Google’s Smart Bidding will happily report a 40% jump in conversions while your incremental revenue flatlines. That’s not a bug. It’s the entire business model of attribution-based reporting colliding with a question it was never built to answer: would this sale have happened anyway? For brands running automated bidding on Maximized Conversions or Target ROAS, the gap between attribution and incrementality is where marketing budgets quietly go to die.
The Metric You’re Optimizing Isn’t the Metric That Matters
Automated bidding systems are optimization machines. Feed them a goal, and they’ll chase it relentlessly. The problem is the goal itself: conversions, as counted by platform pixels, measure correlation, not causation. Google Ads, Meta’s Advantage+, and TikTok’s Smart Performance Campaigns all default to last-touch or data-driven attribution models that credit the algorithm for conversions that would have happened without any ad spend at all.
Think about branded search. Someone types your company name into Google, clicks a paid ad sitting right above the organic listing, and completes a purchase. Attribution says the ad worked. Incrementality asks a harder question: would they have scrolled down three pixels and clicked the organic result instead? For most retailers, the honest answer is yes, most of the time.
A well-documented pattern in performance marketing: automated bidding tends to over-index on branded search and retargeting audiences because these are the cheapest conversions to “win,” not the most valuable ones to create.
What Maximized Conversions Actually Optimizes For
Google’s Maximized Conversions bidding strategy uses machine learning to spend your entire budget in pursuit of the highest possible conversion volume, based on historical account and auction-time signals. Sounds great on paper. In practice, it means the algorithm will chase low-cost, low-friction conversions wherever it finds them, and that often means existing customers, brand loyalists, and people already deep in the purchase funnel.
That’s not incrementality. That’s efficient harvesting of demand you already created through other channels — organic content, influencer partnerships, email, word of mouth. The algorithm doesn’t know the difference between a new customer and a repeat buyer who was going to reorder anyway. It just knows a conversion fired, and it wants more of those.
This is the same blind spot showing up across marketing measurement broadly. As AI service agents break brand attribution models across the funnel, the industry is being forced to confront how fragile last-click and platform-reported attribution really is.
Incrementality Testing: The Uncomfortable Reality Check
Incrementality testing answers a different question: what happens to conversions if you turn the spend off? Geo-holdout tests, ghost ads, and PSA (public service announcement) conversion lift studies all try to isolate the causal effect of ad exposure by comparing an exposed group against a matched control.
The results are frequently humbling. Meta’s own conversion lift studies have historically shown that a meaningful share of “attributed” conversions would have occurred anyway. Nielsen and various marketing-mix modeling studies over the years have pointed to similar patterns: platform-reported ROAS regularly overstates true incremental return, sometimes by two or three times.
That doesn’t mean the channel is worthless. It means the attribution number on your dashboard is answering “who touched this conversion,” not “did this spend cause this conversion.” Those are very different questions, and automated bidding algorithms are optimized entirely around the first one.
A Quick Gut Check for Your Own Account
- Pull your branded search conversion share. If it’s above 20-30% of total paid conversions, you’re likely paying for demand you’d capture organically anyway.
- Check new-customer vs. returning-customer conversion mix inside Google Ads’ audience segments or Meta’s value optimization reporting.
- Compare platform-reported conversions against a geo or holdout test over a 4-6 week window. The delta is your incrementality tax.
- Look at conversion lag and view-through windows — a 30-day click window with a 1-day view window will inflate attribution differently than a 7-day click window.
Why This Problem Is Getting Worse, Not Better
Automated bidding is becoming more opaque, not less. Google has steadily deprecated granular controls in favor of “trust the algorithm” defaults. Meta’s Advantage+ shopping campaigns pool creative, audience, and placement decisions into a black box that reports strong ROAS numbers while giving advertisers almost no visibility into why a given ad won an auction.
Layer on AI shopping agents — tools like OpenAI’s Atlas, Perplexity’s Comet, and Google’s Gemini-powered shopping flows — and attribution gets even murkier. When an agent completes a purchase on a user’s behalf, which touchpoint gets credit? The checkout behavior differences between these agents already show inconsistent tracking pass-through, which means your attribution data may be missing conversions entirely, or crediting the wrong channel by default.
