The Four Attribution Problems Draining Your Paid Media Budget

Attribution is the conversation every learning and events business needs to have — but almost nobody wants to. It’s technical, it challenges assumptions about which channels are “working”, and fixing it properly requires investment before it produces any visible return. So it gets deferred. Campaigns run on last-click data. The wrong channels get cut. Budget gets concentrated in channels that look good in reports but are often harvesting demand created elsewhere.

We’ve audited over forty learning and events paid media accounts in the past three years, and the same attribution problems appear with striking consistency. This piece identifies the four most common ones, explains what’s actually happening, and sets out what good looks like.

Problem 1: Duplicate conversion counting

The most common tracking error we find — present in roughly 60% of accounts we audit — is duplicate conversion counting. It typically happens when a Google Ads conversion tag is firing alongside a GA4 goal import, both counting the same form submission. In some cases, multi-page form completions are also tracking intermediate steps as separate conversions.

The result: reported conversion volumes can be inflated by 40–150%. This means CPA looks better than it is, bidding strategies are trained on corrupt data, and the account appears to be performing better than it actually is. The fix is to audit every active conversion action in Google Ads, verify each one against real form submission data in GA4, and remove duplicates. Implement conversion action sets if you have multiple goal types (lead form, phone call, brochure download) and want to bid against a specific primary action.

Problem 2: Last-click attribution collapsing multi-touch journeys

Learning and events purchases routinely involve three to eight touchpoints across two to six weeks. A prospective delegate might see a LinkedIn post about your conference, click a Google Display ad two weeks later, search your conference name, see a Meta retargeting ad, and then book direct. Under last-click attribution, this conversion is assigned entirely to branded search or direct. LinkedIn, Display, and Meta receive zero credit.

This systematically undervalues top-of-funnel channels and leads to budget decisions that gradually concentrate spend in branded and high-intent search — which then struggles to grow because no prospecting activity is building the pipeline that feeds it. Google’s Data-Driven Attribution (DDA) is a meaningful improvement over last-click for single-channel within Google Ads. But for true cross-channel understanding, you need GA4 with Exploration reports, or ideally a BigQuery export feeding a custom attribution model that includes off-Google touchpoints.

Problem 3: No visibility into the offline conversion gap

For professional qualifications with a phone or email enquiry component — which is most of them — there’s a gap between the digital lead and the actual enrolment that most analytics setups cannot see. You know your Google Ads generated 120 enquiry form completions last month. You don’t know how many of those became paying students, because that data lives in a CRM or booking system that isn’t connected to your ad platforms.

Closing this gap requires offline conversion tracking: uploading CRM data (matched by email or phone number) back to Google Ads and Meta to tell the platforms which leads became revenue. This has two effects: it trains smart bidding algorithms on actual revenue rather than lead proxy signals, and it gives you true channel-level ROAS rather than CPL. For businesses where lead-to-enrolment rates vary significantly by channel — which is almost always the case — this matters enormously for budget allocation.

Problem 4: UTM parameter inconsistency

This is the most preventable problem on the list, and the most frustratingly common. UTM parameters are the strings appended to URLs to identify traffic source, medium, and campaign in analytics. When they’re applied inconsistently — or not at all — GA4 assigns traffic to “Direct” or “Organic” even when it came from a paid campaign. We regularly see accounts where 20–40% of paid social traffic is misattributed to direct because UTMs were stripped by link shorteners, omitted from some but not all ad variants, or formatted inconsistently (utm_source=linkedin vs utm_source=LinkedIn vs utm_source=linkedin-ads).

The fix is a UTM governance document: a standardised naming convention, maintained in a shared spreadsheet, with mandatory review before any new campaign or ad variant goes live. It takes one afternoon to build and eliminates this class of attribution error entirely. We build one for every new client in the first week of an engagement.

Getting this right before your next budget sign-off

Attribution problems tend to surface at budget review time, when someone asks which channel is driving revenue and the data can’t give a reliable answer. By then, it’s too late — decisions get made on flawed data or on gut instinct, and the cycle continues.

The investment to fix attribution properly is typically two to four weeks of technical work and a modest ongoing maintenance overhead. The return is the ability to make confident budget decisions, identify underperforming channels before they drain budget for six months, and make the case to stakeholders for investment in channels that last-click reporting systematically undervalues. In our experience, learning and events businesses that fix their attribution first consistently outperform those that optimise campaigns on top of bad data.