Blog

<h2>The Gap Between Promise and Reality</h2>In an effort to democratize marketing mix modelling (MMM), tech leaders like Google Meridian, Google Lightweight and Meta Robyn have released free, open-source tools. The stated goal? To make MMM more transparent, customizable, and accessible.Yet despite their technical sophistication, these tools remain out of reach for the very people who need them most: non-technical marketers.This article examines:<ul><li>Why these tools were developed—and why their design overlooks real marketing needs. </li><li>Key limitations of Google Meridian, Google Lightweight, and Meta Robyn—based on hands-on user frustrations. </li><li>Why MMM success hinges on marketer adoption—not just statistical metrics. </li><li>What a truly marketer-centric MMM solution should offer. </li></ul><h2>Why Google &amp; Meta Released Open-Source MMM Tools</h2><h3>The Stated Vision</h3><ul><li>Reduce reliance on costly vendors (e.g., Nielsen, Analytic Partners). </li><li>Encourage customization for unique business needs. </li><li>Promote transparency in an often opaque field.</li></ul><h3>The Unspoken Realities</h3><ul><li>Data ecosystem lock-in: Tools subtly encourage reliance on Google/Meta platforms. </li><li>Talent recruitment: Open-source projects attract top data science minds. </li><li>Industry influence: Framing their methods as "standard" shapes market preferences.</li></ul>While these motivations aren’t inherently problematic, they’ve led to tools that serve data teams far better than marketing teams.<h2>The Reality: Why Open-Source MMM Tools Struggle with Adoption</h2><h3>Designed for Coders, Not Marketers</h3><ul><li>Google Meridian &amp; Lightweight MMM: Python-based Bayesian frameworks requiring statistical expertise. </li><li>Meta Robyn: R-based evolutionary computing models that operate like a "black box." </li><li>No intuitive interfaces—just code repositories and command-line operations.</li></ul>Result: Most marketers cannot use these tools independently, forcing reliance on data scientists.<h3>Ignoring Critical Business Needs</h3><ul><li>Inflexible Adjustments</li></ul><ol><li style="list-style-type: none;"><ol><li>Media channel tuning: No easy way for marketers to adjust expected media performance (e.g., "TV should have higher impact than social"). </li><li>Prior distributions: Even data scientists struggle to define Bayesian priors in Meridian/Lightweight. </li><li>Re-running models: Robyn’s evolutionary approach requires restarting the entire process for minor tweaks.</li></ol></li></ol><ul><li>Data Preparation Hurdles</li></ul><ol><li style="list-style-type: none;"><ol><li style="">No plug-and-play connectors: Marketers must manually assemble data from disparate sources (Google Ads Manager, or other ad platforms). </li><li style="">Opaque formatting rules: Only data scientists understand how to structure inputs correctly.</li></ol></li></ol><ul><li>Delayed Insights</li></ul><ol><li style="list-style-type: none;"><ol><li>Meta Robyn’s evolutionary computing is slow for large datasets. </li><li>MCMC (Markov chain Monte Carlo) sampling process inherent in Bayesian methods also creates significant computational bottlenecks.