I just wrapped a conversation with Julian Modiano, founder of Acuto and Weavely. His entire business is built on a simple premise: platforms are becoming black boxes, and if you don't own your data, you're flying blind. What makes Julian different is he understands what marketers actually need because he's been one, and he knows what's technically possible because he builds the solutions himself.
Listen here: Spotify, Apple Podcasts, YouTube
Platform reporting isn't getting better. It's getting worse by design.
Julian was direct about it: "Apple on its own is just spearheading a complete crackdown on cookie tracking. The platforms are taking advantage of that to model conversions more and more. You kind of have to just trust that what the algorithms tell you is correct." When you add up conversions from Google and Meta, they never match what you're seeing in Shopify. Every platform wants to take credit, and the ones that do it the most are email and Pinterest with their 30-day lookback windows.
For 2026: Stop trusting platform dashboards as your source of truth. If you're making budget decisions based solely on Meta or Google reporting, you're optimizing for their business model, not yours.
You need both attribution models, even though they're both flawed
Julian's take on the MTA vs. MMM debate: "The short answer is it depends. You probably should be doing a little bit of both if scale allows." Multi-touch attribution gives you something instant and reactive for day-to-day decisions. It's not perfect, but it's better than platform reporting. MMM gives you a top-level view of where budget should go strategically. It won't tell you which ad performs better, but it will tell you if you should shift 20K from Facebook to Google.
For 2026: Use attribution for tactical decisions (which creative is working, should we kill this ad set). Use MMM for strategic decisions (which channels deserve more budget, where are the real opportunities). Trying to force one model to do both jobs is where brands get lost.
Data warehouses aren't enterprise tech anymore. They're table stakes.
Five years ago, only huge brands had the scale to justify a data warehouse. Now Julian sees brands and agencies of all sizes realizing they need one. The reason is simple: as platforms take more control away, the only competitive advantage left is having consolidated data that you own. AI makes this even more critical. You can't feed scattered data from five different dashboards into an AI tool and expect useful output. You need a foundation.
For 2026: If you're still duct-taping together Dash This or other skinned reporting tools, you're behind. The agencies that win will have their data consolidated in one place where they can actually build on top of it. This isn't optional anymore.
The AI advantage goes to whoever owns the most data
Here's what Julian sees happening: "What really sets agencies apart is having all of that data across multiple clients and for many years. You can consolidate all of this stuff in one place and use AI and all sorts of other tech that's much easier to build on top of that data foundation instead of having to rely purely on these black boxes." Brands can hire their own PPC specialists, but they can't replicate years of cross-client data. That's your edge, but only if you actually own it.
For 2026: Start thinking about your data as your competitive moat. Not your creative team, not your media buying skills, not your platform relationships. The agency with the deepest data foundation is the one that can leverage AI effectively while everyone else is feeding ChatGPT screenshots of spreadsheets.
Listen here: Spotify, Apple Podcasts, YouTube
Julian put it perfectly: platforms don't want to give you control. They're doing their own AI. They're modeling their own conversions. They're becoming more opaque by the quarter. The brands and agencies that survive 2026 aren't going to be the ones with the best platform relationships. They're going to be the ones who stopped trusting the black box and built their own data infrastructure.