Archives, Audits, & AI
Share
The Audit That Pays for Itself
Ask a mid-sized firm to fund a content audit and watch the budget conversation stall. Auditing what an organization actually has, who owns it, and what should survive a transition reads, to most finance leaders, as overhead: a line item that produces no product and closes no deal. Dr. Tyler Musgrave, an information scientist at Wayne State University and a former startup operations lead, thinks that reading is exactly backward.
Dr. Musgrave built the case the hard way. As the first operations employee at an early-stage virtual reality training company, she was, in her words, the concentration risk problem. Every process she built lived in her head or in whichever document happened to be open that year. When she left for a PhD, the company called two years later asking where things were. Nothing had failed. Nobody had deleted anything. The knowledge simply had nowhere else to live.
The bind, as Dr. Musgrave frames it, is that fortune-500 firms can absorb this cost because they already staff for it, while the nonprofit or the startup treats governance as a discretionary expense rather than a structural one. That is a mistake of accounting, not of intent. Every rebuild after a departure is redone work: hours spent reconstructing decisions that were never written down the first time. Dr. Musgrave calls this what it is, a direct hit to return on investment, dressed up as an HR problem.
Artificial intelligence sharpens the stakes rather than easing them. A model trained on scattered, unlabeled content will answer confidently from stale data or simply fail to find the institutional knowledge it needs. The firms positioned to benefit, Dr. Musgrave argues, will not be the fastest adopters but the ones that already know what they have. She lays out the fix as three unglamorous decisions: take inventory of what exists, name an owner for each collection, and schedule a recurring review of what should be retired. None of it requires new software. It requires someone willing to own it, the way a family's designated keeper of photo albums simply decides the job is hers.
Reframed this way, an audit stops looking like a cost center and starts looking like a hedge. Firms that skip it are not saving money. They are borrowing against a revenue model that gets more expensive to rebuild with every person who walks out the door.
Your work shouldn't leave when you do.
The HCD Guide Series walks you through documentation at every stage of a project, from discovery to measurement, so what you build doesn't die in your head or your inbox. Follow the method once and the next person who picks up your work won't be starting from zero.
Get the HCD Guides →
You can't grow a revenue model on a content mess.
An hour with Ana starts where the real risk lives: what you actually have, who owns it, and where the gaps sit. Audit first, organize second, fill what's missing third. That's the sequence that keeps revenue models compounding instead of quietly leaking.
Book Office Hours →
What we’re into this week
Aaron: Ohhhhhh THIS reminds me of the conversation with Jarrett Reed of PinkSlipped that we have coming up!
Ana: Oh yes, the part where Tyler reminds everyone that institutional knowledge walks out the door with every termination round?
Aaron: Bingo
Ana: Yeah, that occurred to me, too! Which is why Jarrett’s conversation is coming out September 15th! Going to be a really good follow-up to Tyler’s.
Scott: Are we supposed to be talking about this yet?
Ana: Jarrett’s episode? No, not technically, but it has to do with Tyler’s talk, so…yes?
Scott: Fair point, but I do have one thing to add that’s ACTUALLY what we’re into this week.
Tyler’s perspective reminds me of a very recent Economist article that asks, Will Anyone Use AI As Much As Coders Do? Their point is that if perfect, contained data is required for AI to be useful, who has that? Only people working in bounded worlds that can perfectly contain data. And that’s just not many of us.
Ana: But it’s you. And I think, to your point and the point of this article, all the implementation barriers that we’ve been talking about for a year now are really coming to surface.
Scott: That’s true. I do use AI a lot in my workflows. But it can be brittle. And it definitely creates a mess.
Ana: I hate a mess.
Aaron: That’s kind of the premise of Ishmael Interactive, no?
Ana: YEP!
Open Call
Do you know someone who would be a great interview for the CX Pod? Nominate them! Email abigail@ishmaelinteractive.com to let us know their name and why their work needs to be highlighted. We would really love to hear more from our community & highlight the great work that’s going on.
Share
Someone at your organization is the concentration risk right now. They know it. You probably do too. Forward this episode to them, or to whoever signs off on the budget that keeps pretending governance is optional.
Credits
Guest: Dr. Tyler Musgrave / Host: Sheev Dave / Line Producer: Aaron Meyers / Producer: Ana Monroe / Words: A. Adams