&nbsp;</li></ol></li></ol><h3>Community Pain Points</h3><ul><li>Google Meridian &amp; Lightweight MMM</li></ul><ol><li style="list-style-type: none;"><ol><li>"Documentation is overly technical": Filled with intimidating equations, lacking practical guidance. </li><li>"Adjusting priors is guesswork": No clear way to align assumptions with business reality. </li><li>"Lightweight is too simplistic": Lacks nuance for multi-channel strategies.</li></ol></li></ol><ul><li>Meta Robyn</li></ul><ol><li style="list-style-type: none;"><ol><li>"Uncontrollable model-building": Evolutionary algorithms optimize without user constraints. </li><li>"Adstock and diminishing return transformation tweaks are tedious and time consuming": Changing decay rates requires re-running the entire model.</li></ol></li></ol><h2>The Root Problem: Built for Data Teams, Not Marketing Decision-Makers</h2><h3>The "Lab vs. Field" Disconnect</h3><ul><li>These tools prioritize:</li></ul><ol><li style="list-style-type: none;"><ol><li>Statistical novelty (e.g., Robyn’s evolutionary optimization). </li><li>Methodological purity (e.g., Meridian’s Bayesian rigor).</li></ol></li></ol><ul><li>But marketers need:</li></ul><ol><li style="list-style-type: none;"><ol><li>Speed (days, not weeks). </li><li>Control (adjust assumptions without coding). </li><li>Clarity (insights that align with business intuition).</li></ol></li></ol><h3>MMM Has No "Source of Ground Truth Answer"—So Why Obsess Over Metrics?</h3>Unlike controlled experiments, MMM relies on interpretation. Yet open-source tools fixate on:<ul><li>R², MSE, back-testing error (withholding some data points and using past observations to predict the withheld values) Instead of: </li><li>"Do these results feel reasonable?" </li><li>"Can we act on them tomorrow?"</li></ul><h3>The Forgotten Success Metric: Marketer Trust</h3>A model’s real value isn’t its statistical score—it’s whether:✅ Marketers trust it. ✅ Marketing Stakeholders believe it. ✅ It improves marketing decisions.Most open-source tools fail this test. Consider the case of a Fortune 500 company where internal data scientists employed Google Meridian for MMM - the marketing team has repeatedly questioned both the methodology and business relevance of the outputs.<h2>What a Marketer-Centric MMM Tool Would Offer</h2>A truly effective MMM solution would:✅ Require zero coding—button-click workflows, easy data prepare, and auto transformation parameter tuning. ✅ Let marketers set goals (e.g., "Prioritize sales activation within short-term"). ✅ Obtain what marketers expect (achieve expected media impact without back-and-forth running models). ✅ Deliver insights in hours—not weeks of iterations. ✅ Speak the language of business—not p-values and confidence intervals.<h2>Conclusion: A Call for Truly Accessible MMM</h2>Google and Meta’s tools represent progress—but they’re stuck in the data science lab. Until they address:<ul><li>Usability (for non-technical users). </li><li>Flexibility (to match business requests). </li><li>Speed (to keep pace with marketing cycles).</li></ul>They’ll remain academic curiosities, not practical solutions.The future of MMM belongs to tools that empower marketers—not just impress data scientists.<h2>Final Thought </h2>If MMM is meant to guide marketing decisions, why are the tools designed for people who don’t make them?The industry needs a shift—from model-centric to marketer-centric.

