Marketing Attribution Models

Attribution is overrated. Incrementality is what actually matters Every new-age brand wants to know what’s working. Meta ROAS is looking good. CAC is steady. Revenue is growing But here’s the truth: Your Meta ad might get the conversion. But did it cause the conversion? That’s the difference between attribution and incrementality. Most dashboards, attribution tools, and agency reports stop at attribution. But if you’re a brand selling across Amazon, Flipkart, GT, MT, Q-com, and D2C—pure attribution will always lie to you Because the sale might happen on Amazon. But it might have been nudged by a Meta video or a YouTube bumper ad 4 days ago. You don’t need a full-blown Marketing Mix Model to get started. There are simpler, street-smart ways to directionally understand what’s working—and what’s not. Here are 4 that have worked for us at Atomberg: 1. Geo Split Testing Pick two similar markets. Run campaigns in one. Don’t run in the other. Then track: • Branded search volume • Sell-through on marketplaces • Secondary sales from GT counters If the test market moves faster than the control, you’re seeing true lift. That’s incrementality. 2. First-Time Buyer Growth vs Returning Buyer Growth Track whether your growth is coming from first-time buyers or repeats. If your campaigns are just bringing back old customers—you’re not creating net new demand. But if there’s a spike in new buyers across Amazon, Flipkart, D2C—your campaigns are likely working at an incremental level 3. Paid Traffic vs Organic Trend Lines If paid traffic, clicks and spends are going up—but your organic sales or branded search isn’t moving—you’re likely just harvesting demand that already existed. But if organic lifts alongside paid—your ads are creating interest. Not just closing it. Directionally, this is one of the simplest sanity checks most teams ignore. 4. Channel Crossover + Offline Signal Mapping Your Meta ad may not show up in last-click attribution. But it might have nudged the consumer to visit your store or buy on Amazon. You can detect this through: • Post-purchase surveys (Where did you first hear about us?) • Branded search + store footfall spikes in campaign-active cities • And most powerfully—offline signals passed back to Meta At Atomberg, we pass back data from installations and warranty registrations—including pincode and purchase timelines Sometimes, we’re even able to identify this at a unique customer level through their cookies for warranty registration This has helped us understand true incrementality of perf marketing campaigns even for offline sales If you’re only measuring ROAS, you might scale what’s only taking credit for sale about to happen anyway If you chase incrementality, you’ll scale what’s working. For more details, read the full post- link in first comment.

Head of Marketing: We're turning off all attribution tracking. CEO: Now I know you've lost it! Explain.. Head of Marketing: Hear me out. Attribution is why we're losing. CEO: We need to know what's working. Head of Marketing: We know exactly what's working. We just refuse to believe it. 73% of our closed deals touched 8+ marketing assets. Our attribution gives 100% credit to the last click - usually a brand search. CEO: So fix the attribution model. Head of Marketing: I did. Six times. Multi-touch, linear, time-decay, custom ML model. You know what happened? We spent more time debating the model than doing actual marketing. CEO: But how do we optimize spend? Head of Marketing: By talking to customers. Novel concept, right? Last week I called 20 closed deals. Not one mentioned the channel we credit. They all mentioned that one LinkedIn post from 6 months ago that made them rethink their entire workflow. CEO: The board wants numbers. Head of Marketing: The board wants revenue. Our competitor grew 300% YoY. Their attribution? "Marketing works when you do good marketing." They invest in what customers actually talk about, not what pixels claim. CEO: This is insane. Head of Marketing: You know what's insane? We killed our podcast because attribution said zero ROI. Three months later, our biggest enterprise deal told us they binged all 40 episodes before reaching out. CEO: Sales will revolt. Head of Marketing: Sales already agrees. They're tired of leads who hit 47 touchpoints but have zero intent. They want the one person who read our deep dive and is ready to buy. CEO: One quarter. That's it. Head of Marketing: Deal. But when revenue jumps 50%, I want that podcast back. CEO: We'll see. PS - The best marketers measure what matters, not what's measurable. Sometimes the most important touch is the one you'll never track. I'm Chris Cunningham - I run social media at ClickUp. Follow me for more actionable marketing tips & tricks.

