A 39 Percent ROAS Gain Should Change How You Think About GEM Budget
If you’re still treating GEM as experimental line-item spend, these numbers demand a rethink. Jellyfish’s documented results — a 17% CTR lift, 14% conversion boost, and 39% ROAS gain for Gentle Monster, plus a 45% revenue jump for MSC Industrial — represent some of the most concrete early-performance data the industry has seen from Google’s AI-powered campaign format.
The question isn’t whether GEM works. It’s whether your team is set up to capture those gains — or whether you’ll allocate budget, see middling results, and blame the channel.
What GEM Actually Is (and Why the Benchmark Math Is Different)
Google’s Generative Experience Marketing — GEM — uses multimodal AI to dynamically assemble ad creative, audience signals, and bidding logic in real time. It’s not a placement. It’s a system. That distinction matters enormously when you’re setting ROI expectations, because traditional benchmarking logic — CPM, CTR by creative variant, ROAS against a fixed media plan — doesn’t map cleanly onto how GEM operates.
GEM pulls from first-party data signals, product feeds, audience intent layers, and creative assets simultaneously. Performance is emergent, not engineered upfront. Which means the brands seeing the biggest lifts are the ones who walked in with strong asset libraries, clean conversion tracking, and realistic ramp timelines — not the ones who expected instant returns.
GEM performance is emergent, not engineered. The brands seeing 39% ROAS gains fed the system well: clean data, strong creative assets, and enough runway to let the AI learn.
For more on how AI-native campaign structures are reshaping brand team responsibilities, see our coverage of AI-native campaign orchestration readiness — the same operational gaps that trip up AI search also surface in GEM deployments.
The Jellyfish Numbers, Unpacked
Jellyfish, one of Google’s most closely-watched agency partners, ran GEM across two very different verticals. That’s the part worth slowing down on.
Gentle Monster is a fashion-forward eyewear brand with high creative equity, strong visual identity, and an audience that skews aspirational. Their 17% CTR lift and 14% conversion boost reflect what happens when a brand with rich, high-quality visual assets gives GEM enough variety to work with. The 39% ROAS gain is remarkable, but contextually it makes sense: GEM’s generative assembly favors brands where creative diversity and audience intent signals are both strong.
MSC Industrial Direct is the opposite profile — B2B, industrial supply, lower visual creativity ceiling, but an extremely well-structured product catalog and first-party data from decades of customer relationships. Their 45% revenue jump suggests GEM’s AI doesn’t need glamour to perform. It needs signal density. MSC had it in abundance.
Two different verticals, two different asset profiles, both showing significant gains. That’s not a coincidence. It’s a pattern that should inform how brand teams diagnose their own readiness before committing GEM budget.
What This Means for ROI Expectation-Setting
Most brand teams approach new channel budgeting the same way: find an industry benchmark, apply a conservative multiplier, build a 90-day projection, and present it to finance. That process breaks down with GEM for three reasons.
- GEM has a learning curve that eats early performance. The first three to four weeks of a GEM campaign are often the AI’s calibration period. Brands that evaluate GEM on month-one ROAS are measuring the wrong window. Jellyfish’s reported results likely reflect post-ramp performance, not week-two numbers.
- Asset quality is a hidden variable in every benchmark. A 39% ROAS lift from Gentle Monster doesn’t transfer to a brand feeding GEM three static images and a product description. The system scales what you give it. Garbage in, mediocre out.
- Attribution model mismatches inflate skepticism. GEM’s AI-driven bidding often captures assist touches and mid-funnel interactions that last-click attribution frameworks miss entirely. If your measurement stack runs on last-click, your GEM numbers will look worse than they are. Fix the measurement model before you judge the channel.
The GEO versus GEM distinction is also worth understanding here — particularly for teams managing both AI search visibility and paid performance simultaneously. Our analysis of GEO vs GEM full-funnel strategy covers how these two AI-era disciplines interact at the CMO level.
The Readiness Audit You Should Run Before Allocating Budget
Before you take these benchmark numbers to your CFO, run an honest internal audit across four dimensions.
First-party data quality. MSC Industrial’s 45% revenue jump wasn’t luck — it was the payoff of years of structured customer data. If your CRM data is fragmented, your GEM performance ceiling is lower than the benchmarks suggest. Platforms like Meta Business and Google Ads Help both document how data quality affects AI campaign performance — the same logic applies here.
Creative asset depth. GEM generates variations, but it works from what you supply. Audit whether you have enough format diversity — video, static, product-focused, lifestyle-focused — for the AI to work with meaningfully. If you’re asset-thin, budget for production before you budget for media spend.
Measurement infrastructure. Are you running data-driven attribution, or are you still on last-click? GEM’s performance will look artificially weak under legacy attribution models. Tools like EMARKETER have documented the attribution gap in AI-driven campaigns extensively.
Internal team capability. GEM campaigns require ongoing creative refresh, data hygiene, and signal optimization. They are not set-and-forget. If your team doesn’t have capacity for active management, you’re not ready to scale GEM — you’re ready to test it at a controlled budget floor. Understanding how AI-native team roles are evolving helps frame who owns this work internally.