The Open-Source MMM Paradox: Why Google and Meta’s Tools Miss the Mark for Real-World Marketers

<h2>Introduction</h2>Marketing Mix Modelling (MMM) is a powerful tool for measuring the effectiveness of marketing campaigns, optimizing budgets, and improving ROI. Yet, despite its potential, many non-technical marketers—both on the brand side and within media agencies—have been slow to adopt it.The co-founder of More Than Data has worked closely with both groups and understands their unique challenges. While they share common pain points, their perspectives on MMM differ. This article explores why MMM hasn’t gained widespread traction among non-technical marketers, the key differences between agency and brand-side marketers, and how More Than Data facilitates a technical-to-non-technical conversation to drive adoption.<h2>The Challenge: Why Non-Technical Marketers Struggle with MMM</h2>Marketing Mix Modelling is often seen as:<ul><li>Too complex – Many marketers lack statistical expertise and find MMM intimidating. </li><li>Time-consuming – Traditional MMM requires extensive data collection and long analysis cycles. </li><li>Lack of clear communication – Technical teams struggle to translate insights into actionable strategies for marketers.</li></ul>As a result, many marketers—especially those without a data background—rely on simpler (but less accurate) metrics like last-click attribution or platform-reported performance.But this is changing. With advancements in advanced analytics and automation, MMM is becoming more accessible. The key is bridging the gap between data science and marketing execution—which is exactly what More Than Data does.<h2>Media Agency Marketers vs. Brand-Side Marketers: Similarities and Differences</h2>While both groups share the same end goal—driving effective marketing campaigns—their priorities, constraints, and perspectives differ.<h3>1. Media Agency Marketers: The Performance-Driven Partners</h3>Who they are:Media agency marketers work across multiple clients, managing ad spend, optimizing campaigns, and proving Return-on-ads-spend (ROAS). They operate in fast-paced environments where agility is key.Their challenges with MMM:<ul><li>Short-term focus – Clients demand quick results, making long-term MMM analysis seem impractical. </li><li>Multi-client complexity – Each brand has different KPIs, making one-for-all MMM difficult. </li><li>Pressure to justify spend – Without clear MMM insights, they rely on platform metrics that may overstate performance.</li></ul>What they need from MMM:<ul><li>Faster, real-time insights to adjust campaigns on the fly. </li><li>Clear visualizations (not raw data) to present to clients. </li><li>Proof that their media mix recommendations are optimal.</li></ul><h3>2. Brand-Side Marketers: The Strategic Decision-Makers</h3>Who they are:Brand marketers focus on long-term growth, brand equity, and customer loyalty. They work closely with leadership to align marketing with business objectives.Their challenges with MMM:<ul><li>Budget constraints – MMM is often seen as a luxury, not a necessity. </li><li>Internal silos – Sales, finance, and marketing teams may not share data seamlessly. </li><li>Fear of complexity – Many assume MMM requires a data science team.</li></ul>What they need from MMM:<ul><li>A clear link between marketing investment and business outcomes (e.g., sales, brand awareness, market share). </li><li>Guidance on budget allocation across channels (traditional vs. digital). </li><li>Simplified reporting for C-level stakeholders.</li></ul>Where They OverlapDespite their differences, both groups:<ul><li>Need actionable insights, not just data. </li><li>Want faster, more transparent measurement. </li><li>Struggle with attribution in a multi-channel world. </li></ul><h2>How More Than Data Bridges the Gap</h2>The co-founder of More Than Data has worked with both agency and brand-side marketers, recognizing that MMM adoption isn’t just about better models—it’s about better accessibility and communication.Here’s how they make MMM accessible:<h3>1. Speaking Their Language (No Jargon Allowed)</h3><ul><li>For agencies: Focus on speed, flexibility, and client-facing reports. </li><li>For brands: Tie MMM to business outcomes (e.g., “This is how TV ads impact quarterly revenue”).</li></ul><h3>2. Automating the Heavy Lifting</h3><ul><li>Traditional MMM takes months; More Than Data delivers insights in minutes (our product Minute MMM can do so). </li><li>No need for a PhD in statistics—marketers get clear dashboards, not complex equations.</li></ul><h3>3. Aligning with Their Goals</h3><ul><li>Agencies → Prove your media mix is optimal. </li><li>Brands → Justify budget driving business performance to leadership.</li></ul><h3>4. Building Trust Through Education</h3><ul><li>Free tools as high-value offers, simplified guides, and case studies to demystify MMM. </li><li>Showing real-world examples of how MMM optimized media mix and delivered business outcomes.&nbsp;</li></ul><h2>The Future: MMM as a Standard Tool for All Marketers</h2>The gap between technical MMM and non-technical marketers is closing. With tools like Minute MMM and Budget Optimizer, agencies can measure media performance and optimize future campaigns faster, and brands can make data-driven budget decisions with confidence.The key is not just better models, but better conversations. By translating MMM into practical, marketer-friendly insights, More Than Data is helping both agency and brand-side marketers unlock the full potential of their marketing.<h2>Final Thought</h2>Marketing Mix Modelling shouldn’t be reserved for data scientists. Whether you’re an agency marketer juggling multiple clients or a brand leader making strategic budget calls, MMM can—and should—work for you.The question isn’t “Why hasn’t MMM been adopted?” but rather “How can we make MMM work for non-technical marketer?”And that’s exactly what More Than Data is solving.Would you like to explore how MMM can fit into your marketing strategy? Let’s start the conversation—no technical expertise required.&nbsp;

Why Marketing Mix Modelling Hasn’t Been Adopted by Non-Technical Marketers—And How We Bridge the Gap

