More people are discovering products on TikTok, researching on Amazon, and buying through either platform. Yet most platform measurement tools were built for a world where each channel operated independently.
TikTok shows you TikTok conversions. Amazon shows you Amazon sales. But neither provides a solution to answer the question every CMO is asking: What is our TikTok spend actually driving?
This isn’t a campaign problem. It’s a measurement problem.
The Measurement Gap
As dashboards measure direct in-app conversions, they don’t capture the spillover effect – otherwise known as the halo effect – when TikTok awareness drives Amazon search behavior and purchase decisions. In other words, native reporting can’t answer the critical question: How much of my Amazon revenue was influenced by my TikTok campaigns, even if the customer never clicked a TikTok link? This measurement gap could show up in different ways:
- Misattributed credit: A customer sees a TikTok video and doesn’t click or convert in-app. However, they search and purchase from your brand on Amazon days later. Attribution credits Amazon Ads or organic search, not TikTok.
- Invisible impact: Your TikTok Standard Ads drive zero direct conversions, so the dashboard shows $0 ROAS. Meanwhile, those same videos lifted Amazon brand page views by 15% and downstream sales by 8%. The campaign worked. You just couldn’t see it.
When people rely solely on platform native reporting, they’re flying blind on the channel’s true contribution.
What We Built
At Pacvue, our data science team built a halo model to measure the halo effect. Instead of tracking clicks, we use a mixed marketing model using your TikTok signals and Amazon outcomes to see the full picture.
As a result of the halo model, you can attribute how much of Amazon revenue came from TikTok, campaign by campaign, week by week.
We also measure three distinct outcomes: sales, page views, and search volume. This matters because if your TikTok campaigns drive massive page view halo, but minimal sales, you know your listings are driving engagement and awareness. On the other hand, if you’re driving sales halo without the traffic spike, you’re likely reaching loyal, brand-aware customers directly. Understanding the difference between these patterns is key to campaign optimization.
Our halo solution also adjusts to business needs whether you’re campaigning for a single hero product, multiple products, or building broader brand awareness. Within the UI, you can opt for product-level insights or brand-level insights, and each report will offer campaign-level breakdowns, and weekly aggregates that your teams can reference for budget planning.
Technical Breakdown
Our pipeline fits a Bayesian Media Mix Model that breaks down your daily Amazon KPI into additive components:
The baseline captures your organic activity (what would have happened without TikTok, including seasonality, promotions, and Amazon’s own ad ecosystem). The halo term captures TikTok Social Ads’ incremental contribution, accounting for delayed effects. An ad this week can still influence sales two weeks later, but with diminishing returns. Priors are calibrated from your brand’s own historical data and anchored to realistic industry ranges, so the model neither blindly trusts platform ROAS nor allows wild speculation.
We fit this model daily per KPI, per brand, per region. The output is a daily table where baseline plus social contribution equals your observed Amazon KPI.
But an MMM is only as good as its safeguards. We built three:
- Calibration: Priors are anchored in your own history and cross checked against what plausible cross platform halo actually looks like. A channel that spent nothing on a given day contributes nothing on that day, so the model’s reporting aligns with the media buyer’s dashboard rather than fighting it.
- Decomposition discipline: Baselines are not fitted freely so that components reconcile exactly. If the model ever tries to credit TikTok with more sales than you actually made, it’s automatically flagged, re-run under tighter specification, and withheld from delivery if it still fails.
- Cross KPI corroboration: We don’t just look at whether sales went up. We also check that the story makes sense across other metrics like page views and search volume. Our composite gate catches cases where a single KPI might fit cleanly but the overall halo story does not hold up across metrics.
What’s Next?
Measuring the halo effect is an ongoing feedback loop. You measure, optimize, and measure again. Our solution tracks shifts as you scale TikTok spend, test new campaigns, or enter new categories.
Rather than optimizing each channel in isolation, the brands that are winning at cross-platform advertising are measuring incremental contribution and reallocating budget accordingly.
TikTok’s halo effect isn’t hidden. It’s just hidden from standard measurement tools.
Ready to understand TikTok’s true impact on Amazon? Schedule a demo of Pacvue’s TikTok Ads solution and see your halo effect for yourself.