Identity resolution compounds the issue. As identity resolution gets rebuilt for AI shopping agents, the cookie-based and pixel-based signals that automated bidding relies on are becoming less reliable at exactly the moment brands are leaning harder into automated, “set it and forget it” bidding strategies. It’s a bad combination: less signal fidelity, more algorithmic trust.
If your CFO asks why marketing-attributed revenue keeps climbing while overall company revenue growth is flat, you’re looking at an incrementality problem, not an attribution problem.
Marketing-Mix Modeling as the Corrective Lens
This is exactly why marketing-mix modeling (MMM) has come roaring back into favor, even among brands that swore off it as “too slow” a decade ago. MMM doesn’t rely on individual-level tracking. It looks at aggregate spend and outcome data over time, across channels, and statistically isolates the incremental contribution of each one, including offline and non-clickable channels like influencer seeding and podcast sponsorships.
For influencer and creator spend specifically, this matters enormously, because platform attribution for creator content is notoriously unreliable. A marketing-mix modeling approach for influencer spend can isolate the lift a creator campaign actually drove, independent of whatever last-click credit a retargeting ad grabbed three days later.
The practical move for most mid-size and enterprise brands: run MMM or geo-lift testing quarterly as a check against platform-reported numbers, rather than replacing platform reporting altogether. Use attribution for day-to-day optimization signal. Use incrementality testing for budget allocation decisions. Confusing the two roles is where most performance marketing teams go wrong.
What to Actually Do About It
You don’t need to abandon automated bidding. Maximized Conversions and Target ROAS strategies genuinely outperform manual bidding in many accounts, particularly at scale, where the algorithm can react to auction dynamics faster than any human trafficker could. The fix isn’t turning off automation. It’s putting a governance layer around it.
- Separate branded and non-branded campaigns. Never let an automated bidding algorithm optimize across both pools simultaneously. Branded search should run on manual or a tightly capped ROAS target, since its incrementality ceiling is low.
- Run a holdout test before scaling any Maximized Conversions campaign. A 2-4 week geo holdout will tell you more about true lift than 6 months of dashboard-watching.
- Set a new-customer value modifier. Both Google and Meta allow you to bid differently for new vs. returning customers. Use it. It nudges the algorithm away from harvesting existing demand.
- Audit attribution windows quarterly. Shorter windows generally produce more conservative, more honest numbers. This matters even more as AI Overviews reshape attribution windows for search-driven traffic broadly.
- Report incrementality alongside attribution, not instead of it. Give stakeholders both numbers. The gap between them is a KPI in its own right.
Platforms have every incentive to keep attribution generous. Google, Meta, and TikTok get paid regardless of whether a conversion was incremental or not (TikTok Ads, Meta Business, and Google Ads Help all document their default attribution models, but none of them will proactively tell you those defaults inflate perceived performance). Independent measurement is on you.
Third-party data on this trend keeps accumulating. eMarketer and Statista have both tracked growing advertiser skepticism toward platform-reported ROAS over the past several ad cycles, and HubSpot‘s benchmarking research on marketing attribution echoes the same conclusion: the tools measuring performance are rarely neutral about the performance they’re measuring.
The Takeaway
Next quarter, before you let an automated bidding campaign scale further, run a four-week geo holdout against your top-spending campaign and compare the lift to what the dashboard claims. If the gap is wider than 20%, you’re not funding growth — you’re funding a more expensive way to report on demand you already had.
FAQs
What’s the difference between attribution and incrementality?
Attribution assigns credit for a conversion to a specific touchpoint or channel based on tracking data. Incrementality measures whether that conversion would have happened anyway, using controlled tests like geo holdouts or lift studies. Attribution tells you what touched the customer; incrementality tells you what caused the outcome.
Why does Maximized Conversions inflate conversion counts?
Because the bidding strategy is designed to spend budget wherever conversions are cheapest and most likely, which often means branded search, retargeting pools, and existing customers who were likely to convert regardless of ad exposure. The algorithm optimizes for conversion volume, not incremental lift.
How do I test incrementality without a data science team?
Geo holdout tests are the most accessible option: pause spend in a set of matched regions while continuing normally elsewhere, then compare conversion rates. Google, Meta, and most MMM vendors offer templated conversion lift or geo-testing tools that don’t require custom statistical modeling.
Should brands stop using automated bidding altogether?