<h2>Introduction</h2>In today’s competitive health and wellness market, brands must leverage data-driven insights to optimize their marketing strategies. For a leading men’s health and general wellbeing brand, measuring the effectiveness of Above-the-Line (ATL) and Below-the-Line (BTL) campaigns was a significant challenge. With fragmented data, overlapping audience segments, and a need for real-time measurement, the brand partnered with More Than Data to build a granular Marketing Mix Model (MMM) that quantified media performance, promotional synergy, and cross-product impact.This article explores how More Than Data’s solution helped the brand:<ul><li>Measure and quantify ATL &amp; BTL media performance </li><li>Track the synergy between TV/Radio and digital promotions </li><li>Optimize media mix for different audience groups </li><li>Analyse cross-product bundle sales impact </li><li>Enable real-time marketing planning &amp; forecasting</li></ul>The Challenge: Fragmented Data &amp; Complex Audience TargetingThe men’s health brand primarily targeted three key audience segments:<ul><li>Men aiming to lose body weight </li><li>Men looking to grow muscle </li><li>Men seeking a balanced everyday diet</li></ul>Each segment responded differently to media channels, making it crucial to tailor the marketing mix. However, the brand faced several challenges:<h3>1. Disorganized &amp; Fragmented Data</h3><ul><li>Data from Meta, Google Ads, TikTok, Snapchat, TV, and Radio was mismanaged. </li><li>Manual data collection took 2-3 weeks, delaying insights. </li><li>Mismatched reporting led to version control issues.</li></ul><h3>2. Measuring ATL &amp; Promotional Synergy</h3><ul><li>The brand needed to measure both standalone and combined effects of ATL (TV, Radio) and BTL (social, search, influencers). </li><li>They also wanted to track how promotions (online vouchers) interacted with media spend.</li></ul><h3>3. Cross-Product Purchase Behaviour</h3><ul><li>Many customers bought product bundles (e.g., weight loss + muscle growth). </li><li>Traditional MMM struggled to attribute sales correctly when promotions overlapped.</li></ul><h3>4. Real-Time Decision-Making Needed</h3><ul><li>The in-house marketing team required a live measurement system to adjust campaigns dynamically.</li></ul><h2>The Solution: Automated Data Pipelines &amp; Granular MMM</h2><h3>1. Automated &amp; Semi-Automated Data Collection</h3><ul><li>Fully Automated: APIs from Google Ads, Meta, TikTok, Snapchat pulled real-time ad performance data. </li><li>Semi-Automated: TV &amp; Radio data was ingested from publisher reports in just 3 days (vs. 2-3 weeks manually).</li></ul><h3>2. Granular Marketing Mix Modelling</h3><ul><li>Measured ATL &amp; BTL contributions separately and their synergistic effects. </li><li>Modelled bundle promotions to understand cross-product sales impact. </li><li>Built response curves to predict sales based on media spend.</li></ul><h3>3. Optimized Media Mix for Each Audience Segment</h3><ul><li>Weight Loss Audience: Best responded to online video, influencers, TikTok, and social ads. </li><li>Muscle Growth Audience: More influenced by search and performance marketing. </li><li>Diet Balancers: Engaged via ATL (TV/Radio) + educational promotions.</li></ul><h3>4. Real-Time Planning &amp; Forecasting</h3><ul><li>The brand could simulate budget scenarios before launching campaigns. </li><li>Incremental testing helped validate media effectiveness without overspending.</li></ul><h2>The Results: Higher Sales, Better ROAS &amp; Smarter Budgeting</h2>✅ Identified High-Performing Bundle Promotions<ul><li>Certain product bundles (e.g., weight loss + diet balancers) drove higher conversion rates.</li></ul>✅ Optimized Media Mix per Audience<ul><li>Weight loss ads performed best on TikTok &amp; influencer marketing. </li><li>Diet-focused campaigns benefited from TV ads + voucher promotions.</li></ul>✅ Sales Uplift: 2% In-Store, 8% Online<ul><li>In-store sales grew 2% YoY due to better-targeted ATL campaigns. </li><li>Online sales surged 8% from optimized digital promotions.</li></ul>✅ 16% ROAS Improvement<ul><li>Synergy between ATL + digital promotions amplified returns without increasing budget.</li></ul><h2>Testimonial from the Brand</h2>"Working with More Than Data transformed our marketing measurement approach. Their automated data pipelines eliminated weeks of manual work, and their MMM provided crystal-clear insights into how ATL, digital ads, and promotions work together. We now know exactly which media mix works for each audience segment and how bundle promotions drive sales. Most importantly, we’ve seen an 8% boost in online sales and 16% higher ROAS—all without increasing our budget. Their real-time planning tools empower our in-house team to make faster, smarter decisions."<h2>Conclusion: Data-Driven Decisions = Higher ROAS</h2>For men’s health brands, precision in media measurement is key. By leveraging automated data collection, granular MMM, and response curve modelling, More Than Data helped this brand:<ul><li>Eliminate guesswork in media planning </li><li>Maximize ATL + BTL synergy </li><li>Boost sales &amp; ROAS efficiently</li></ul>The future of marketing is real-time, data-driven, and audience-specific—and this case study proves it.

