Seeing the Whole Funnel, Not Just the Last Click
StraighterLine, an online education platform running paid acquisition across Meta, Google, Bing, and TikTok to enroll new students in its self-paced online courses. One of AxiaOS's six grandfathered founding case-study clients.
First-party visitor and order data that split every channel into demand creation and demand conversion.
Re-evaluated Meta's role from last-click conversions to top-of-funnel engine.
The Setup
StraighterLine measured every channel the same way: by the conversions each ad platform reported, and by a generic web-analytics conversion rate on the marketing site. That framework has a blind spot built into it. It credits whoever touches the customer last, which quietly punishes any channel whose actual job is creating demand rather than closing it.
The Break
Reconciling all four ad platforms against StraighterLine's own first-party visitor and order data split every channel into two separate numbers instead of one blended conversion count: how many net-new visitors it brought in, and how many of those visitors converted to a verified new customer. The split told a very different story than last-click reporting ever could.
In June 2026, Meta brought in 100,063 net-new visitors, 45% of all new top-of-funnel traffic, the single largest source in the account, but converted only 15 of them to a verified new customer in-session. Read on last-click terms alone, that looks like a channel to cut. Read against where the rest of the funnel's new customers actually came from, it looks like the engine feeding everyone else: Direct converted 1,005 new customers from 75,641 new visitors (34% of new traffic), Organic converted 544 from 23,443 (10%), Google converted 803 from 17,979 (8%) at a verified cost of roughly $31 per new customer, and Bing converted 206 from 2,816 (1%) at roughly $15. Across the trailing eleven months the shape holds: Meta accounts for roughly 28% of all new visitors (532,000+), with the resulting conversions landing downstream in direct, organic, and search.
A second gap surfaced in the same review. Roughly one in five sessions on the marketing site (97,629 of 467,081 in June) were existing students logging in, not prospects, quietly diluting the conversion rate the team was trying to improve. Stripped of that noise, StraighterLine has a clean, true non-paid prospecting conversion rate of 0.86% for the first time.
The Diagnosis
Judging Meta by its last-click conversion count is exactly the mistake a first-party system should catch, not repeat. Meta is functioning as an upper-funnel demand-generation channel. It puts net-new visitors into the top of the funnel, and direct, organic, and search close the ones who convert. First-click and last-click attribution structurally under-credit exactly this kind of channel: view-through impressions, multi-session journeys, and cross-device paths mean the person Meta introduced often converts days or weeks later, through a different, lower-funnel channel, and gets credited there instead. The funnel is working as designed. It's just invisible to anyone only counting last clicks. Cutting Meta on its 15 last-click conversions would have starved the demand pipeline that direct, organic, and search all depend on to convert.
The Fix
The reconciled dashboard splits every channel into the two questions that actually matter: how much new demand does it create, and how much does it convert, instead of collapsing both into one last-click number. New-customer revenue is separated from returning/renewal revenue using the store's own new-customer flag, so marketing can see the dollars its acquisition spend actually drives. Traffic is filtered to real prospects, existing users, logged-in sessions, and login-page visits removed, producing an honest prospecting conversion rate instead of one diluted by StraighterLine's own returning students.
Where It Stands
StraighterLine can now weigh every channel on the job it's actually doing: Meta as the top-of-funnel engine (45% of new visitors in June), search converting efficiently and reconciling almost exactly with platform-reported claims (verified new-customer cost of $31 for Google, $15 for Bing), and direct capturing the largest share of the demand the upper-funnel channels create. June revenue totaled $2.60M, with $627k (24%) attributable to new customers specifically, a distinction StraighterLine couldn't previously isolate. Over the trailing eleven months, new-customer revenue totals $7.4M. A weekly plain-English brief (new customers, new-customer revenue, verified cost per new customer by channel, true site conversion) refreshes automatically, in the language the marketing team actually uses.
Why It Matters
Most attribution tools repackage whatever the ad platforms report, one last click at a time, which means they'll always tell you to cut the channel that's filling the funnel everyone else converts from. A first-party system built on the client's own data skips that blind spot, showing what each channel is actually for, not just which one happened to touch the sale last. For StraighterLine, that turned "which channel gets the credit?" into "what is each channel actually doing for us?", the difference between managing a funnel and guessing at one.
That distinction only survives because the funnel view lives in a system StraighterLine owns outright. A rented attribution tool has no reason to keep telling you a channel is working once it stops being the one taking last-click credit for it.