Before presenting GEM benchmarks to finance, audit your first-party data quality, creative asset depth, attribution model, and internal capacity. Missing any one of these will drag performance below what the Jellyfish numbers suggest is possible.
Vertical Signals Worth Watching
The Gentle Monster and MSC Industrial cases together suggest GEM is vertically agnostic in a way that other AI ad formats haven’t been. Fashion and industrial supply shouldn’t both be posting significant gains from the same campaign architecture. But they did.
That said, the specific levers differ. For consumer brands in fashion, beauty, and lifestyle, creative variation and audience segmentation do the heaviest lifting. For B2B and catalog-heavy verticals, product feed structure and first-party audience data matter more. Brands in verticals with long consideration cycles — think B2B software, high-ticket consumer goods, financial services — should expect GEM’s value to concentrate in mid-funnel engagement metrics before it surfaces in ROAS. The 45% revenue figure for MSC Industrial likely reflects a pipeline that was already partially warm.
For brands experimenting with content format ROI by vertical, the GEM data adds another layer: format performance within GEM doesn’t follow the same hierarchy as organic or traditional paid. Video may dominate in some verticals; dynamic product ads outperform in others. Test before assuming.
How to Structure Your GEM Budget Allocation
Given the early-stage nature of GEM’s benchmark data, a tiered allocation approach is more defensible than a full commitment.
- Test phase (weeks 1–6): Allocate 10–15% of a relevant campaign budget to GEM. Don’t evaluate on ROAS alone. Track CTR trends, audience segment performance, and creative variant engagement. This is your signal-gathering phase.
- Optimization phase (weeks 7–12): Feed back what you learned. Refresh creative, refine audience data inputs, and adjust attribution modeling if necessary. ROAS should start approaching meaningful territory here if your readiness factors are solid.
- Scaling phase (month 4+): If ROAS is trending above your baseline and creative refresh cadence is sustainable, scale incrementally. The Jellyfish numbers suggest the ceiling is high — but ceiling and floor aren’t the same number.
Resources like HubSpot and Statista track broader AI marketing adoption curves that provide useful context for how early-mover advantage plays out in channel adoption cycles — GEM follows a similar pattern.
One more thing: tie your GEM investment to a parallel creator content strategy where possible. GEM’s generative assembly performs better when brand signals are strong across channels. If your organic and creator-driven content is building cultural relevance and search intent, your GEM campaigns inherit that signal density. Brands that treat paid AI and creator programs as separate silos are leaving performance on the table. For context on how creator-driven signals feed paid performance, see our piece on micro-communities and creator trust.
Bottom line: The Jellyfish GEM data is real, but it’s a ceiling, not a floor. Run the readiness audit, fix your attribution model, and build the asset depth before you build the budget case.
Frequently Asked Questions
What is GEM and how is it different from standard Google ad campaigns?
GEM (Generative Experience Marketing) is Google’s AI-powered campaign format that dynamically assembles ad creative, bidding logic, and audience signals in real time using multimodal AI. Unlike traditional Google campaigns where advertisers control individual placements and creative variants, GEM uses generative AI to optimize and combine inputs continuously. Performance is therefore emergent and improves over time as the system learns from first-party data and audience behavior, rather than being fixed by upfront creative decisions.
Are the Jellyfish GEM results applicable to all brands?
Not automatically. The 17% CTR lift, 14% conversion boost, 39% ROAS gain for Gentle Monster, and 45% revenue jump for MSC Industrial reflect brands that entered GEM with strong first-party data, deep creative asset libraries, and proper attribution infrastructure. Brands lacking these inputs should expect lower initial performance. The results set an aspirational benchmark, not a guaranteed baseline, and readiness factors significantly affect where a brand’s performance will land.
How long does it take for GEM to show meaningful ROAS results?
Most practitioners and agency teams working with GEM report a calibration period of three to six weeks before ROAS data becomes meaningful. Early-phase measurement should focus on CTR trends, creative variant performance, and audience segment behavior rather than ROAS alone. Evaluating GEM on month-one ROAS is a common mistake that leads to premature budget cuts before the system has learned enough to optimize effectively.
What attribution model should brands use for GEM campaigns?
Data-driven attribution is strongly recommended for GEM. Last-click attribution models systematically undervalue GEM’s contribution because the format excels at capturing mid-funnel intent signals and assist touches that last-click frameworks don’t credit. Brands running GEM on last-click attribution will routinely see performance appear weaker than it actually is, which can lead to incorrect budget decisions. Auditing and updating your attribution model before launching GEM is a prerequisite, not an afterthought.
How should B2B brands approach GEM budget allocation differently from B2C brands?
B2B brands should expect GEM’s value to surface first in mid-funnel engagement metrics — qualified traffic, lead quality, time-on-site — before it translates into direct revenue ROAS. MSC Industrial’s 45% revenue gain reflects a brand with a very well-structured product catalog and decades of first-party customer data. B2B brands with clean CRM data and structured product feeds are well-positioned for GEM, but should build longer evaluation windows into their performance expectations, typically 90 days rather than 30, given longer B2B consideration cycles.
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 →