I Told You So, Part II: Apparently There's a Price Tag Now
- Rich Washburn
- 20 minutes ago
- 6 min read
Well. That escalated quickly.
I keep saying I'm not an "I told you so" guy. I promise I'm not that guy. It's just getting harder to make that argument with a straight face. Because this is now the second time in about a week I've put "I told you so" in a headline. And the first article already had me saying it again inside the article.
So, yeah. This is getting a little silly.
A few days ago I wrote "I Told You So: McKinsey Just Proved the Value of Everything You Know."
The point was pretty simple.
Companies have spent decades obsessing over the data sitting neatly inside systems while ignoring a gigantic pile of value sitting in plain sight: everything people know about how the place actually works.
The conversations...The weird exceptions...The workarounds...The arguments...The mistakes...The stories.
The thing Bob knows because Bob's been doing this for 27 years and nobody ever thought to ask Bob to write it down.
I called it human operational data.
McKinsey basically came along and said, yeah, this matters.
Great.
Point made. And then Google apparently said: cool, how much for Spirit's?
Answer: $10 million.
I mean, come on. That's funny.
Google won a bankruptcy auction for a massive chunk of Spirit Airlines' internal business data with a $10 million bid. And the thing that matters here isn't really Google.
It isn't really Spirit. It isn't even the $10 million. It's what they're buying.
Approximately 100 million emails.
Roughly 500 million Microsoft Teams messages.
Around 30 million lines of code.
Spreadsheets.
Calendars.
Internal documents.
Software.
Operational information.
Pricing information.
Marketing information.
HR information.
Basically decades of digital exhaust from thousands of people trying to run an airline. Except, apparently, it's not exhaust anymore.
That's the whole thing.
Google isn't really buying 100 million emails because emails are exciting. It's buying what's buried inside them. It's not buying Teams chats because chat bubbles are valuable. It's buying the trail between a problem and the decision that came out of it.
Somebody notices something's wrong.
Somebody sends an email.
Someone else says that won't work.
A meeting happens.
Someone builds a spreadsheet.
Operations says the spreadsheet's wrong.
Finance changes an assumption.
IT discovers some cursed legacy system from 2008 is still quietly holding the whole thing together.
Somebody creates a workaround.
The workaround becomes a process.
Nobody remembers why.
Three years later somebody asks, "why do we still do this?"
And somebody else says, "because of that thing that happened in 2019."
There it is.
Institutional memory.
That's the thing. And the funny part is, this didn't actually start with the McKinsey article last week. Unfortunately for everyone, including me, I have receipts.
Back in October 2025, I wrote "The Inevitable Obsolescence of Consulting Firms."
Subtle title.
Very restrained.
In that one, I was talking about distributed cognition.
The fact that companies have knowledge scattered everywhere.
Inside senior people.
Inside conversations.
Inside weird little exceptions.
Inside judgment.
Inside things nobody wrote down because everybody just "knew" them.
And I wrote that firms should be extracting that tribal knowledge like it was the most valuable asset they own.
That was October.
Then last week, McKinsey helped validate the idea.
And now Google's standing over Spirit Airlines with a $10 million check.
So, yeah.
Cue the carnival music.
AI has read what we wrote. Now it wants to see how we work.
That's the actual shift.
The internet can teach an AI what airline revenue management is. Books can explain it. Research papers can explain it. Consulting reports can explain it. Documentation can explain it.
All useful.
But Spirit's internal history can potentially show something very different: what happened when revenue management met staffing.
Or maintenance.
Or software limitations.
Or customer behavior.
Or operations.
Or some system that was supposed to integrate with another system and absolutely did not.
That's real work. And real work is mostly edge cases.
It's Jane in accounting needing column F formatted a certain way because it feeds something nobody mentioned.
It's Frank in operations manually correcting something every Thursday because the vendor system's been wrong since 2021.
It's the API documentation saying one thing while the API does another.
It's somebody quietly knowing that if you restart Server Three before Server Two, everything breaks.
That stuff isn't glamorous.
It's also exactly the stuff that separates a nice-looking AI output from something that actually works inside a company.
This is why I keep talking about the corpus.
The model isn't the moat.
I keep saying this because I think it matters. Eventually everybody gets access to very capable models. Then the question becomes: what does your model know that mine doesn't?
What context does it have?
What history does it understand?
What exceptions does it know?
