For a long time, the status of every venture I was running lived in my head, scattered across a few documents, a handful of notes, and whatever I could remember from the last time I'd checked in on something. It worked, barely, when there was one main thing to track. It stopped working the moment there were several ventures moving at once, each with its own state, its own open questions, and its own next steps.
The problem wasn't that I lacked information. The problem was that the information lived nowhere consistent. Ask me about one venture on a Tuesday and I could give you a sharp answer. Ask me about a different one three weeks later and I'd have to reconstruct the whole picture from memory before I could say anything useful. That's not a sustainable way to run a portfolio, and I knew it long before I fixed it.
Why scattered context is more expensive than it looks
The cost of not having a single source of truth isn't just inconvenience. It's decision quality. When you're making a call about a venture based on a half-remembered status instead of an accurate one, you're more likely to repeat a mistake, miss a dependency, or waste effort on something that's already been addressed. Multiply that across a dozen active efforts and the compounding cost becomes real, even if it never shows up as a single visible failure.
There's also a trust cost. If I can't quickly and confidently state where a venture actually stands, I can't hand off a piece of that work to anyone or anything else, human or automated, without a lengthy re-briefing every time. That re-briefing is exactly the kind of overhead that kills momentum when you're trying to move fast across multiple fronts.
What I actually built
The fix was less exciting than it might sound and more valuable than I expected. I built one operating repository, a single place that holds the current status, roadmap, and key context for every venture, kept current instead of stale. Not a wiki that rots after the first month. An actively maintained source that every workflow, every automated check, and every planning session pulls from.
The discipline mattered more than the tooling. Any reasonably organized system, a shared doc, a lightweight database, a structured repo, can serve this purpose if it's actually kept current. What matters is committing to update it every time something changes, and treating "check the source of truth first" as the default habit before making any real decision.
Once that existed, a few things changed immediately. Planning sessions got shorter, because half of a planning session used to be reconstructing where things stood. Automated systems and AI-assisted workflows got dramatically more useful, because they could work from accurate current context instead of a stale mental model I was feeding them inconsistently. And handoffs, even the internal kind where I'm just resuming my own work three weeks later, got faster and more reliable.
The lesson that came two years too late
I should have built this the moment I had more than one serious project running at once. Instead, I tolerated the scattered-context tax for a long time because building the system felt like overhead on top of already-full days. That's backwards. The system is what makes the full days sustainable instead of just chaotic.
If there's a single piece of advice in this for anyone else trying to run more than one thing at a time: build the single source of truth before you think you need it, not after the scattered version has already cost you real time and real decisions. It's not a glamorous project. It's one of the highest-leverage things I've built this year.
What almost stopped me from building it
The honest reason I delayed this for so long is that it felt like time taken away from the actual ventures. Every hour spent building an internal system is an hour not spent on something a customer or reader would see. That framing is exactly backwards, but it's a very natural way to think when you're busy and every venture is asking for attention. What finally changed my mind was noticing that the scattered-context tax was costing more hours across all the ventures combined than building the system would ever cost once.
Keeping it from becoming its own burden
There's a real risk that a single source of truth becomes just another thing to maintain, one more surface competing for attention instead of reducing the number of surfaces. I've tried to guard against that by keeping the format simple and the update habit tied directly to the natural rhythm of working on a venture, updating it as part of finishing a work session rather than as a separate scheduled task. If maintaining the source of truth ever starts to feel like its own project, that's a sign it's gotten too complicated for what it actually needs to do.
How this connects to everything else I've built since
Every other system I've built this year, the automated checks, the AI-assisted workflows, the repeatable content processes, depends on this one existing and staying accurate. None of those would be trustworthy if they were pulling from stale or inconsistent context. That's the real argument for building this first, before any of the more visible automation. The flashier systems only work as well as the foundation underneath them, and this is the foundation.
