.

The four places b2b ppc waste actually lives

1. 1

   Retargeting at frequency nobody sane would tolerate

   Windows set to 180 days, no frequency cap, chasing everyone who ever loaded a blog post. It reports beautifully because it takes credit for people who were already coming back. Cut the window to 30 days and cap frequency, and watch how little changes.

2. 2

   Branded search defended against nobody

   Worth paying for when competitors bid on your name. Worth much less when they do not, and you are paying to appear above an organic result you already own. Test it by pausing for two weeks and measuring total branded clicks, not paid clicks.

3. 3

   Broad awareness with no downstream measurement

   Not because awareness does not work, but because nobody defined what would count as it working. Spend with no falsifiable expectation attached is not a bet, it is a habit.

4. 4

   The channel somebody senior likes

   Every account has one. It survives review after review because the argument costs more than the budget. It is usually the cleanest cut available.

Spend that cannot be defended with a falsifiable expectation is not an investment. It is a subscription.

Why cost per lead hides all of it

Cost per lead is the metric that keeps waste alive, because the cheapest leads usually come from the least serious audiences. A campaign producing leads at a third of the account average is normally not efficient, it is attracting people with nothing at stake.

Cost per lead vs pipeline per euro

How to run the cut without losing the argument

• Write the prediction down first. State what you expect to happen to pipeline over the next six weeks. A cut without a stated expectation becomes a permanent debate.
• Pause, do not reallocate. If you move the money the same day, you will never know which change caused the outcome.
• Give it a full sales cycle before the verdict. Most B2B pauses look terrifying in week two and fine in week seven.
• Cut one thing at a time when the stakes are high, and several at once when the budget is small and the learning matters more than the precision.

The best case for leaving it alone

What to do with the money

Not necessarily anything, at first. The reflex to redeploy immediately is what created the waste in the first place: a budget that must be spent gets spent. Hold it for a cycle and let the account show you where the constraint actually is.

When you do redeploy, the honest ranking is usually the same. Bottom of funnel capture first, because it converts and it is almost always underfunded. Then creative, because in a saturated auction the asset does more work than the targeting. Then a considered bet on one new channel, with a stated expectation and a date to review it.

The audit checklist

A worked example of the full audit

The forecasting mistake that hides waste even longer

Pipeline forecasts built off historical conversion rates make waste harder to see, not easier, because a forecast built on a blended average bakes the underperforming campaigns into next quarter's target. If the account has been carrying a third of spend that produces almost nothing, the forecast for next quarter assumes that third will keep producing almost nothing, and the plan quietly works around it instead of questioning it.

This is why the audit has to happen before the planning cycle, not during it. Once next quarter's targets are set using this quarter's blended numbers, cutting a channel becomes a forecasting problem as well as a budget problem, and the two arguments get tangled together in a way that makes the cut much harder to make cleanly.

Where the line gets drawn on brand and awareness spend

Nothing in this argument is against brand spend. It is against unmeasured brand spend that has never had a falsifiable expectation attached to it. There is a real difference between a campaign built to build awareness with a stated proxy metric, branded search volume, share of voice in a specific channel, direct traffic growth, and a campaign that runs because awareness is generally considered good and nobody has been asked to define what winning would look like.

The fix is the same in both cases: name what you expect to happen and by when, before the money goes out the door. A brand campaign with a six month view and a stated proxy metric is an investment. The same spend with no stated expectation is the fourth category of waste from the list above, dressed up in better language.

Why this argument applies below the account level too

The same logic scales down to a single campaign's ad sets or a single search campaign's keyword list. Inside almost any active search campaign, a handful of keywords are absorbing the majority of spend at a cost per opportunity nobody has checked in months, while a smaller set of keywords produces most of the pipeline at a fraction of the spend. The instinct to optimise bids campaign wide misses this every time, because the campaign level average hides the same split that the account level average hides.

A second worked example, at the keyword level

The account level cut is the easy version of this argument to accept, because a whole campaign is a visible line item. The same math holds one level down, inside a single search campaign, and it is worth walking through once because it is where most teams stop looking.

How to run this test on your own account

1. 1

   Export search terms, not just campaigns, for the last two full quarters

   Campaign level reporting hides keyword level waste the same way MQL reporting hides lead level waste. The unit of analysis has to match the unit where the decision gets made.

2. 2

   Join spend to pipeline, not to clicks or leads

   This usually means a manual join between the ad platform export and the CRM, since few tools do it natively at the keyword level. It takes an afternoon, and it is the afternoon that finds the money.

3. 3

   Sort by cost per opportunity, then look at where spend actually sits

   The uncomfortable pattern in most accounts is that spend is weighted toward the keywords with the worst ratio, because those are usually the highest volume, highest cost terms that felt too important to touch.

4. 4

   Reallocate budget toward the ratio, not the volume

   Moving spend from four expensive keywords into twelve cheap ones that already convert better does not require new creative or a new landing page. It requires only the decision to do it.

Table: what each audit level actually catches

The objection worth taking seriously here too