Attribution in marketing is mostly bullshit. We’ve become obsessed with modelling every click, view, and conversion like we’re tracking particles in a lab. But here’s the truth, most buying journeys are chaotic, messy, and totally untrackable. Case in point: Whilst in Vegas at the start of the month, sat at brunch, one of those mobile billboard trucks rolled past. It was advertising tickets to a college basketball game. No QR code. No CTA. Just a truck with a giant sign. But it caught my eye. So I Googled it. Found the ticket site and got distracted when my food arrived (friend chicken on waffles with extra syrup for anyone interested). Later, I’m scrolling Instagram and get hit with a retargeting ad. I click. I buy. Guess who takes the credit? Instagram. Paid Social. They’ll report it as a “conversion,” like they owned the journey. But they didn’t. The truck did. That very physical, analog, unmeasurable truck is what sparked the whole thing. And that’s the problem. Marketing attribution isn’t science. We need to stop pretending we can track and model everything. You can’t map human behaviour with a spreadsheet. Not everything that works is measurable. And not everything that’s measurable matters.

ChatGPT just hit 800 million weekly active users, up from 500 million in March. AI discovery is the fastest-growing channel to reach your customers. But more than that, it forces us to rethink buyer journeys for AI discovery. The team at Docebo told me AI discovery now accounts for 12.7% of their high-intent leads, up 429% (!) from last year. They're doing this with one dedicated team member (Valeriia) plus an assist from AI tools like AirOps & XFunnel. Here's the thing: ChatGPT leads don't look like SEO or referral leads. They show up ready to buy, either going to your website directly (*after* they've made a decision) or searching for your brand on Google. Branded search now accounts for 85% (!) of Docebo's search traffic -- meaning 85% who come from Google already know about the company. ▶️ The old buyer journey: Problem awareness > feature research > vendor comparison > brand selection. ▶️ The new buyer journey: AI-assisted research > network validation > direct brand search > selection. Read more in Growth Unhinged about what this means & how to get more pipeline from AI search: https://lnkd.in/ee2wKTbR The TL;DR: 1. Measure your AI search performance through self-reported attribution, branded search traffic, and share-of-voice in AI queries. 2. Late-funnel SEO is the foundation for great answer engine optimization (AEO). 3. The dirty work of AEO has to get done: updating, formatting, and structuring content for LLMs. 4. You can see tangible AEO results with a modern tech stack. #aeo #ai #seo

B2B executives don’t realize that department-level source attribution is the *cause* of their pipeline problem, not the solution. I sat in yet another meeting yesterday where Marketing, Marketing Ops, and Biz Dev leaders debated which attribution model they should use…..5 minutes in, I wanted to blow my brains out ☠️ I just can't take it anymore. These debates are useless. The SDR leader wants a first-touch attribution model so SDRs get credit when they load leads into the database, citing that the SDRs get “demoralized” when they book a meeting that gets tagged as inbound. The Marketing team wants it to be “Marketing-sourced” if there’s any Marketing touch within 45 days. The Partner team wants it to get tagged as “Partner” and take precedence over everything when a Partner is involved. Every department just fights over credit for sourcing the lead, every department celebrates “sourcing” a % of revenue, while the financial planning model breaks down IRL and the company misses growth targets by a mile. Then - and this is the hilarious part - they try to fix the problem by changing the attribution model 😭 😭 It’s like admitting you have a drug problem, and then deciding to switch from cocaine to meth. So, to reinforce the fundamental flaws of this approach: 1. Using lead source attribution credit by department, regardless of which attribution model you use, is a terribly sub-optimal way to plan, measure, and optimize your GTM when you need to deliver scalable growth aka sustainable unit economics. 2. Comparing Demand/Marketing (triggers and detects buying signals) with Prospecting (engage buyers 1:1 and book qualified meetings/opps) is like comparing apples to monkeys. They are NOT the same thing and any logical human would arrive at this conclusion with a small amount of critical thinking. 3. Using this incredibly short-sighted and simplistic planning model forces all Marketing investments to appease the attribution model - skewing investments all to 1 part of the buyer journey (database capture) and causing extreme self-inflicted pain & dysfunction in the Pipeline Creation process. 4. We should be measuring, comparing and optimizing the effectiveness of our Supply Chain (all of the 1st-party and 3-party signals that enter our Revenue Factory for Prospecting). 5. We should be measuring, comparing, and optimizing the effectiveness of our Prospecting (all of our efforts to engage buyers 1:1 and create qualified opportunities, including AI, SDRs, and AE prospecting motions). ___ If you spend just a small amount of time thinking about the current situation, you’ll too see that all of these debates & meetings & tech projects are a total waste of time. It's time for C-Level executives to fix the real problem. Stop planning, measuring, and optimizing your GTM like 4 low-quality department-level assembly lines. And start running your GTM like a Unified Revenue Factory. #b2b #gtm #marketing #sales #finance

