Auditing a paid programme that's plateaued

The call we get most often from B2B tech firms is some version of “the paid numbers used to grow, now they don’t, and we don’t know why”. The spend is similar, the campaigns are running, the leads are still arriving in roughly the right numbers, but pipeline has stalled, cost-per-opportunity is creeping up and nobody on the team can tell leadership what changed. The instinct is usually to fire the agency or switch tools. The right move is almost always to audit.

We have audited a lot of plateaued B2B tech paid programmes, and they tend to share a small set of root causes. Below is the structure we work through, with the diagnostic questions we ask first and the order in which to investigate.

Define the plateau before diagnosing it

The first move is establishing what plateau actually means on this account. Three numbers we want before any campaign is opened.

First, the trend across at least 18 months on cost per qualified opportunity, sourced from the CRM. Not cost per lead. Cost per opportunity, weighted to account for cycle changes. If this number has been flat for six months and the team has been adding budget, the actual situation is worse than flat (you are spending more for the same outcome, which is decline, not plateau).

Second, the trend on share of pipeline by source. If LinkedIn used to drive 35 per cent of new pipeline and now drives 18 per cent, that is a different problem than if LinkedIn share has held but absolute numbers have dropped. Channel mix shifts often reveal the issue before the campaign-level data does.

Third, the trend on win rate by source. A plateau where win rate is healthy is a volume problem. A plateau where win rate has dropped is a quality or fit problem. The interventions are different.

This baselining usually takes two or three days and is non-negotiable. Without it, every “fix” is a guess. The reporting setup that supports this work usually echoes our pipeline-tracking framework and the underlying conversion tracking guide for long B2B sales cycles.

The five most common root causes

In our experience, plateaued B2B tech paid programmes tend to suffer from one or more of five issues, in roughly this frequency.

1. Audience saturation

The most common cause. The ICP audience has been worked through and the channel is now showing the same ads to the same people. Frequency has crept up, click-through rates have softened, the auction is more expensive because everyone in scope has been bid against repeatedly.

Diagnostic: pull the LinkedIn or Google audience overlap and frequency reports across the last six months. If frequency is above three to four impressions per week per user across the addressable ICP, saturation is real. The fix is broadening the ICP definition (cautiously), refreshing creative aggressively or accepting that the channel has hit a natural ceiling and reallocating budget. Our LinkedIn Ads by vertical breakdown is a useful sense-check on what the addressable audience actually is for the relevant niche.

2. Creative fatigue

The same set of ads has been running for six to twelve months, the early winners have stopped winning and nobody has refreshed the rotation. Click-through rates and engagement metrics decline gradually, frequency stays high, the audience disengages.

Diagnostic: pull creative-level CTR and conversion data across the campaign window. A clear downward trend on the same creatives is the signal. The fix is a creative refresh against the same audiences, with at least three to five new variants per stage. The cost is real (creative pipelines are not cheap) but the ROI on a refresh is usually faster than any other intervention.

3. Attribution drift

The numbers look worse than reality because the tracking has degraded. iOS updates, browser-side restrictions, expired pixels, broken offline conversion imports, CRM stage changes that have not been mirrored in the conversion event setup. None of these changes the underlying pipeline. They just change what gets reported.

Diagnostic: walk through the conversion events in each platform, confirm that they fire correctly on a real test, check the offline conversion import logs, verify the pixel coverage on the website. We see broken conversion tracking on a meaningful proportion of plateaued accounts. The reports are wrong, the pipeline is fine.

The fix is the rebuild work covered in conversion tracking for long B2B sales cycles, often paired with a server-side tagging project for accounts at scale.

4. Quality drift in the CRM

The campaigns are still producing the same lead volume, but lead quality has shifted. Audience targeting has loosened over time, negatives have not been maintained, search-term reports have not been mined and the SDR team has quietly stopped working the Bing or LinkedIn leads because the SQL rate dropped.

Diagnostic: pull SQL and SAL rates by source and by campaign across the last 12 months. If a particular campaign or channel has decayed disproportionately, focus there. The fix is usually negative-keyword pruning, audience tightening and creative realignment with what the better-converting leads said about the journey. Our negative keyword strategy for tech piece covers the search side, and the wider demand-gen vs lead-gen budget framework helps reset the funnel mix.

5. Budget allocation against last year’s market

Budget splits often reflect the channel mix that worked 12 to 18 months ago rather than the current market. Google Ads has gentrified in some categories, LinkedIn CPMs have risen, AI search is taking informational queries off Google entirely. The old 60-30-10 split between Google, LinkedIn and other becomes structurally wrong.

Diagnostic: compare the source mix of the last 12 months of pipeline against the spend mix. Where pipeline share has decayed faster than spend share, budget is misallocated. The fix is reallocation, often into channels that were not in scope a year ago: more LinkedIn, less search; or more programmatic, less display; or a deliberate Reddit or Microsoft Ads test. Our Reddit Ads for B2B tech piece is the honest read on the most common “should we test this” channel question.

Structural questions to ask early

Beyond the five root causes, three structural questions catch issues that are easy to miss.

First, has the ICP changed? Plateaus often mask product or pricing shifts. If the firm now sells to larger organisations than 18 months ago, the campaigns calibrated for the old ICP are competing for the wrong audience.

Second, has the sales motion changed? A move from inbound-led to ABM-led changes which paid signal is valuable. The full demand-gen vs lead-gen budget frame applies.

Third, has the website changed? A homepage redesign that quietly reduced conversion rates, a new contact form requiring more fields, both show up as paid campaigns “plateauing” when in fact the destination is the issue. Our landing page CRO for paid traffic piece covers what to look for.

A typical audit timeline

For a meaningful B2B tech account, a paid audit usually takes two to three weeks of structured work.

We deliver the audit as a written document, not a deck. The recommendations come ranked by expected impact and effort, with a clear sequence rather than a wish-list.

What we will not recommend

A few things we deliberately avoid recommending in audits, even when they are the obvious-sounding fix.

Fully replacing one channel with another. Plateaued accounts rarely get out of the plateau through a single channel swap. The work is structural.

Adopting a new platform tool. The tooling is rarely the bottleneck. The campaigns, the data and the reporting framework usually are.

Doubling the budget. Adding budget to a plateaued account without diagnosing the cause is the most expensive way to confirm the problem is not budget.

When the plateau is real and structural

Sometimes the audit produces an honest answer the team did not want to hear: the channel has done what it can do, the ICP is fully addressed, the cost of incremental pipeline is genuinely higher than the value, and the right move is to plateau the spend rather than push through it. That is occasionally the right answer, and naming it explicitly is more useful than pretending growth is always available.

If your paid programme has plateaued and the marketing meeting has gone in circles for two quarters, an external audit is usually the cheapest way to break the loop. If you’re rebuilding a paid programme that’s drifted off-strategy, we’re happy to take a look. The wider context sits on our paid media service page.