A recent piece from my colleague got me thinking about AI's impact on the SaaS world that came before it, and I finally feel like I can see clearly what's missing from it.
This is the first in a series of posts about Azalea, and I want to start with the thing that matters most: what it actually provides. Not a feature list, but the specific gap in every CMO's day that nothing else is filling. Here's how we got there.
SaaS started off as a simple idea: applications in the cloud instead of on-prem. Somewhere along the way, it became the default way to consume information technology. Need to run your finance or marketing function? You went and found a SaaS solution to do it, and often that worked well: when the underlying job is truly universal (everyone runs payroll the same way, everyone processes an invoice the same way), a single, uniform piece of software genuinely is the better answer. SaaS might offer some configuration knobs, but underneath, you were running the same solution as everyone else. That's fine when the underlying job is the same for everyone. But when it isn't there is a mismatch.
Marketing is where that mismatch shows up. SaaS got so standardized that product and UX teams stopped asking "how do I build a system that works the way my customers actually run their business?" and started asking "how do I get users to intuitively understand how our application works?" That's a subtle but important shift: it optimizes the software for the average buyer instead of any one team's actual workflow. Marketing organizations bought expensive platforms built for that average buyer, and then quietly struggled to make them fit. It shows up in the numbers. Gartner's 2025 Marketing Technology Survey found that marketers now actively use only 49% of the tools in their stack, and just 15% of organizations qualify as "high performers" seeing real ROI from what they've bought. I don't think this is a training problem. It's the uniform-software, non-uniform-job mismatch playing out at scale.
AI, paired with protocols like MCP that let it actually reach into your systems, offers a way out of that bind. Instead of being boxed in by whatever workflow a monolithic platform assumed you'd have, a marketing team can use AI to knit together the processes and systems they actually have, to create, execute, and measure campaigns on their own terms. That's a real productivity unlock: data that used to be locked away in a system nobody could query becomes accessible on demand, and moving between systems to actually get a campaign live becomes far less painful.
But only up to a point. If every staff member on the team builds their own process, their own dashboard, their own definition of what "working" looks like, the CMO doesn't get freedom, they get ten people with ten different definitions of ROI and no way to reconcile them. The same flexibility that solves the SaaS mismatch problem creates a new rationalization problem if nothing holds it together. The unlock is real, but it needs structure, particularly around the one thing that's always been hardest to unify: tracking spend and results across systems that were never built to talk to each other. Today, doing that well takes data engineering expertise that simply doesn't sit inside most marketing teams, and it's rarely available to the CMO on demand.
So we asked a different question: what if we gave CMOs an analytics application built around their specific requirements, instead of the average requirements of every buyer? And what if we brought the idea of a forward-deployed engineer (someone who configures the system around your processes rather than asking you to adapt to its processes) into marketing analytics itself?
That's the heart of Azalea. Concretely, here's what it provides: a single, standing view that connects marketing spend to sales and revenue, built around how your team actually works rather than a template every buyer gets. No data engineering hire required to get there. No twelve-tab reconciliation before the Monday meeting. One system of record for "is this working," built once, for your business specifically, and kept current automatically as your campaigns and pipeline change.
SaaS gave every team the same software and asked them to adapt to it. Azalea gives every CMO their own answer to "what's working," built around their business, without the chaos of everyone on staff building it their own way.
We built Azalea because we needed it and couldn't find anything else that actually did it. The best products tend to start that way, don't they, not as a plan to build a company, but as the tool you wished already existed. This post is where we start talking about it more broadly, and we'll keep building on it from here.