OpenAI’s Commerce announcement today is a turning point for how we think about earned media, trust and ROI in PR. Comms leaders have long argued that media coverage, brand reputation and word-of-mouth validation are foundational to long-term influence. But until recently, it’s been challenging to causally connect earned media to direct purchase behavior. We know that earned media fuels the vast majority of LLM citations, and now ChatGPT has removed any friction on the path of discovery to checkout. Here’s how we’re thinking about it at Muck Rack: 1. Earned media immediately becomes more actionable. Mentions, features, reviews and product roundups in trusted media are now direct gateways to transactions. A consumer reading a review on a product in ChatGPT might click “Buy” directly, turning a story into a sale. That shifts the balance: media is no longer just awareness or credibility– it becomes part of the purchase funnel. 2. Attribution and ROI gain new clarity in PR. Historically, PR’s impact on revenue has been indirect and hard to trace. With agentic commerce, we can quantify how many media-driven discussions led directly to completed checkouts. That means PR investments can be benchmarked against tangible outcomes, more like paid or performance marketing. 3. Comms plays a direct role in commerce. PR teams can identify which mention types (e.g., “best of,” feature comparisons, product spotlight, gift guides) are likely to convert in a checkout-enabled environment. Pitch with an eye toward conversion: think ahead to how media content can feed into discovery inside ChatGPT agents. 4. Measure reach, sentiment, share-of-voice AND conversion lift, checkout volume and per-mention ROI. This is where leveraging Google Analytics with your earned media measurement tooling can offer great insights.

