Predictive Modeling for Conversion Improvement

What if you could predict which users are actually valuable before they convert? Most performance marketing strategies focus on what’s already happened - who clicked, who converted, and how much they spent. But what if you could optimise campaigns based on what will happen? Well that’s exactly what propensity models enable. By analysing user behaviour and intent signals, we can predict the likelihood of a conversion - allowing brands to make smarter, faster decisions across paid search and social. Understanding what a Propensity Model is A propensity model is a machine learning approach that predicts how likely a user is to take a specific action - whether it’s making a purchase, signing up, or returning to your site. Instead of treating all users the same, it helps advertisers: ✅ Identify high-value users before they convert ✅ Adjust bids dynamically based on predicted value ✅ Prioritise ad spend toward users who are more likely to convert Why Does This Matter? Ad platforms like Google and Meta rely on past conversion data. But for brands with long purchase cycles, waiting weeks or months for that actual revenue to come in isn’t practical. With propensity modelling, we estimate conversion value earlier and feed that data directly into bidding algorithms—enabling real-time optimisation. How It Works: 1️⃣ Data Collection – Analyse behavioural signals (session length, page views, interactions, historical purchases, etc). 2️⃣ Model Training – Machine learning identifies patterns that indicate conversion likelihood. 3️⃣ Real-Time Scoring – Every user gets a propensity score, predicting their likelihood to convert. 4️⃣ Activation in Paid Media – These scores are pushed to ad platforms, dynamically adjusting bids based on predicted value. Some results: Over the past 12 months, some brands using propensity models that we have built have seen ROI increase by 40% and conversion volume grow by 150% - driving significantly higher revenue at improved efficiency. But propensity modelling isn’t just for performance marketing. Its insights can help predict total future customer value and inform CRM, communication strategies, financial modelling, and beyond. Behavioural Insights The screenshot below is an example of a behavioural importance analysis, showing which user actions influence future value most. How to interpret the plots: - Each point represents a user record. - X-axis (SHAP Value): Left = lower probability of conversion, Right = higher probability. - Colour Scale: Blue = lower impact, Red = higher impact. Key takeaways - Propensity models provide a critical data point for understanding future customer value. - Integrating these signals into ad platforms can give brands a major advantage in bidding. - Their applications extend beyond performance marketing—impacting CRM, financial modelling, and overall business strategy.

Most teams think of predictive models as answer machines. You ask a question, you get a score. Will this customer churn? How likely is this lead to convert? That's valuable. But it's only half the story. The real gold sits in the SHAP values behind each prediction. SHAP (SHapley Additive exPlanations) breaks down exactly which variables pushed a prediction in each direction, for every single customer or record. Not just "this customer is likely to churn," but why. Was it their purchase frequency? The channel they came from? How much revenue they've generated? Think of it as doing BI on AI. When you analyze SHAP values with a business lens, you stop looking at individual predictions and start seeing patterns. You can identify entire segments that share the same risk drivers. Maybe your high-revenue customers from one acquisition channel are three times more likely to leave than those from another. That's not just a prediction. That's a strategy. This is one of the most overlooked benefits of having a strong predictive model in place. The predictions tell you what's coming. The SHAP values tell you what to do about it. Want to go deeper? Here's a solid breakdown of how SHAP values work: https://lnkd.in/dxMMyhFH

