Winding down the MessageWorks platform — and what we're building instead

We recently made the decision we'd been circling for months: MessageWorks is winding down its product business. We've stopped developing and supporting the platform. Our customers were told directly, and their commitments are being honored. And the company is pivoting to the thing I've done for most of my career — services.

This post is for everyone else we built relationships with along the way: the marketing and go-to-market leaders who took our calls, tested our thinking, argued with our roadmap, and made the product better. You deserve more than a quiet redirect on the homepage. Here's what happened, what we learned, and what comes next.

The two assumptions

MessageWorks was built on two assumptions.

The first: that product marketing and demand generation teams have urgent messaging-alignment and validation needs. AI made it trivial for anyone in a company to generate content, and nearly impossible to keep the story coherent. Positioning lives in slide decks; the tools optimize for plausible copy, not your actual strategy; and when the strategy shifts, the new deck exists while every team and every AI tool keeps producing from the old one. If you've lived it, you know.

The second assumption: that a SaaS platform was the right way to solve it.

The first assumption proved very true. The second did not. AI moved faster than our platform vision — in nine months, the workflows and toolchains marketers actually use shifted materially, and kept shifting. There was still a path to building it, but it ran through a few million dollars of seed financing. I've self-funded this company from the start, and raising venture money to chase a platform target that moves every quarter is a fine decision for someone. It wasn't mine.

So we made the call the moment it became clear, rather than two years and a fund-raise later.

What the platform taught us

Here's the uncomfortable part: the lesson was sitting inside our own product thesis the whole time.

You've seen the headline — MIT's finding that 95 percent of AI pilots fail (a small early study, and its own authors call it preliminary). There's a quieter statistic on the other end: FTI found that 95 percent of private equity funds with AI programs actually in production report those initiatives meeting or beating their business case. Both are true. They're two ends of one funnel. Most AI efforts die before they reach production; most of what reaches production pays. Which means the game is played before launch — in choosing which problem, which use case, which way, and whether at all.

MessageWorks was our own trip through that funnel. The technology was never the hard part. The hard part was the selection — connecting AI capability to a strategy, and to numbers someone would defend in a board meeting. We started this company believing AI should do more than make teams faster; it should make them better. We still believe that. We just learned that the leverage isn't in shipping another tool. It's in the judgment about where AI actually creates value — and the unglamorous work of building it into how a business runs.

What comes next: Stetson Value Creation Partners

That judgment, and that work, is the new firm: Stetson Value Creation Partners (stetsonvcp.com).

Stetson works with private equity firms and their portfolio companies — the arena where the "does AI actually move the P&L" question gets asked with real money attached. But the work itself will look familiar to anyone who talked with us at MessageWorks: honest AI-readiness diagnosis, roadmaps that start with where AI pays and where it doesn't yet, custom workflow implementation, agent deployment, and automation for the places where standard tools fall short. We build what we recommend.

Two things make it different from the AI consulting you've probably been pitched. First, diagnosis comes before building, always — and the diagnosis is allowed to conclude "not here, not yet." A recommendation that can't say no is a sales document. Second, the economics are shaped to keep it honest: the early reads are free, because that's how we choose the clients we build with long term, and we only earn when a company decides something is worth building. The How We Work page lays the whole engagement out plainly — what's free and why, what your hours look like, and where we earn.

If this maps to your world

Many of you lead marketing or GTM at software and technology companies, and you're carrying some version of the same mandate: leadership wants AI to show up in the numbers, not just in the tooling. If you're staring at a pile of pilots and wondering which of them will ever survive contact with a P&L — that's precisely the work now. Start with the How We Work page, and if it reads true, book the 30-minute framing call (Note: while our go-to-market is built around the private equity universe, the value we can provide is independent of a company’s ownership structure). Worst case, you get an honest no in thirty minutes.

To everyone who believed in MessageWorks enough to spend time on it: thank you. The problem we set out to solve is real, and someone will solve it well. We learned more from building — and from stopping — than we ever could have from almost building. That lesson is now the business.

— Marshall

‍