No. Automated bidding often outperforms manual bidding at scale. The fix is adding governance: separate branded from non-branded campaigns, apply new-customer value modifiers, and validate performance periodically with holdout tests rather than trusting platform-reported ROAS alone.
How often should brands run incrementality tests?
Quarterly is a reasonable cadence for most mid-size to enterprise advertisers, with additional tests triggered any time you scale spend significantly, launch a new campaign type, or notice a widening gap between attributed and actual business results.
FAQs
What’s the difference between attribution and incrementality?
Attribution assigns credit for a conversion to a specific touchpoint or channel based on tracking data. Incrementality measures whether that conversion would have happened anyway, using controlled tests like geo holdouts or lift studies. Attribution tells you what touched the customer; incrementality tells you what caused the outcome.
Why does Maximized Conversions inflate conversion counts?
Because the bidding strategy is designed to spend budget wherever conversions are cheapest and most likely, which often means branded search, retargeting pools, and existing customers who were likely to convert regardless of ad exposure. The algorithm optimizes for conversion volume, not incremental lift.
How do I test incrementality without a data science team?
Geo holdout tests are the most accessible option: pause spend in a set of matched regions while continuing normally elsewhere, then compare conversion rates. Google, Meta, and most MMM vendors offer templated conversion lift or geo-testing tools that don’t require custom statistical modeling.
Should brands stop using automated bidding altogether?
No. Automated bidding often outperforms manual bidding at scale. The fix is adding governance: separate branded from non-branded campaigns, apply new-customer value modifiers, and validate performance periodically with holdout tests rather than trusting platform-reported ROAS alone.
How often should brands run incrementality tests?
Quarterly is a reasonable cadence for most mid-size to enterprise advertisers, with additional tests triggered any time you scale spend significantly, launch a new campaign type, or notice a widening gap between attributed and actual business results.
Top Influencer Marketing Agencies
The leading agencies shaping influencer marketing in 2026
Agencies ranked by campaign performance, client diversity, platform expertise, proven ROI, industry recognition, and client satisfaction. Assessed through verified case studies, reviews, and industry consultations.
Moburst
-
2
The Shelf
Boutique Beauty & Lifestyle Influencer AgencyA data-driven boutique agency specializing exclusively in beauty, wellness, and lifestyle influencer campaigns on Instagram and TikTok. Best for brands already focused on the beauty/personal care space that need curated, aesthetic-driven content.Clients: Pepsi, The Honest Company, Hims, Elf Cosmetics, Pure LeafVisit The Shelf → -
3
Audiencly
Niche Gaming & Esports Influencer AgencyA specialized agency focused exclusively on gaming and esports creators on YouTube, Twitch, and TikTok. Ideal if your campaign is 100% gaming-focused — from game launches to hardware and esports events.Clients: Epic Games, NordVPN, Ubisoft, Wargaming, Tencent GamesVisit Audiencly → -
4
Viral Nation
Global Influencer Marketing & Talent AgencyA dual talent management and marketing agency with proprietary brand safety tools and a global creator network spanning nano-influencers to celebrities across all major platforms.Clients: Meta, Activision Blizzard, Energizer, Aston Martin, WalmartVisit Viral Nation → -
5
The Influencer Marketing Factory
TikTok, Instagram & YouTube CampaignsA full-service agency with strong TikTok expertise, offering end-to-end campaign management from influencer discovery through performance reporting with a focus on platform-native content.Clients: Google, Snapchat, Universal Music, Bumble, YelpVisit TIMF → -
6
NeoReach
Enterprise Analytics & Influencer CampaignsAn enterprise-focused agency combining managed campaigns with a powerful self-service data platform for influencer search, audience analytics, and attribution modeling.Clients: Amazon, Airbnb, Netflix, Honda, The New York TimesVisit NeoReach → -
7
Ubiquitous
Creator-First Marketing PlatformA tech-driven platform combining self-service tools with managed campaign options, emphasizing speed and scalability for brands managing multiple influencer relationships.Clients: Lyft, Disney, Target, American Eagle, NetflixVisit Ubiquitous → -
8
Obviously
Scalable Enterprise Influencer CampaignsA tech-enabled agency built for high-volume campaigns, coordinating hundreds of creators simultaneously with end-to-end logistics, content rights management, and product seeding.Clients: Google, Ulta Beauty, Converse, AmazonVisit Obviously →