More Than Data Transformed a Men’s Health Brand’s Marketing Measurement & Strategy

<h2>Understanding the Basics of a Marketing Mix Model</h2>
A Marketing Mix Model (MMM) is an analytical tool used to evaluate the impact of various marketing tactics on sales and other performance metrics. By understanding this model, you can allocate your budget more effectively and optimize your marketing strategies. The MMM considers four key marketing elements: Product, Price, Place, and Promotion, often referred to as the 4Ps.

Before diving into building your own model, it's essential to get familiar with these components. The Product refers to what you are selling, Price is how much you charge, Place involves distribution channels, and Promotion covers advertising and other marketing efforts. Mastering these elements will be critical in creating an effective model.

<figure id="figure-dclpgek5vbm9b7kp60"><img id="dclpgek5vbm9b7kp60" src="https://media.gettyimages.com/id/629303228/photo/im-happy-to-explain.jpg?b=1&s=2048x2048&w=0&k=20&c=WEmzS8V0sMIhn9i-_wyF5Cs9Iyk4Ye2AfzGK4jmkLt4=" alt="marketing strategy" class="object-cover bg-gray-100" style="aspect-ratio: 16/9"></figure>

<h2>Setting Up Your Data</h2>
The first step in building a Marketing Mix Model is gathering the right data. You will need historical data on sales, marketing spend, and any external factors that might influence your results, such as economic conditions or seasonal trends. Accurate data is crucial as it forms the foundation of your model.

Once you have collected your data, organize it in a spreadsheet. Ensure that each variable is clearly labeled and that the data covers a sufficient time period to allow for meaningful analysis. Ideally, you should have at least two years of data to identify trends and patterns.

<figure id="figure-soaylopau5m9b7kp60"><img id="soaylopau5m9b7kp60" src="https://media.gettyimages.com/id/1954841243/photo/data-analysis-chart-graph-3d-statistics-background.jpg?b=1&s=2048x2048&w=0&k=20&c=JqlWDrRuqpKotSJiWlw9XEKfqmlar9qw3g5TGMGWdVk=" alt="data analysis" class="object-cover bg-gray-100" style="aspect-ratio: 16/9"></figure>

<h2>Choosing the Right Tools</h2>
With your data in place, the next step is selecting the appropriate tools for building your model. There are several software options available, ranging from Excel to more advanced analytics platforms like R or Python. For beginners, starting with Excel might be the easiest way to get a grip on the basics.

If you're more experienced or willing to learn, using R or Python can provide greater flexibility and more robust modeling capabilities. These tools offer powerful statistical packages that can handle complex data sets and provide more accurate results.

<figure id="figure-4kasi2gznbm9b7kp60"><img id="4kasi2gznbm9b7kp60" src="https://media.gettyimages.com/id/2174071798/photo/maintenance-repair-and-improvement-concept-tools-3d-icon-on-the-glossy-futuristic-surface.jpg?b=1&s=2048x2048&w=0&k=20&c=kTE5mxyqtZOeO86BdxSdo3hShGKpODKO1sjCoG-1EaA=" alt="software tools" class="object-cover bg-gray-100" style="aspect-ratio: 16/9"></figure>

<h2>Building Your Model</h2>
Now that you're equipped with data and tools, it's time to start building your Marketing Mix Model. Begin by inputting your data into the chosen software and setting up a basic regression model. Regression analysis will help you understand the relationship between different marketing variables and sales outcomes.

Pay attention to key metrics like Return on Investment (ROI) and Elasticity. These will help you evaluate which marketing activities are driving the most value and where adjustments can be made for better performance. Remember, the goal is to identify which elements of your mix are working and which need improvement.

<h3>Testing and Refining Your Model</h3>
Once your model is built, it's crucial to test its accuracy by comparing its predictions against actual results. This step will help you identify any discrepancies and refine your model further. Don't be discouraged if your first attempt isn't perfect; iterative testing and tweaking are part of the process.

Consider running simulations to predict future outcomes based on different marketing scenarios. This can provide valuable insights and guide decision-making for upcoming campaigns.