𝐅𝐨𝐫 𝐲𝐞𝐚𝐫𝐬, 𝐦𝐚𝐫𝐤𝐞𝐭𝐢𝐧𝐠 𝐫𝐚𝐧 𝐨𝐧 𝐡𝐢𝐧𝐝𝐬𝐢𝐠𝐡𝐭. Dashboards told us what already happened—open rates, MQLs, churn numbers. By the time we saw the problem, it was too late. 𝐋𝐞𝐚𝐝𝐬? 𝐃𝐞𝐚𝐝. 𝐂𝐮𝐬𝐭𝐨𝐦𝐞𝐫𝐬? 𝐆𝐨𝐧𝐞. 𝐁𝐮𝐝𝐠𝐞𝐭? 𝐁𝐮𝐫𝐧𝐞𝐝. But AI and predictive analytics are flipping the game. 𝐌𝐚𝐫𝐤𝐞𝐭𝐢𝐧𝐠 𝐢𝐬𝐧’𝐭 𝐫𝐞𝐚𝐜𝐭𝐢𝐯𝐞 𝐚𝐧𝐲𝐦𝐨𝐫𝐞. 𝐈𝐭’𝐬 𝐩𝐫𝐞𝐝𝐢𝐜𝐭𝐢𝐯𝐞. 🔹 𝐋𝐞𝐚𝐝 𝐅𝐨𝐫𝐞𝐜𝐚𝐬𝐭𝐢𝐧𝐠 Traditional lead scoring is broken. A whitepaper download? That’s not intent—it’s noise. When we actually analyzed behavioral data using platforms like HubSpot, we found that multiple pricing page visits and engagement with onboarding content predicted conversions 3x better than generic lead scores. 𝐖𝐢𝐭𝐡 𝐦𝐮𝐥𝐭𝐢-𝐭𝐨𝐮𝐜𝐡 𝐚𝐭𝐭𝐫𝐢𝐛𝐮𝐭𝐢𝐨𝐧 𝐦𝐨𝐝𝐞𝐥𝐬 and 𝐛𝐞𝐡𝐚𝐯𝐢𝐨𝐫𝐚𝐥 𝐜𝐨𝐡𝐨𝐫𝐭 𝐚𝐧𝐚𝐥𝐲𝐬𝐢𝐬 ✔ Leads with 𝐫𝐞𝐩𝐞𝐚𝐭 𝐯𝐢𝐬𝐢𝐭𝐬 𝐭𝐨 𝐭𝐡𝐞 𝐩𝐫𝐢𝐜𝐢𝐧𝐠 𝐩𝐚𝐠𝐞 had a 𝟑𝐱 𝐡𝐢𝐠𝐡𝐞𝐫 𝐥𝐢𝐤𝐞𝐥𝐢𝐡𝐨𝐨𝐝 𝐨𝐟 𝐜𝐨𝐧𝐯𝐞𝐫𝐬𝐢𝐨𝐧 ✔ Prospects engaging with 𝐢𝐧𝐭𝐞𝐫𝐚𝐜𝐭𝐢𝐯𝐞 𝐝𝐞𝐦𝐨𝐬 moved through the funnel 𝟒𝟐% 𝐟𝐚𝐬𝐭𝐞𝐫 ✔ Combining 𝐢𝐧𝐭𝐞𝐧𝐭 𝐬𝐢𝐠𝐧𝐚𝐥𝐬 𝐰𝐢𝐭𝐡 𝐟𝐢𝐫𝐦𝐨𝐠𝐫𝐚𝐩𝐡𝐢𝐜𝐬 increased lead quality 𝐰𝐢𝐭𝐡𝐨𝐮𝐭 𝐢𝐧𝐟𝐥𝐚𝐭𝐢𝐧𝐠 𝐚𝐜𝐪𝐮𝐢𝐬𝐢𝐭𝐢𝐨𝐧 𝐜𝐨𝐬𝐭𝐬 We stopped chasing the wrong leads. And our pipeline? Tighter than ever. 🔹 𝐂𝐮𝐬𝐭𝐨𝐦𝐞𝐫 𝐑𝐞𝐭𝐞𝐧𝐭𝐢𝐨𝐧 A churn report tells you what you lost. But by then, it’s a post-mortem. Advanced platforms flag disengagement before it happens. A simple tweak—triggering check-ins for inactive accounts—cut churn by 15% in six months. A simple intervention—𝐭𝐫𝐢𝐠𝐠𝐞𝐫𝐢𝐧𝐠 𝐚𝐮𝐭𝐨𝐦𝐚𝐭𝐞𝐝 𝐫𝐞-𝐞𝐧𝐠𝐚𝐠𝐞𝐦𝐞𝐧𝐭 𝐰𝐨𝐫𝐤𝐟𝐥𝐨𝐰𝐬 when customers showed 𝟑+ 𝐝𝐢𝐬𝐞𝐧𝐠𝐚𝐠𝐞𝐦𝐞𝐧𝐭 𝐭𝐫𝐢𝐠𝐠𝐞𝐫𝐬—led to a 𝟏𝟓% 𝐫𝐞𝐝𝐮𝐜𝐭𝐢𝐨𝐧 𝐢𝐧 𝐜𝐡𝐮𝐫𝐧 𝐢𝐧 𝐬𝐢𝐱 𝐦𝐨𝐧𝐭𝐡𝐬. 🔹 𝐏𝐫𝐨𝐝𝐮𝐜𝐭 𝐅𝐢𝐭 Guessing what users want is a waste of time. Predictive analytics showed us which features had a 𝟒𝟎% 𝐥𝐢𝐤𝐞𝐥𝐢𝐡𝐨𝐨𝐝 𝐨𝐟 𝐚𝐝𝐨𝐩𝐭𝐢𝐨𝐧 before launch. The result? No wasted dev cycles, no misfires—just 𝐝𝐚𝐭𝐚-𝐛𝐚𝐜𝐤𝐞𝐝 𝐝𝐞𝐜𝐢𝐬𝐢𝐨𝐧𝐬. If you’re still relying on past data to drive strategy, 𝐲𝐨𝐮’𝐫𝐞 𝐩𝐥𝐚𝐲𝐢𝐧𝐠 𝐲𝐞𝐬𝐭𝐞𝐫𝐝𝐚𝐲’𝐬 𝐠𝐚𝐦𝐞. 𝐌𝐚𝐫𝐤𝐞𝐭𝐢𝐧𝐠 𝐢𝐬𝐧’𝐭 𝐚𝐛𝐨𝐮𝐭 𝐥𝐨𝐨𝐤𝐢𝐧𝐠 𝐛𝐚𝐜𝐤. 𝐈𝐭’𝐬 𝐚𝐛𝐨𝐮𝐭 𝐤𝐧𝐨𝐰𝐢𝐧𝐠 𝐰𝐡𝐚𝐭’𝐬 𝐧𝐞𝐱𝐭. #PredictiveAnalytics #MarketingStrategy #DataDriven #Growth