What decisions can it trace?
What institutional memory can it retrieve?
What does your organization know about itself that exists nowhere else?
That's where the value moves. And Spirit's such a bizarrely perfect example because the airline can fail.
The airplanes can be sold. The gates can be reassigned. The employees can leave. The signs can come down. And the accumulated digital record of people trying to operate the business still survives.
Apparently with a market price.
That's a pretty weird new thing.
The company can die. The cognition doesn't have to.
That might actually be the line. Because the knowledge didn't suddenly become valuable when Google showed up.
It was already valuable. The difference is we now have machines capable of consuming enough of it to make the value obvious.
So something that used to be fuzzy becomes measurable. And once it becomes measurable, somebody eventually puts a number on it. That's what AI's doing here. It's making institutional knowledge economically legible.
Which brings me back to the older consulting piece.
I wrote: "This isn't about robots replacing consultants. It's about machines remembering what humans forget."
That's still basically the whole story.
The risk was never just automation. It was amnesia.
Companies forgetting themselves.
Losing the judgment... Losing the reasons... Losing the context... Losing the strange little operational truths nobody thought were worth preserving.
And now we're seeing the inverse.
Memory has value.
Apparently, in this case: $10,000,000.
So, yeah. I told you so. Again.
I know.
I know.
I'm trying not to become that guy. This isn't helping.
But the sequence is now: Last year: preserve the knowledge, it's an asset.
Last week: McKinsey says institutional knowledge is becoming a critical AI layer.
This week: Google bids $10 million for a bankrupt company's accumulated work history.
At some point I'm allowed to point at the screen for a second.
That's all I'm asking.
Then we move on.
Because the bigger question isn't whether every company has a $10 million archive sitting around.
Obviously not.
The bigger question is whether founders, boards, investors and executives are starting to understand that there's a category of value inside companies that most of them have barely bothered to preserve.
Everything the company learned while doing the work.
Everything people figured out.
Everything they argued about.
Everything they fixed.
Everything they stopped doing.
Everything they learned never to do again.
Everything they know.
We spent decades digitizing business. We thought we were creating systems of record. What we may have accidentally created was something much more interesting: a recorded memory of how the business thinks.
Last year I said preserve it. Last week McKinsey helped make the case. This week Google put a price tag on it.
So, yeah.
I told you so. Again.
I promise I'm not making this a series.
Probably.
Sources
Reuters — Google to buy Spirit Airlines business data for $10 million https://www.reuters.com/legal/litigation/google-buy-spirit-airlines-business-data-10-million-2026-08-17/ Google’s $10 million winning bid, intended use for product development and AI models, and details of the bankruptcy sale.
The Wall Street Journal — Spirit Flight Attendants Fight Google’s Data Bid for AI https://www.wsj.com/pro/bankruptcy/spirit-flight-attendants-fight-googles-data-bid-for-ai-e939e049 Covers the employee-privacy objections and the court’s decision to delay approval of the transaction until September 9.
ITPro — Google just spent $10 million on Spirit Airlines data https://www.itpro.com/technology/artificial-intelligence/google-just-spent-usd10-million-in-an-auction-for-spirit-airlines-data-100-million-emails-500-million-microsoft-teams-chats-and-30-million-lines-of-code-will-be-used-to-improve-ai-models-and-products Breakdown of the dataset: approximately 100 million emails, 500 million Teams messages, 30 million lines of code, spreadsheets, calendars, workflow/process data and internally developed software. Also reports Google’s $10 million bid beating Mercor’s $7.5 million bid.
Rich Washburn — I Told You So: McKinsey Just Proved the Value of Everything You Know https://www.richwashburn.com/post/i-told-you-so-mckinsey-just-proved-the-value-of-everything-you-know The immediate predecessor to this piece: institutional knowledge, human operational data and the argument for building the organizational corpus.
Rich Washburn — The Inevitable Obsolescence of Consulting Firms https://www.richwashburn.com/post/the-inevitable-obsolescence-of-consulting-firms The earlier thesis: distributed cognition, tribal knowledge as an organizational asset, and machines preserving what organizations otherwise forget.
Rich Washburn is a technologist, strategist, and Founder & Chief AI Architect of ARIA AI Labs, working at the intersection of AI, infrastructure, communications, and capital. He also serves as Managing Partner and Chief AI Officer at Eliakim Capital.
Comments