<figure id="figure-ryilvgsg2qm9b7kp60"><img id="ryilvgsg2qm9b7kp60" src="https://media.gettyimages.com/id/2041488351/photo/artificial-intelligence-artificial-general-intelligence-neural-science-computing-node.jpg?b=1&s=2048x2048&w=0&k=20&c=jNywzXLhgT6WVvFTQGGkeLZ98BCiT-4hPWCRl1D2Y_M=" alt="testing model" class="object-cover bg-gray-100" style="aspect-ratio: 16/9"></figure>

<h3>Interpreting Results</h3>
Interpreting the results of your Marketing Mix Model is perhaps the most critical part of the process. Look for trends and insights that can inform your strategy moving forward. For instance, if promotions consistently show high ROI, you might consider increasing your investment in this area.

Share these insights with your team to ensure everyone is aligned on strategic changes. Effective communication of your findings can lead to more informed decisions and ultimately improve overall marketing success.

<h3>Continuous Improvement</h3>
The world of marketing is ever-evolving, so it's important to continually update and refine your Marketing Mix Model. As new data becomes available or market conditions change, revisit your model to ensure it remains relevant and accurate.

Remember, a well-maintained MMM can be a powerful tool in your marketing arsenal, helping you make data-driven decisions that enhance performance and drive growth.

