Attribution in e-commerce rarely follows a neat, linear path. A brand may see a 4x ROAS in Facebook Ads, another 3x in Google Ads, and yet find that the commercial reality feels far less straightforward when revenue, margin, and cash flow are reviewed together.
Across modern acquisition channels, it’s common to see a persistent gap, often around 20%, between what individual platforms report and what the business ultimately experiences. At Positive Sparks, we see this as a natural consequence of fragmented measurement, overlapping channel credit, and attribution models that were not built for today’s customer journey.
The opportunity is not simply to identify the gap, but to interpret it well. Once brands understand where reported performance diverges from commercial truth, they can make sharper investment decisions, protect margin, and scale with greater confidence.
What is the "20% Gap" (And Why Should You Care?)
Why is there always a discrepancy? If you’re using standard "last-click" attribution, you’re basically giving a trophy to the person who handed the runner a bottle of water at the finish line, while completely ignoring the athlete who actually ran the 26 miles.
In the world of e-commerce, that 20% gap is where your profit goes to hide. This gap typically exists because of three main culprits:
- Platform Overlap: Meta, Google, and your affiliate networks all want to take credit for the same sale.
- Model Bias: Last-click models ignore the 5 to 15 touchpoints that happened before the final click.
- The "Ghost" Conversions: Sales that would have happened anyway, even if you hadn't spent a penny on that specific ad.
When you move from a messy, platform-siloed view to a unified, first-party data approach, you almost always see a 20% "deflation" in your reported ROAS. It feels like bad news at first, but it’s actually the most valuable data you’ll ever own. It’s the difference between "Vanity ROAS" and True ROAS.
The Last-Click Trap: A Familiar Model with Strategic Limits
Are you still relying on last-click attribution? Bear in mind, this model was built for a much simpler digital environment. Today, customers move from Instagram Reels to Google Search, then back to email, and finally convert through a retargeting ad on another device altogether.
When businesses evaluate performance through the final click alone, budget decisions can become distorted. A brand might see that its Google Ads appear to be driving strong returns and decide to reduce spend on Influencer Marketing. Yet the broader sales picture can weaken quickly, because those influencers may have been creating the very demand that search later captured.
We’ve seen clients assume their Meta Ads were underperforming because last-click ROAS looked modest. In practice, those campaigns were often doing the foundational work of building awareness and consideration. Without that contribution, the entire growth system becomes less effective.
The Cannibalization Question: Enter Intend and Essendia
Let’s turn to one of the more commercially sensitive areas of attribution: affiliate cannibalization.
Have you ever noticed how platforms like Intend or Essendia, along with other coupon or cashback partners, can appear to deliver exceptionally strong performance? On paper, the ROAS can look remarkable. The more important question, though, is whether that affiliate interaction genuinely created the sale or simply appeared at the final stage of a purchase already in motion.
A common scenario is what we might call "the checkout hunt." A customer reaches the checkout page, notices a discount code field, opens a new tab, searches for "[Your Brand] Coupon," clicks an affiliate link, applies a 10% code, and completes the order.
In reporting terms, a platform such as Intend or Essendia may then receive full credit for the conversion. In commercial terms, however, the picture is often more nuanced. The customer may already have been fully committed to purchasing, with the affiliate merely inserting itself at the last moment.
That is the essence of cannibalization. The business is not only paying commission, but often giving away margin on a sale that was already highly likely to convert. When we look at True ROAS, the goal is to separate genuine demand creation from late-stage interception and understand who truly originated the sale.
Why Multi-Touch Attribution (MTA) Isn’t Just a Buzzword
You might be wondering, "Okay, if last-click is lying, what’s the alternative?"
The answer lies in Multi-Touch Attribution. Instead of giving 100% of the credit to the last click, we spread the credit across the entire journey. We ask questions like:
- Who introduced the customer to us? (First Touch)
- Who kept them interested? (Middle Touches)
- Who finally convinced them to buy? (Last Touch)
By using a more sophisticated model, that 20% gap starts to make sense. You realize that your affiliates might only deserve 20% of the credit they're claiming, while your top-of-funnel ads deserve 3x what they’re currently getting.
At Positive Sparks, we help brands navigate this shift, moving them away from Amazon-only sales to a thriving Direct-to-Consumer (DTC) model where you actually own your customer data.
The Solution: First-Party Data & True ROAS
So, what creates a more reliable view of performance? It comes down to First-Party Data.
As third-party cookies continue to disappear and privacy restrictions across Meta and Google become more significant, platform-reported numbers are naturally becoming less complete. Each platform offers useful signals, but no single one should be treated as the sole source of truth.
True ROAS becomes possible when a business uses its own internal tracking as that source of truth, allowing it to de-duplicate sales and measure incrementality with greater confidence.
What is incrementality? Put simply, it is the business question behind the metric: "If we stopped investing in this channel tomorrow, how many sales would we actually lose?"
If an affiliate such as Essendia is paused and total sales remain unchanged, that suggests little to no incremental value was being added. By contrast, if TikTok Ads are reduced and total revenue falls by 30%, the commercial contribution of that channel becomes much clearer, regardless of what a last-click report may suggest.
Actionable Steps to Close the Gap Today
Looking to create a clearer view of performance? Here are three things you can do right now:
- Audit Your Affiliates: Look at your "Time to Conversion" for affiliate clicks. If the click happens within 2 minutes of the purchase, it’s almost certainly a coupon-hunt at checkout. Consider reducing commissions for these "last-minute" interrupters.
- Compare Your Models: Use a tool (or a partner like us) to compare your Last-Click ROAS against a Position-Based or Data-Driven model. Seeing the difference is usually a "lightbulb" moment for e-commerce owners.
- Focus on New Customer Acquisition (NCA): Don't just look at total ROAS. Look at how much it costs to acquire a new customer. That is the only metric that truly scales a business long-term.
Let’s Get Real About Your Numbers
At the end of the day, we’re all trying to build profitable, sustainable businesses. We want to spend $1 and see $5 come back into the bank, not just appear on a dashboard.
Is your data telling a complete commercial story, or simply the most convenient platform version of events? We’ve helped dozens of brands uncover their "True ROAS" and identify the inefficiencies that were holding back smarter growth. Whether you’re scaling Microsoft Ads or trying to understand why your retargeting performance feels difficult to validate, we’re here to help you make sense of the numbers.
What about you? Have you noticed a gap between your platform data and your actual bank balance? Which channel do you think is claiming more credit than it really deserves?
Drop us a comment or reach out to our team to start your True ROAS audit today. Let’s uncover that missing 20% together.