If more of your store sales start on TikTok lately, you might wanna read this. 𝘛𝘩𝘦 𝘴𝘢𝘭𝘦 𝘪𝘴 𝘥𝘦𝘤𝘪𝘥𝘦𝘥 𝘣𝘦𝘧𝘰𝘳𝘦 𝘺𝘰𝘶𝘳 𝘤𝘶𝘴𝘵𝘰𝘮𝘦𝘳 𝘦𝘷𝘦𝘯 𝘦𝘯𝘵𝘦𝘳𝘴 𝘺𝘰𝘶𝘳 𝘴𝘵𝘰𝘳𝘦. The checkout happens in-store. But the sale happens everywhere else. Here's the reality: This year 60%+, and in 2027, 70% of retail sales will be digitally influenced. I can't emphasize this enough; here's what most brands miss—digital influence isn't just about online sales. It's about shaping every moment before the customer even walks into your store. L'Oréal cracked this code: 100M+ AR try-on sessions driving real conversions. 31 brands orchestrating seamless experiences across 72 countries. No.1 in beauty influencer marketing (29% market share), 20-80% higher conversion rates through enhanced digital experiences. The new customer journey isn't linear—it's layered: - They discover you on social - Research you through reviews and UGC - Try your product virtually through AR - Get retargeted with personalized content - Finally purchase in-store (feeling confident they're making the right choice) Every touchpoint matters, and every interaction influences the final decision. The brands winning today aren't just selling products—they're orchestrating experiences across owned, paid, and earned media that guide customers from curiosity to checkout. Digital discovery is increasingly pay-to-play and shoppers are paying attention. ++ Tactical Recommendations for CPG / FMCG Brands ++ 1. Beyond just having perfect, high SOV product pages, create discovery ecosystems. - Optimize for "zero-moment-of-truth" searches. - Activate shoppable content at scale. - Leverage user-generated content as social proof. Brands that do these see a 35% higher conversion rate from digital touchpoints to in-store purchases. 2. Connect digital engagement directly to retail execution. - Geo-target digital campaigns to drive foot traffic - Create "store-specific" digital content CPG brands using geo-targeted social ads see a 23% higher in-store sales lift in targeted markets. 3. Most important one; stop flying blind—measure digital influence on offline sales. - Implement unique promo codes for each digital touchpoint to track conversion paths. - Use customer surveys at point of purchase. - Partner with retailers on shared data insights Brands with proper attribution see 15-25% improvement in marketing ROI within 12 months. 𝗧𝗼 𝗮𝗰𝗰𝗲𝘀𝘀 𝗮𝗹𝗹 𝗼𝘂𝗿 𝗶𝗻𝘀𝗶𝗴𝗵𝘁𝘀 𝗳𝗼𝗹𝗹𝗼𝘄 ecommert® 𝗮𝗻𝗱 𝗷𝗼𝗶𝗻 𝟭𝟰,𝟲𝟬𝟬+ 𝗖𝗣𝗚, 𝗿𝗲𝘁𝗮𝗶𝗹, 𝗮𝗻𝗱 𝗠𝗮𝗿𝗧𝗲𝗰𝗵 𝗲𝘅𝗲𝗰𝘂𝘁𝗶𝘃𝗲𝘀 𝘄𝗵𝗼 𝘀𝘂𝗯𝘀𝗰𝗿𝗶𝗯𝗲𝗱 𝘁𝗼 𝗲𝗰𝗼𝗺𝗺𝗲𝗿𝘁® : 𝗖𝗣𝗚 𝗗𝗶𝗴𝗶𝘁𝗮𝗹 𝗚𝗿𝗼𝘄𝘁𝗵 𝗻𝗲𝘄𝘀𝗹𝗲𝘁𝘁𝗲𝗿. #CPG #FMCG #AI #ecommerce Procter & Gamble PepsiCo Unilever The Coca-Cola Company Nestlé Mondelēz International Kraft Heinz Ferrero Mars Colgate-Palmolive Henkel Bayer Haleon Kenvue The HEINEKEN Company Carlsberg Group Philips Samsung Electronics Panasonic North America

Perhaps the key to fixing our perennial attribution problem in B2B isn't to focus on "marketing sourced pipeline" but rather to eliminate that metric altogether. In complex selling situations, first-touch is a joke. A mere starting point, a blip on the journey. Last-touch is icing. It's the culmination of countless other touches, activities and influences. There is no marketing-sourced or sales-sourced, there's just we-sourced. That's an inconvenient truth for those that want a cleaner dashboard. The reality is that buying journeys are extremely messy, and a body of work mentality is required by integrated go-to-market teams to lasso that behavior into any sort of predictable, repeatable pipeline. For organizations that still worship at the altar of the almighty MQL, establishing a marketing-sourced pipeline goal may feel like a step in the right direction. And absolutely the farther you take accountability deeper into the pipeline the better. Sourced doesn't matter nearly as much as velocity. Consensus-building. Commitment to change. Measured, predictable and repeatable sequencing of cross-channel and cross-team motions that drive engagement and conversion. It's messier. And it's far more effective.

