12 MarTech Questions Answered | The MarTech Matrix

Q: What's the difference between MMM, MTA, and incrementality testing?

Three measurement approaches, three different questions. Marketing Mix Modeling (MMM) uses statistical regression on top-down data (sales, spend, external factors) to answer "what's working at the macro level?" Multi-Touch Attribution (MTA) uses bottom-up user-level data to answer "which touchpoints contributed to this conversion?" — but it's been gutted by privacy changes and walled gardens. Incrementality testing runs controlled experiments (geo-holdouts, ghost ads) to answer "would this conversion have happened anyway?" Most sophisticated programs use all three: MMM for budget allocation, incrementality for validation, MTA only where the data is clean.

Q: How can brands measure marketing without third-party cookies?

Three plays, used together. First: invest in first-party data — customer accounts, email, loyalty, on-site behavior, anything you own outright. Second: use modeled data — MMM and incrementality testing don't depend on individual-level cookies. Third: use clean rooms and ID partnerships — Google's Ads Data Hub, Amazon Marketing Cloud, and walled-garden audiences let you measure within their environment without exporting individual data. The brands winning right now treat measurement as a portfolio of methods, not a single source of truth.

Q: How does AI personalize eCommerce experiences?

AI personalization in eCommerce works across three layers: discovery (recommending which products to surface based on intent signals and similar shopper behavior), presentation (changing site copy, hero images, and product order per visitor), and post-purchase (replenishment, cross-sell, retention triggers). The brands getting it right aren't generating individual experiences from scratch — they're segmenting shoppers into 8-20 high-confidence cohorts and serving each one a tuned version of the experience. Pure 1:1 personalization is expensive, fragile, and rarely measurably better than tight cohort targeting.

MarTech questions deserve straight answers — but the MarTech vendor landscape is messy. The vendor landscape is bloated, the jargon is exhausting, and every analyst report contradicts the last one. This page answers the questions brands and agencies actually ask us — distilled from 40+ podcast conversations with the people building, buying, and breaking MarTech. Got a question we haven’t answered? Drop it below and we’ll cover it on the show.

Submit a question

← Back

Thank you for your response. ✨

Common MarTech Questions Answered

What is MarTech?

MarTech (marketing technology) is the stack of software brands and agencies use to plan, execute, measure, and optimize marketing. It spans everything from your email platform and CRM to attribution tools, ad platforms, personalization engines, and AI-driven creative. The category has exploded — the MartechMap now tracks 14,000+ tools. Most marketers use 20-50 of them, and the work is less about picking individual tools and more about getting them to talk to each other.

Learn more: AI in MarTech hub · Harnessing AI for Customer Identity (episode) · Browse vendors on Blurbs

How do brands choose the right MarTech vendor?

Start with the business problem, not the category. Most vendor selection fails because the buyer started with “we need an attribution tool” instead of “we don’t know which channels actually drive new customers.” Once the problem is clear, three filters matter: does it integrate with your existing stack, can you see real outcome data (not just vendor case studies), and is the team behind it people you’d actually want to work with for the next three years. Skip the analyst-grid theater. Talk to actual customers.

Learn more: Vendor Spotlights hub · Best MarTech Podcasts 2026 · Find vendors without the noise on Blurbs

What’s the difference between MMM, MTA, and incrementality testing?

Three measurement approaches, three different questions. Marketing Mix Modeling (MMM) uses statistical regression on top-down data (sales, spend, external factors) to answer “what’s working at the macro level?” Multi-Touch Attribution (MTA) uses bottom-up user-level data to answer “which touchpoints contributed to this conversion?” — but it’s been gutted by privacy changes and walled gardens. Incrementality testing runs controlled experiments (geo-holdouts, ghost ads) to answer “would this conversion have happened anyway?” Most sophisticated programs use all three: MMM for budget allocation, incrementality for validation, MTA only where the data is clean.

Learn more: Insighta on Marketing Measurement · Modern Media Measurement (Data Speaks) · Incrementality Measurement

What is incrementality testing in marketing?

Incrementality testing isolates the true lift of a marketing investment by comparing exposed groups to controlled holdout groups. The classic version is a geo holdout — turn ads off in a matched region for 4-6 weeks, measure the gap in conversions vs. the region where ads stayed on. The harder it is to fake, the more reliable the answer. ROAS and last-click attribution can’t tell you what would have happened without the ad. Incrementality can. The trade-off: it’s slower and requires marketing teams to give up some near-term spend to learn.

Learn more: Incrementality Measurement episode · Kill the ROAS Crutch · Audiohook on Blurbs (incrementality-based audio CPA)

How can brands measure marketing without third-party cookies?