Quick Guide to Building Your First Marketing Mix Model in Under 10 Minutes

Marketing Mix Modelling (MMM) has been around for decades. It’s the OG of marketing measurement—older than Facebook, older than Google Ads, older than that one Excel formula you keep forgetting.Yet, despite its proven ability to answer critical questions like "Is my TV ad actually working?" or "Why am I spending so much on LinkedIn ads that only my ex-colleagues see?", many marketers still treat MMM like a mysterious black box.So why aren’t strategists, media planners, account managers, and brand managers using MMM more? Let’s break it down—with a side of brutal honesty and a dash of humor.<h3>Reason 1: “We Don’t Have the Data… Or Do We?” </h3>The Problem:Many marketers believe they need perfect data to run MMM. Newsflash: No one has perfect data.<ul><li>Media Planners: "Our agency reports are a mess—some are in PDFs, some in Excel, and some are just scribbled on a napkin from last year’s media lunch." </li><li>Brand Managers: "We don’t track in-store sales properly, so MMM won’t work for us." </li><li>Media Investment Managers: "Half our digital spend is dark social—how do we even measure that?"</li></ul>The Reality: MMM doesn’t need perfect data—it just needs enough data. More Than Data has worked with:<ul><li>Brands that thought their data was unusable (spoiler: it wasn’t). </li><li>Companies still using spreadsheets from 2015 (yes, really). </li><li>Teams who didn’t even know where their media spend reports were stored (we found them).</li></ul>Lesson:&nbsp; If you’re waiting for "perfect" data, you’ll be waiting forever. Start messy, refine later.<h3>Reason 2: “MMM Sounds Complicated… Like, ‘Call-IT-Support’ Complicated” </h3>The Problem: Somewhere along the way, MMM got a reputation for being:<ul><li>Too technical (What’s a Bayesian model? Is that like a Star Wars thing?) </li><li>Too slow (We need answers now, not in six weeks!) </li><li>Too academic (If I see one more regression analysis, I’m switching careers.)</li></ul>The Reality: Modern MMM isn’t your grandpa’s econometrics project. At More Than Data, we:<ul><li>Explain results in plain English (No jargon, just "This is what’s working, this is what’s not.") </li><li>Deliver insights fast (Weekly updates, not quarterly post-mortems.) </li><li>Make it visual (Dashboards &gt; 100-page PDFs.) </li></ul>Lesson:&nbsp; If your MMM provider makes you feel like you’re back in stats class, fire them.<h3>Reason 3: “But Last Time We Did MMM, Nothing Changed!” </h3>The Problem: Some marketers have tried MMM before… and then promptly ignored the findings.<ul><li>Account Managers: "The model said to cut radio spend, but the client loves radio!" </li><li>Marketing Managers: "The report said influencers don’t work, but our CEO’s daughter is an influencer!" </li><li>Strategists: "The data says we should shift budget, but we’ve always done it this way!"</li></ul>The Reality: MMM isn’t a magic wand—it’s a flashlight. It shows you where the money’s working (and where it’s disappearing into a black hole). But if you don’t act on it, nothing changes.Lesson:&nbsp;MMM only works if you actually use it. Shocking, we know.<h3>Reason 4: “We’re Too Busy Fighting Fires to Think Long-Term” </h3>The Problem: Most marketers are stuck in:<ul><li>Last-minute campaign tweaks ("The creatives aren’t performing—panic!") </li><li>Endless meetings ("Let’s discuss the meeting about the meeting.") </li><li>ROI justification ("The CFO wants to know why we spent $500K on TikTok.")</li></ul>Who has time for strategic measurement when you’re putting out daily fires?The Reality:MMM actually saves you time. Instead of guessing:<ul><li>You’ll know exactly which channels drive sales. </li><li>You’ll stop wasting money on underperforming tactics. </li><li>You’ll have data to shut down pointless budget debates.</li></ul>Lesson:&nbsp; You’re not too busy for MMM—you’re too busy NOT using it.<h3>Reason 5: “We’re Addicted to Last-Click Attribution - Help Us!” </h3>The Problem: Many marketers are hooked on:<ul><li>Digital attribution ("Look, this Google Ad led to a sale!") </li><li>Vanity metrics ("Our Instagram post got 10K likes!") </li><li>Short-term thinking ("Sales dipped this week—ABORT THE STRATEGY!")</li></ul>The Reality: Last-click attribution is like judging a movie by its last scene. MMM looks at the whole story, including:<ul><li>How TV builds brand awareness that later converts via search. </li><li>Why radio drives in-store purchases that never get tracked online. </li><li>When social media buzz leads to sales three months later.</li></ul>Lesson:&nbsp; If you’re only measuring what happens right before a sale, you’re missing 90% of the picture.<h3>The Bottom Line: Stop Making Excuses, Start Using MMM </h3>For Media Planners:Stop guessing which channels work best. Use MMM to allocate budgets with confidence.For Brand Managers:Stop relying on gut feel. Use MMM to prove your campaigns actually drive brand awareness.For Marketing Managers:Stop wasting time on pointless debates. Use MMM to shut down budget arguments with data.For Everyone Else:Stop pretending you don’t need MMM. You do.<h3>What’s Your Excuse?</h3>✅ "Our data is messy." → We’ve seen worse.✅ "It’s too complicated." → Not the way we do it.✅ "We tried it once and nothing changed." → Then you did it wrong.✅ "We don’t have time." → You don’t have time NOT to.<h3>Final Thought </h3>The brands winning right now aren’t the ones with the biggest budgets—they’re the ones using MMM to spend smarter.So… when are you starting?<h3>Key Takeaways - For Those Who Skipped to the End </h3><ul><li>No one has perfect data. Start with what you’ve got. </li><li>MMM shouldn’t feel like a math exam. If it does, get a better provider. </li><li>Insights are useless if you ignore them. Actually use the findings. </li><li>Last-click attribution is lying to you. MMM tells the full story. </li><li>The best time to start MMM was last year. The second-best time is now.</li></ul>Ready to stop guessing and start measuring? Let’s talk.

