If your dashboard keeps telling you to do more retargeting, more brand search, and pour more budget into the bottom of the funnel, you know the feeling.
The numbers say yes. Your gut says no.
And the brand and mid-funnel campaigns you actually believe in keep losing the budget fight.
This post explains why marketing attribution alone keeps giving you that answer, why mid-funnel and top-funnel campaigns get systematically undervalued, and what enterprise ecommerce and retail teams need to add to attribution to measure marketing accurately.
In the video below, Juha Nuutinen, Co-founder and CEO of Sellforte, walks through why attribution is not enough on its own.
I actually love attribution, because it gives you highly granular data at campaign, ad-set, and keyword level. It's really, really important data to have. But it is not enough.
— Juha Nuutinen, CEO & Co-founder, Sellforte
There are two specific problems with using attribution as your only measurement system, and once you see them, you cannot unsee them.
The first problem is that attribution doesn't understand incrementality.
Attribution will always tell you: do more Meta retargeting, do more Google Search brand, because these are the channels that have high ROAS in your attribution model. But these are also channels that capture existing demand, so the incrementality of these channels is typically low.
Your dashboard rewards channels that take credit for demand that was already there. It does not reward channels that created the demand in the first place.
💡 To understand why average ROAS and incremental ROAS tell completely different stories about your campaigns, read: ROAS, iROAS, miROAS: Choosing the Right KPI for Optimizing Media Spend
The second problem is that your attribution model really doesn't like mid-funnel and top-funnel campaigns. It punishes them.
If you love to put everything in the bottom of the funnel and just capture demand, attribution is great for you. But if you are thinking in funnel stages, and you believe mid-funnel and awareness channels are really important for growing and creating demand, then attribution is not the tool for you.
You want a tool that can include all of the channels in one system: Meta awareness, traffic campaigns in paid social, influencers, TV, radio, YouTube, out of home. Not only optimizing your bottom of funnel.
💡 To see how upper funnel and brand channels actually drive new customer acquisition that attribution cannot see, read: Why Measuring Upper Funnel Marketing Is So Hard in Ecommerce
For omnichannel retailers and enterprise fashion ecommerce brands, the limitations of attribution-only measurement are especially severe. TV, radio, DOOH, print catalog, and brand awareness campaigns get systematically undercredited, while retargeting and brand search absorb credit for demand those upper funnel channels actually created.
In omnichannel, the picture is even worse: online ads often drive in-store purchases that attribution cannot see at all. The digital team looks undervalued. The brand team has to defend TV spend in every budget meeting. And the controlling team is left wondering whether any of these numbers can be trusted.
There is a second omnichannel-specific problem worth naming: promotions and seasonality contaminate the signal. Discount weeks, end-of-season sales, weather, and holiday peaks all move sales independently of media. If your measurement system can't separate the promo-driven baseline from the media-driven incremental sales, every channel gets credited for lift it didn't cause. For fashion and retail buyers, this is the difference between a model finance trusts and a model that gets quietly ignored.
A complete measurement system is not optional for enterprise retail. It is the only way to give every channel a comparable signal that finance and marketing can both work from.
Here is where the attribution problem gets operational. The Target ROAS values you set in Google Ads and Meta are derived from your attributed ROAS data. If that attributed data systematically overcredits retargeting and undercredits upper funnel, your bid algorithms are optimizing toward the wrong target on every channel.
This is the pain enterprise marketing teams articulate most often once they look closely: not just "I don't know which channels work", but "my bidding strategies are pointed at numbers I don't believe."
The fix is not to abandon Target ROAS bidding. The fix is to feed those bid algorithms a calibrated number. Incremental ROAS as the input, attributed ROAS as the legacy reference. That is what changes how budgets actually flow into platforms day-to-day.
💡 To learn how to set Target ROAS values that account for true incrementality, read: How to Set a Target ROAS That Reflects True Incrementality
For most enterprise ecommerce and retail teams, the hardest budget conversation is not with the agency or the platform reps. It's with the CFO.
Marketing believes in the brand spend. Finance does not. The numbers from the platform dashboards don't survive a procurement review, because procurement knows that Meta and Google are both grading their own homework. What finance wants is a number produced by a method that does not depend on the platforms reporting on themselves.
That is the role marketing mix modeling plays. An independent, methodology-transparent measurement of how media drove revenue, built on first-party sales data and external factors, with incrementality experiments validating the result.
When the CFO asks "how do you know that TV campaign worked?", you do not want to answer with a screenshot from a media platform. You want to answer with a number that came from an MMM that triangulates against a controlled experiment.
This is one of the most common concerns we hear from enterprise ecommerce teams. We have built a dynamic attribution model, it is integrated into our reporting and steering processes, and we do not want to throw it away.
The good news is that you do not have to. The calibration multiplier approach, anchoring your existing attributed ROAS to incrementality-based truth through calibration factors from a marketing mix model and incrementality experiments, is exactly how Sellforte works.
You preserve the operational benefits of your attribution model and the campaign and ad-set level granularity it gives you. But you recalibrate the numbers it produces so they reflect incremental sales, not just attributed sales. Those recalibrated numbers are then what feeds your Target ROAS bidding strategies, your CFO conversations, and your channel-mix decisions.
💡 To learn the methodology behind calibration and causal attribution, read: What is Causal Attribution in Marketing?
A complete measurement system combines three methods, each playing a distinct role:
Attribution for granular campaign-level performance data
Marketing Mix Modeling (MMM) for total causal impact, lagged effects, and cross-channel comparability
Incrementality testing for the truth check that calibrates everything else
This is what Google calls the modern measurement triangle. No single method is enough. Together, they cover what attribution misses: incrementality, lagged effects, and the value of upper funnel and mid-funnel channels.
💡 To learn what incrementality means and how it is measured, read: What is Incrementality in Marketing?
Attribution gives you really, really important data. It is not the enemy. But on its own, it will systematically push you toward channels that capture existing demand and away from channels that create new demand.
If you only steer by attribution, you will keep optimizing toward what looks good in the dashboard. You will keep underinvesting in mid-funnel and upper funnel. You will keep setting Target ROAS values your team doesn't trust. And you will keep losing the brand budget conversation with finance.
Most enterprise ecommerce and retail teams already have the building blocks. The attribution model is in place. The experiments are running. The MMM exists somewhere in the organization. What is missing is the connection between them.
That is the gap Sellforte closes.
Sellforte unifies Marketing Mix Modeling, Incrementality Testing, and Attribution into a single always-on platform, delivering true incremental ROAS at campaign and ad set level for enterprise retail and ecommerce brands.
To learn more how it works, book a demo with Sellforte.
Daria Alén is Senior Marketing Manager at Sellforte, where she builds educational programs, webinars, and events for ecommerce and DTC growth teams. She has over 10 years of marketing experience in B2B SaaS and Tech with specialization in go-to-market strategy and marketing analytics. Follow Daria on LinkedIn for more about marketing and growth.