We changed one button on a client’s website and watched acquisition costs drop by a third overnight. Same ads, same audience… just tracking what Meta ACTUALLY values instead of what everyone thinks it values. Here’s the exact framework: 1. Fix Your Funnel Mechanics Standard e-commerce flows create massive inefficiencies when they don't align with platform event schemas. Multi-page checkouts, delayed confirmation signals, and fragmented purchase paths all force algorithms to work harder to find your customers. 2. Implement Strategic Conversion Paths Single-page checkout flows increase "InitiateCheckout" events by 20%, giving Meta earlier signals that immediately improve auction performance. Email-capture modals treated as "Lead" events let you optimize for actions Meta can deliver at a fraction of "Purchase" event costs. Progressive form fields create additional data points that feed algorithms the optimization signals they crave. 3. Optimize for Predictive Events While everyone obsesses over "add-to-cart," events like "complete registration" often predict lifetime value more accurately and convert at substantially lower costs. The accounts we've restructured around these insights consistently see 30%+ CPA improvements within weeks. 4. Sequence Your Channels Strategically Start with Pinterest/YouTube for cold reach. Transition to Meta Lead/Form campaigns, optimizing toward micro-conversions. Finally, move to Meta Conversion campaigns using fresh "AddToCart" seed audiences. This sequence leverages each platform's attribution window to maximize incremental lift while preventing platform competition for conversion credit. The brands beating CAC benchmarks in competitive markets have simply restructured their funnel mechanics to align with how algorithms really value conversions. This approach requires zero additional spend; just a strategic reconfiguration of your customer journey.

Your lead scoring is broken. Here's the model that predicts revenue with 87% accuracy. Most B2B companies score leads like it's 2015. ┣ Downloaded whitepaper: +10 points ┣ Attended webinar: +15 points ┗ Opened email: +5 points Meanwhile, 73% of these "hot" leads never convert. Here's what we discovered after analyzing 10,000+ B2B leads: The leads scoring highest in traditional systems aren't buyers. They're information collectors. They download everything. Open every email. Click every link. But when sales calls? ↳ "Just doing research." ↳ "Not ready yet." ↳ "Send me more info." The leads that DO convert show completely different signals: They don't just visit your pricing page. They spend 8 minutes there, come back twice more that week, then search "[competitor] vs [your company]." They're not reading blog posts. They're calculating ROI and researching implementation. Activity doesn't equal intent. And that's where most scoring models fall apart. We rebuilt lead scoring from the ground up. Instead of rewarding every action equally, we weighted four factors based on what actually predicts revenue: ┣ Intent signals (40%) - someone searching "implementation" is closer to buying than someone downloading an ebook ┣ Behavioral depth (30%) - how someone engages tells you more than what they engage with ┣ Firmographic fit (20%) - perfect ICP match or bust ┗ Engagement quality (10%) - quality of interaction matters The framework is simple. The impact isn't. We map every lead to one of four tiers: ┣ 90-100 points → Sales gets them same-day ┣ 70-89 points → Automated nurture + retargeting ┣ 50-69 points → Educational content track ┗ Below 50 → Long-term relationship building No more dumping mediocre leads on sales and wondering why they don't follow up. Results after 6 months: ┣ Sales acceptance rate: +156% ┣ Sales cycle length: -41% ┗ Lead-to-customer rate: +73% The biggest shift wasn't the scoring model. It was the mindset. 🛑 Stop measuring marketing by MQL volume. ✔️ Start measuring it by how many MQLs sales actually wants to talk to. Your automation platform will happily score 500 leads as "hot" this month. But if sales only accepts 50, you don't have a volume problem. You have a scoring problem. Traditional scoring optimizes for activity. And fills your pipeline with noise. Revenue-predictive scoring optimizes for intent and fills it with buyers. If you'd like help with assessing your current lead scoring logic, comment "SCORING" and I'll get in touch to schedule a FREE consultation.