🔎 Just released on Water & Music: A brand-new, 4,500-word report on how music AI attribution actually works. -- The state of music AI today can be measured in the millions — millions of dollars raised, millions of users engaged, millions of AI-generated tracks uploaded to streaming services. As the market continues to grow at breathtaking speed, an urgent business question has emerged: Who deserves credit and compensation when AI generates music? The answer could determine how billions in potential revenue flow through the music economy in the coming years. The core concept underpinning the ideal of a more granular rights holder compensation framework for AI is ATTRIBUTION, or the linking of AI outputs back to the specific training inputs that influenced them. Unlike copyright or deepfake detection systems that simply flag similarity or infringement, attribution aims to establish causal relationships between inputs and outputs. For music, attribution can help answer questions like: ❓ Which songs in the training data most influenced this output? ❓ How strong was each influence, and how should we quantify it? ❓ What specific musical elements were borrowed or transformed? -- Yung Spielburg, Alexander Flores, and I spent the last several months on intensive research and interviews with teams forging new ground in music AI licensing and attribution deals, including Sureel AI, Musical AI, Soundverse AI, Lemonaide Music, LANDR, SourceAudio, Human Native, and more. A preview of some of the insights we covered: 💡 No "perfect" attribution system currently exists, but this hasn't stopped deals from happening. Perfection isn't a prerequisite for business. 💡 AI music generation is not sampling. Models don't copy fragments outright from training data, but rather absorb concepts on a more abstract level across multilayered architectures, making attribution exponentially more complex. 💡 Every attribution system makes inherent value judgments about which musical elements deserve (or don't deserve) compensation, potentially shaping revenue distribution for years to come. 💡 Tech alone won't solve the market coordination problem. Trust between rights holders and AI developers matters as much as technical innovation; a technically perfect solution is practically useless without industry adoption. -- Understanding the realities of attribution isn't merely a matter of technical curiosity — it's business critical for anyone working with music rights today, and will determine how the music industry adapts and thrives in an AI-centric future. Read our full analysis here (members only): https://lnkd.in/eBnHAEGn #MusicTech #AIMusic #MusicBusiness #MusicAI #MusicIndustry #Interpretability #Explainability #XAI

VP Growth: Meta just wants you to use their in-platform attribution.  CMO: [slacks the below summary and link to the Meta paper debunking that myth]  VP Growth: [5 min later] We need to change the way we measure asap. – Summary & Takeaways: - Meta's latest paper on measuring ad effectiveness Incrementality as north star. Incrementality = driving a purchase from someone who would not have already purchased - If ad dollar is not incremental, it's wasting money—paying for a transaction that would have already happened - To maximize incrementality, use more than simple attribution tools - Need statistical modeling (MMM) and incrementality experiments - Those 3 things (attribution + statistical modeling + incrementality experiments) are the ideal triumvirate of measurement tools to maximize actual revenue and profit growth from ads - Expect conflicting results. This is a consistent learning journey, not one-and-done where conflicts cause us to throw babies out with bathwater. Commitment to consistent testing where learning and calibrating are expected - Use marginal return as the guide—constantly measure how much more revenue you get or lose when adding or subtracting dollars from a strategy, tactic, or channel - For diminishing lower-funnel returns, adding broader reach (e.g., non-purchase conversion campaigns) drives more incremental purchases at higher marginal return ($1.50+ revenue per $1 spent vs. <$1.50) Actions: - Start using MMM and incrementality experiments. Bad setups (wrong geos, insufficient time/spend) can be damaging. This is a learning journey - Test incremental attribution—don't be surprised if ROAS is lower. Doesn't necessarily mean 'it doesn't work', could mean previous attribution inflated ROAS by claiming credit for transactions that would have happened anyway - Measure impact using marginal return. Pulse spend up/down to understand revenue response. Results evolve over time/season/stage - Over-invest in measurement tools—tiny % of ad budgets but dramatically reduces money you light on fire -- Full paper: https://lnkd.in/gXDejV-G