5 Real Reasons Marketers Avoid Marketing Mix Modeling - Why They're Wrong

<h3>The Challenge: Untangling a Complex Full-Funnel Strategy</h3>In crowded insurance market, where consumers shop across multiple brands before selecting a policy, one major car and home content insurer faced a pressing question:"How can we generate more insurance quotes without significantly increasing our marketing budget?"The brand was running one of the industry's most sophisticated media mixes, spanning:<ul><li>Traditional channels: TV, radio, outdoor, print, cinema </li><li>Digital channels: Search, social, display, online video, affiliates</li></ul>But three critical challenges stood in their way:1. Data Disorganization at ScaleDespite working with established media agencies, the brand's media data was trapped in:<ul><li>Inconsistent Excel files with mismatched date formats </li><li>PDF reports that required manual extraction </li><li>Disconnected affiliate tracking systems</li></ul>The marketing team was spending 40% of their time on data wrangling rather than strategy.2. Granular Measurement RequirementsThe key metric—insurance quotes—needed to be analyzed across multiple dimensions:<ul><li>Product type: Car insurance vs. home &amp; content </li><li>Demographics: Different conversion patterns for 25-40 vs. 41-65 age groups </li><li>Geography: Varying performance across states</li></ul>Traditional marketing mix models couldn't handle this level of detail.3. Budget Optimization Under ConstraintsWith a mandate to limit budget growth to just 3% annually, the team needed to:<ul><li>Identify the most efficient channels for each product / age group / state </li><li>Predict how reallocations would impact quote volume </li><li>Maintain brand-building while driving immediate quote submission</li></ul><h3>The Solution: A Custom Hierarchical Measurement Framework </h3>1. Streamlining the Data Foundation More Than Data implemented a robust data processing system that:<ul><li>Automated ingestion of agency reports using Python-based cleaning scripts </li><li>Established consistent cost tracking across all channels </li><li>Integrated affiliate data with owned media performance</li></ul>This reduced data preparation time from weeks to days.2. Multi-Dimensional Performance ModellingThe custom-built marketing mix model accounted for:<ul><li>Channel effectiveness by product type (e.g., search worked better for car insurance) </li><li>Regional variations (OOH performed best in metro areas) </li><li>Age-based conversion patterns (social drove younger quotes, radio influenced older demographics)</li></ul>3. Key Insights Uncovered <ul><li>Cinema ads generated 3% more car insurance quotes among 25-40 year old </li><li>Radio drove 8% higher conversion rates in regional Queensland </li><li>Affiliate partnerships accounted for 18% of high-intent quotes</li></ul>4. Dynamic Budget Allocation System The team could now:<ul><li>Test scenarios like "What happens if we shift 15% from print to YouTube?" </li><li>See predicted quote impact before making changes </li><li>Set budget rules and constraints to protect brand-building channels</li></ul><h3>The Results: Doing More With the Same Budget</h3>✅ 11% increase in marketing efficiency (surpassing the 6% target)✅ 5% more quotes generated with just 3% additional spend✅ Ongoing partnership as the brand's marketing effectiveness advisor"Before working with More Than Data, we were making million-dollar decisions based on incomplete information. Their hierarchical modelling approach finally gave us the clarity we needed to optimize across products, regions and age groups. The 11% efficiency gain wasn't just a number—it translated directly to our bottom line."<h3>Why This Matters for Insurance Services Marketers </h3><ul><li>Insurance Purchases Are Complex – Different products appeal to different demographics through different channels </li><li>Every Dollar Counts – In a low-margin industry, small efficiency gains create big impacts </li><li>Static Models Aren't Enough – Consumer behavior changes constantly </li></ul><h3>The Real Competitive Advantage </h3>What set this solution apart wasn't flashy technology, but rather:<ul><li>Deep understanding of insurance purchase journeys </li><li>Rigorous methodology for hierarchical measurement </li><li>Practical tools that marketers could actually use</li></ul><h3>Key Lessons for Marketing Leaders</h3><ul><li>Clean data comes first – No model can overcome poor inputs </li><li>Granularity matters – Average performance hides important variations </li><li>Scenario testing is essential – The market changes constantly</li></ul>For brands ready to move beyond guesswork, this is how modern marketing measurement works.

Precision Marketing Measurement for an Insurance Leader

More Than Data

An Australia Sydney-based pioneer in marketing analytics, offering a free, fast, and user-friendly marketing mix modelling platform. We help marketers build actionable marketing mix models in under 10 minutes for instant insights. Our software Minute MMM is free, DIY, and simple—no costly consultants, complex tools, or data expertise needed. Optimize marketing investments with ease!

Mulish

Sora

Merriweather

Thank You to Our First 100 Clients and Users.
Your support helped shape our products and services—built to make media impact measurement accessible and affordable.
Starting 1 July 2025, Minute MMM and Budget Optimizer move to a paid model with monthly subscription and upfront payment options.
We're no longer free—but we're still powerful, affordable, and focused on your success.

Blog

The Open-Source MMM Paradox: Why Google and Meta’s Tools ...

Why Marketing Mix Modelling Hasn’t Been Adopted by Non-Te...

More Than Data Transformed a Men’s Health Brand’s Marketi...

Quick Guide to Building Your First Marketing Mix Model in...

5 Real Reasons Marketers Avoid Marketing Mix Modeling - W...

Precision Marketing Measurement for an Insurance Leader