The client's campaigns were not badly run. They were run against the wrong measure — and that is a much harder problem to see from the inside.
The Challenge
Solvenza was spending a healthy monthly budget and reporting a healthy cost per lead. The trouble was that "lead" meant anyone who downloaded a PDF. Sales quietly ignored roughly seven in ten of them, so the number that looked good on the marketing dashboard had almost no relationship to revenue.
Meanwhile 63% of the budget sat on broad, high-volume keywords that generated plenty of downloads and very few conversations.
Fixing the measurement first
Before touching a single campaign we spent four weeks connecting ad spend to the CRM, so that every closed deal could be traced back to the click that started it. This is unglamorous work and it is almost always where the real gain is hiding.
With that in place the picture inverted. The cheapest leads were the least likely to close. A handful of expensive, low-volume search terms — the ones with a competitor's name or a specific integration in them — were producing most of the revenue.
What We Did
We rebuilt the account around intent tiers rather than topics. High-intent terms got dedicated budget, dedicated landing pages and no shared budget cap. Broad terms were capped and treated as a research line item, not an acquisition channel.
We replaced the single generic landing page with six, each matching the promise of the ad that led to it. The page for the integration query talks about that integration in its first sentence — which sounds obvious and was, in fact, the single largest contributor to the conversion gain.
Reporting moved from cost per lead to cost per qualified opportunity, agreed jointly with the sales director. Everyone stopped optimising for a number nobody trusted.
What we tested
- Ad copy — specific numbers beat superlatives in eleven of twelve tests.
- Form length — reducing seven fields to three raised submissions 31% and lowered quality by almost nothing, because the qualifying question stayed.
- Bidding — manual control outperformed automation for the first six weeks, then automation won once the conversion signal was clean. Sequence mattered more than choice.
The Outcome
Cost per acquisition fell 38% over the quarter while total spend stayed flat. Return on ad spend reached 4.1x, measured on closed revenue rather than pipeline. Demo-to-close improved 27% simply because the demos were with better-matched prospects.
Budget distribution reversed: 61% now sits on high-intent terms, against 37% at the start.
The first month we spent the same money and generated fewer leads. Our CEO was not thrilled. The quarter it closed on was the best we have had.
Priya Raman — Head of Demand Generation, Solvenza
What we would do differently
We should have agreed the "fewer leads before better leads" trade-off with the board, not only with the marketing team. Four weeks of uncomfortable meetings were entirely avoidable with one slide at the start.
Want a second opinion on your campaigns?
— we will tell you what we would change before you commit to anything.