Three plays, used together. First: invest in first-party data — customer accounts, email, loyalty, on-site behavior, anything you own outright. Second: use modeled data — MMM and incrementality testing don’t depend on individual-level cookies. Third: use clean rooms and ID partnerships — Google’s Ads Data Hub, Amazon Marketing Cloud, and walled-garden audiences let you measure within their environment without exporting individual data. The brands winning right now treat measurement as a portfolio of methods, not a single source of truth.

Learn more: A Window into the Walled Gardens · Future Proofing Your Data · Modern Media Measurement

What is connected TV (CTV) advertising?

CTV advertising places video ads on streaming services delivered through internet-connected TVs — Roku, Fire TV, smart TVs, gaming consoles, and streaming apps like Hulu, Disney+, Tubi, Pluto. Unlike linear TV, CTV is addressable: brands can target specific audiences (households, geos, behaviors) and measure outcomes with digital-style tracking. The ecosystem is fragmented across dozens of platforms and inventory sources, which is why working with experts who have direct publisher relationships matters more than just plugging into a DSP.

Learn more: CTV: Thriving, Confusing, and Ready for Clarity · Current State of CTV Advertising · ClearTrust Media on Blurbs (CTV inventory)

What is programmatic DOOH (digital out of home)?

Programmatic DOOH is the automated buying of digital out-of-home media — billboards, transit displays, gym screens, gas pump TVs, restaurant menu boards — through demand-side platforms. The screens are connected, the inventory is dynamic, and ad serving can be triggered by location, weather, time of day, traffic data, or audience proximity. The format works for awareness at scale and increasingly for performance when paired with mobile retargeting and footfall measurement.

Learn more: The New Frontier: Digital Out of Home · ClearTrust Media on Blurbs (DOOH)

How does AI personalize eCommerce experiences?

AI personalization in eCommerce works across three layers: discovery (recommending which products to surface based on intent signals and similar shopper behavior), presentation (changing site copy, hero images, and product order per visitor), and post-purchase (replenishment, cross-sell, retention triggers). The brands getting it right aren’t generating individual experiences from scratch — they’re segmenting shoppers into 8-20 high-confidence cohorts and serving each one a tuned version of the experience. Pure 1:1 personalization is expensive, fragile, and rarely measurably better than tight cohort targeting.

Learn more: Personalized eCommerce Experiences · The Future of Retail (FindMine) · Harnessing AI for Customer Identity · FindMine on Blurbs

Why is ROAS misleading and what should performance marketers measure instead?

ROAS (return on ad spend) measures revenue against ad cost but ignores three things: incremental contribution (would the customer have bought anyway), gross margin (a 4x ROAS at 20% margin loses money), and customer lifetime value (a discounted first purchase can have a great ROAS and terrible LTV). The fix isn’t to throw ROAS out — it’s to layer profit-based metrics (contribution margin per acquisition), incrementality lift, and payback period on LTV alongside it. Optimize against profit, validate against incrementality.

Learn more: Kill the ROAS Crutch: Build a Profit Stack · See What Your Ads SOLD (Attain) · Incrementality Measurement

How can eCommerce brands use UGC at scale?

User-generated content (UGC) at scale isn’t about reposting customer photos — it’s about building a creator network you can brief, source from, and measure. The brands winning treat UGC like a content engine: a steady-state roster of creators producing on-brand assets that get tested across paid social, email, PDPs, and retention. The investment pays back when UGC outperforms studio creative in paid (which it usually does) and reduces creative production cost per asset.

Learn more: Inside the Blurb: Cohley UGC Engine · The Power of UGC in Marketing & eCommerce · Cohley on Blurbs

How can brands improve eCommerce site speed and conversion?

Three places to look. Core Web Vitals: largest contentful paint, interaction to next paint, cumulative layout shift — get these into the green and Google rewards you with rankings. Time to interactive: the moment your site is actually usable, not just visible. Heavy product images, third-party scripts, and bloated app frameworks are the usual culprits. Edge delivery: moving rendering closer to the user via CDN-level personalization can cut load time in half on mobile. Every 100ms shaved off load time correlates with measurable conversion lift.

Learn more: Website Speed Matters (Nostra) · Cost of Ignoring Website Friction · eCommerce Tech hub

What is “Complete the Look” merchandising?

Complete the Look (CTL) is a merchandising strategy that shows shoppers full outfits or coordinated product sets instead of individual items. The use case spans PDPs (“here’s what works with this jacket”), landing pages (seasonal looks, trend stories), and ad creative (carousels that show a full look click-through to a curated landing page). For apparel and home goods, CTL drives higher average order value, lower returns, and clearer brand storytelling than individual product display. The technology side maps trends, color theory, and brand-specific style rules to your live product catalog.

Learn more: Complete the Look Merchandising · The Future of Retail (FindMine) · FindMine on Blurbs

Have a different question?

Submit it above — or explore our topic hubs: Vendor Spotlights, eCommerce Tech, Brand-Side Stories, and AI in MarTech.