Some of your software is about to become a service instead
Lawhive, Corgi and Eilla didn't out-feature their industries' software: they skipped it and did the job themselves. Here's how to tell which of your own SaaS bets are exposed to the same move.
Some of your software is about to become a service instead
Lawhive lets people get legal work done without hiring a solicitor first. Corgi sells insurance without a broker in the loop, and went from a YC batch to a $1.3bn valuation in four months doing it. Eilla runs M&A mandates end to end, the advisory work a boutique bank used to bill for by the hour. None of them out-featured the software already in that market. They skipped the software and did the job themselves.
Three different markets, the same move: instead of licensing a tool into someone else's process, own the outcome and the client relationship that comes with it.
Why owning the relationship beats owning the feature
A SaaS company's position depends on staying between the vendor and the customer's outcome, a fragile place to stand once the model underneath it gets good enough to deliver the outcome directly. Once a company crosses that line, it stops competing on features and starts competing on results, which is harder to copy and much harder for a client to walk away from. It also reopens markets that were priced out of reach for smaller buyers. A solicitor on retainer, a broker relationship, an M&A advisor: these used to cost enough that most SMBs went without. A service built the way Lawhive, Corgi and Eilla are built doesn't have to charge for a person's time, so the price drops into range.
The firms with the most to lose are worst placed to respond. A legal practice or insurance broker that's been running the same way for thirty years isn't rebuilding its business model at the pace a four-person startup can, and it doesn't need to lose the whole market to feel it, just the price-sensitive end it was relying on.
What that means for your own stack
Not everything you pay for is exposed the same way. Watch the tools where the vendor's whole value is a feature you could imagine a service doing for you instead: book-keeping, scheduling, anything where "log in and do it yourself" is really standing in for "we haven't built the service version yet." Infrastructure, systems of record, anything you configure rather than something meant to save you from doing a job: that layer isn't going anywhere.
Most of your stack is fine. This isn't a call to panic and rip things out, and "SaaS is dying" is a bigger claim than three companies support. But for the handful of tools where you're really just licensing a workaround for a job you'd rather someone did for you, the question worth asking is whether the company selling you the software is about to get out-competed by one selling you the result.
Working out which tool in your stack is a feature and which is a workaround? Get in touch before you renew.
Robin Carswell
More on
Call us when your vibe-coded option needs to grow up
At some point, your experiment becomes infrastructure. Here's how to know when you've crossed that line — and what to do about it.
Use vibe coding to escape the SaaS you've been stuck with — part 2
You built it quickly, it worked, and now the business depends on it. Here's how to tell whether your vibe-coded build is worth stabilising or worth replacing — before the next thing breaks.
Use vibe coding to escape the SaaS you've been stuck with
You've got at least one SaaS tool your team complains about and you've been meaning to replace. Vibe coding has made it cheap enough to run a proper experiment — here's how to find out if your worst offender is simpler than it looks.