The pilots stall. The dashboards multiply. Governance says no, so the business routes around it. I've spent my career fixing exactly this — building the data, AI, and governance foundations that let enterprises move fast with confidence, and killing everything that doesn't earn its place.
I'm Jonathan Salmon — Director of Enterprise Intelligence at Centene, leading a 15-person team in one of healthcare's most demanding data environments.
Different companies, same three failures. If any of these sound familiar, we should talk.
Hundreds of reports nobody opens, each one taxing the platform and the people who maintain it. Waste compounds.
AI demos that never reach production because nobody owns the path from experiment to product. The budget line is on borrowed time.
Controls designed to say no instead of enabling speed. So the business routes around them, and the real risk goes underground. Good governance is automated and invisible: guardrails, not gates.
What I've actually owned and delivered — the kind of scope a VP remit is built on.
Lead a 15-person enterprise intelligence organization inside a Fortune-scale healthcare company — regulated data, real stakes, no room for vanity metrics. Two years at director level and counting.
Own the platforms and the reporting estate. My rule is simple: every report and dashboard must earn its existence — or it goes. High-utility core products, built with best practices, replace sprawl.
Known for translating strategy into executable plans and empowering strong engineers rather than hovering over them. Talent development and mentoring are the multiplier on everything else.
Geodynix — U.S. market-intelligence product, live in production. Profit Sentinel — B2B SaaS for profit-leak intelligence. I don't just govern data platforms; I ship product on them.
The beliefs I actually manage by — not poster slogans.
Hire well, set direction, clear obstacles, get out of the way. Control is the enemy of scale.
Say the thing — to your team, your peers, and up. Ambiguity is where careers and projects quietly die.
Reports, dashboards, meetings, metrics: prove value or kill them. Waste compounds; so does focus.
Your team's growth is your output. Mentoring isn't charity work — it's how you build an organization that outlasts you.
Ask earlier. Advocate louder. Manage up on purpose. Power dynamics are real; pretending they aren't is not a strategy.
Most value concentrates at the edges; most effort bleeds into the middle. Find the vital few inputs, starve the trivial many, and treat waste as the enemy — because it is.
A straight path from reading to conversation.
Read the AI Signal — five stories every morning, each with my take. You'll know within a week whether I think the way you need.
Email me directly. I read everything myself, and I make time for good conversations.
Fractional, advisory, or board conversations. A straight conversation about your problem.
Five stories a day from a short list of sources I actually read — each with my one-line take. This is how I stay in the conversation.
Updated Friday, October 30, 2026
My take: An MIT Technology Review Insights poll finds every organization surveyed is using AI, but fewer than one in five, 16 percent, are extracting transformative value, and just 12 percent have agents in production. The missing layer named here is context: governed, machine-usable enterprise knowledge sitting between business systems and AI platforms. That is the 80/20 of enterprise AI. Most spend chases models and tools, while most value concentrates in the context layer that tells those tools what the business actually means.
My take: OpenAI reorganized its public Model Spec around five values: faithfulness to human intent, harm minimization, honesty and transparency, human agency, and care, with new guidance on context, oversight, and side effects. A vendor publishing its behavior rulebook in public is governance you can actually test against. Leaders should map their own use policies to it, and note OpenAI's own admission that its models do not yet fully reflect the Spec.
My take: Willison highlights Spawn, a game company whose agents change live code and ship to production hourly, at a reported cost of over $3 million in tokens per month, described as 30x payroll, with an hour without a ship treated as a SEV-0 incident. It works there because rollback takes 30 seconds and the stakes are a game. Copy the recovery discipline if you like, but in higher stakes work the approval gates and audit trail are not overhead, they are the product.
Founder work I run outside my day job — where I stay close to the craft of shipping.
U.S. market-intelligence product unifying demographic, economic, housing, and business data — sold as a Snowflake data product and a self-serve insights app.
B2B SaaS: continuous profit-leak intelligence for mid-market trade contractors, auditing the seams between their operational and financial systems.
I'm writing on AI in the enterprise, data leadership, and organizational power — publishing on Medium as part of a deliberate content practice. The AI Signal above is the daily version; the essays are the deep version.
Latest: AI Made Building Free. It Didn't Make Deciding Free. — why review, not output, is the scarce skill now.
I also keep a public reading list — the books that shaped how I think.
I make time for good conversations — with data and AI leaders, founders who want an advisor, or peers working on the same hard problems. I read everything myself.
jonathan.salmon@gmail.comRésumé available on request.
Jonathan Salmon