June has been a busy month for cheque-writing. Salesforce announced it was buying Tableau in an all-stock deal worth something north of fifteen billion dollars, and only days earlier Google had announced it was acquiring Looker for a couple of billion of its own. Two of the more prominent names in the analytics world, both swallowed inside a fortnight. If you build dashboards for a living, your week probably involved a few unsettled Slack messages.
I do not run Tableau in anger, but I do depend on a couple of tools whose corporate ownership I do not control, and these announcements landed in exactly the part of my brain that worries about that. So this is less a hot take on the deals and more an honest think about what it actually means when the company that owns a thing you rely on changes hands.
the immediate answer is "nothing changes," which is true and useless
The first thing every acquirer says is that the product will carry on, the team stays, nothing changes for customers. And in the short term that is genuinely true. Nobody buys a fifteen-billion-dollar asset and then breaks it on a Tuesday. The roadmap you were promised this quarter will almost certainly ship.
The reason it is useless reassurance is that nobody is worried about this quarter. The worry is the slow stuff: the gradual repricing, the bundling into a suite you did not want, the quiet deprecation of the standalone product two or three years out when the integration is "complete." None of that shows up in the press release, because none of that has been decided yet. It gets decided later, by people optimising for the acquirer's strategy, not for your particular use of the tool.
what i actually do about it
I have learned, mostly the hard way, that the right response to this is not panic and not complacency. It is a quiet audit. When a tool I lean on changes owner, I ask myself three blunt questions.
First: how locked in am I, really? Not "could I leave in theory," but "if I had to migrate off this in six months, what would it cost me in actual days?" The answer for a tool where my data lives in an open format and I script against a documented API is "annoying but fine." The answer for a tool that hoards my work in a proprietary blob and an undocumented export is "I am hostage and I should have known better." The acquisition does not change the lock-in. It just makes me finally measure it.
Second: is there an exit that I have tested? An export button I have never clicked is not an exit, it is a rumour. The only migration plan worth anything is one you have actually rehearsed, even once, even badly, on a copy of your real data. I now treat "can I get my data out and into something else" as a thing to verify on the way in, not the way out, precisely because the bad day always arrives unannounced.
Third: am I paying for the tool or for the company? If the value is the software and my data, an acquisition is mostly a billing nuisance. If the value is the company's ongoing attention, their support, their roadmap, their willingness to fix my edge case, then an acquisition genuinely threatens the thing I was buying, because the acquirer bought it for their reasons, not mine.
the broader pattern
Both these deals are the same move: a big platform absorbing the visualisation layer so the whole pipeline lives under one roof. It is a sensible strategy for Salesforce and Google. It is a slow consolidation for everyone else, and the practical effect is that the comfortable middle ground of "best-of-breed tool I bolt onto whatever I like" gets squeezed. The tools survive, but they survive as features of something larger, and features get prioritised differently from products.
I have lived through enough of these to recognise the arc, and it almost always runs the same way. Year one is reassurance and business as usual. Year two is "deeper integration," which sounds like a gift and frequently means the standalone version stops getting the love because the strategic value is now in the bundle. Somewhere around year three the pricing gets "simplified," which has never once in my experience meant cheaper for the way I happen to use the thing. And at some point the tool you bought because it did one thing brilliantly gets repositioned as a component of a platform you were never in the market for. None of that is villainy. It is just what happens when a product's owner changes from a company whose entire reason to exist was that product, to a company for whom it is one line item in a far larger strategy.
The open-source angle is worth a thought too, because it is the asymmetry that makes me sleep at night about some of my stack and not others. When the company behind an open tool gets bought, the worst case is a fork. It is painful and it splits a community, but the code does not vanish and neither does my ability to run it. When the company behind a closed, hosted tool gets bought, the worst case is that the thing simply stops existing on a timeline I do not control, and my data is only as portable as their export button. That is not an argument that open always wins, plenty of closed tools are worth the risk. It is an argument for knowing, before the acquisition, which kind of dependency each tool actually is.
I am not going to rip out anything over a press release, and neither should you. The day-to-day reality of these tools will not shift for a good while. But I did spend an afternoon this week confirming I could get my data out of the two tools I care most about, in a format I could actually use, without anyone's permission. Both passed, which is the only outcome that lets me ignore the next acquisition with a clear conscience. The lesson the analytics world handed me this June, free of charge: the time to test the exit is the boring Tuesday before you need it, not the day the logo on the login page changes.