Funnel analysis is essential for understanding where and why users drop off in structured workflows like onboarding, checkout, or sign-up flows. Unlike clickstream analysis, which maps the broader user journey, or session analysis, which focuses on individual interactions, funnel analysis zeroes in on goal-driven processes, tracking user progression and highlighting abandonment points. What’s evolving today is how we approach funnel analysis. With more natural behavioral data and machine learning enhancements, we’re moving beyond static drop-off reporting. AI-driven insights now allow teams to predict drop-offs before they occur, identifying early warning signs like hesitation patterns or inefficient navigation loops. This proactive approach enables UX researchers to refine workflows dynamically, improving user retention before friction escalates. Advanced segmentation is also revolutionizing funnel tracking. Instead of analyzing drop-offs solely through broad demographic data, researchers can now segment users based on behavioral clusters - how they interact with key touchpoints, their engagement duration, or even their likelihood of return. This behavioral-first approach allows for personalized interventions that cater to different user types, ensuring a more seamless experience for all. Beyond traditional conversion tracking, we’re incorporating statistical methods like survival analysis to estimate how long users remain engaged in a funnel and Markov modeling to understand the probability of transitioning between different steps. Instead of treating drop-offs as simple yes/no outcomes, these approaches quantify the likelihood of users completing a process based on their prior actions, leading to more precise and actionable insights. Funnel analysis is no longer just about counting conversions, it’s about deeply understanding user intent, predicting disengagement, and designing experiences that encourage progression. The shift from static reporting to predictive UX optimization is already underway.

𝐃𝐚𝐭𝐚 𝐒𝐜𝐢𝐞𝐧𝐜𝐞 𝐈𝐧𝐭𝐞𝐫𝐯𝐢𝐞𝐰 𝐐𝐮𝐞𝐬𝐭𝐢𝐨𝐧: You’re using logistic regression in a marketing campaign to predict customer conversion. How would you convert predicted probabilities into actionable decisions (e.g., who to target) To convert predicted probabilities into decisions, I map them to business outcomes—specifically, target customers where expected benefit exceeds cost. For instance, a sample set of steps would be - First, define business objective In a marketing campaign, the goal is to maximize conversions while minimizing spend. Second, compute the expected value per customer 𝘌𝘹𝘱𝘦𝘤𝘵𝘦𝘥 𝘱𝘳𝘰𝘧𝘪𝘵= 𝘗(𝘤𝘰𝘯𝘷𝘦𝘳𝘴𝘪𝘰𝘯)×𝘳𝘦𝘷𝘦𝘯𝘶𝘦−𝘤𝘰𝘴𝘵 𝘰𝘧 𝘵𝘢𝘳𝘨𝘦𝘵𝘪𝘯𝘨 Example, If the model predicts a customer has a 30% chance of converting, and: Revenue per conversion = $50 Cost to target = $10 Then: 0.3×50−10=$5 expected profit → target them Next, Choose a threshold Rather than using a default 0.5, I select a threshold that maximizes ROI, possibly using - Profit curves, Uplift modeling, Custom cost-benefit matrices (comment in the comment if you want to know more about these!) Next, Validate strategy I simulate the campaign using historical data to see if my targeting strategy would’ve outperformed random or rule-based approaches. 𝐒𝐮𝐦𝐦𝐚𝐫𝐲: I turn predicted probabilities into decisions by aligning them with business impact—target only when the expected gain outweighs the cost, not just based on a fixed threshold.

Still optimizing your paid campaigns for signups when they're VERY FAR from all being created equal? For my latest article, I've teamed up with David Loris: a data scientist with nearly 20 years of experience helping brands like Booking, Expedia, and Typeform improve marketing performance. Together, we've noticed a critical mistake: marketing teams spending millions optimizing toward ALL signups instead of THE MOST VALUABLE ones. Here's what happens when you move from tCPA to tROAS with conversion value prediction: 1️⃣ You stop treating all leads equally (when some might be worth 50x others) 2️⃣ You can bid higher for high-value prospects (a CAC is only as high as its return) 3️⃣ You harness the FULL power of ad platform algorithms (instead of manually figuring everything out yourself) Spoiler: For B2B and companies with long sales cycles, this approach can dramatically improve ROAS by predicting which leads will actually convert and generate significant revenue. In this article, you'll learn: 👉 Which businesses need conversion value prediction the most 👉 What's technically required to implement this (hint: it's achievable, but also complex) 👉 How to run your first test without overengineering 👉 A practical rollout plan from assessment to analysis Working with freemium models or lengthy sales cycles? Wanting to scale paid efficiently? Click to read the full article and drop any questions in the comments. https://lnkd.in/dQxNGDgW