AI model spend is becoming very concentrated.
Top 1% of customers account for 46% of all Anthropic spend…
You see a similar concentration at the individual company level with users. And those are the ones we trying to solve for by leveraging more cheaper open-weight models.
I get a ton of requests from “legacy SaaS” for paid product research sessions….
Offering $300+ for 30 mins
Have never taken one, but I know they just want to hear how the new companies are using AI. They were so far behind…
Legacy SaaS is in serious trouble.
By the time your product leaders are offering to pay churned customers to see a demo of the platform winning all your business, the game is already over.
Have talked to several software companies that are adding “AI-enabled managed services” to their offering.
>5-10x their TAM
>Makes them stickier by providing the services
>Incentive to automate more with AI
But they are quickly learning that selling software is so much easier…
When ICONIQ says top quartile you should read that as top decile (or probably higher).
Top VC firm that gets into the best companies and then segments only for AI companies and then the top quartile of those
Just in case you want to feel a tiny bit better about yourself...
Many VCs push Big 4 auditors onto earlier stage companies. Bad advice for most companies <$100M of ARR.
Waste of limited money and time...
1. Series A/B docs default to requiring an audit. Tell the board you want to push it out 1-2 years.
2. You do NOT need Big 4 for your first several audits. They are more expensive and take a lot more of your time. Switch when you have $300M+ in revenue and a path to IPO (hint: most won't ever need one)
3. Already stuck with Big 4 auditor? If an IPO now seems very unlikely or you are 5+ years away, then it’s OK to switch firms to a 2nd or 3rd tier firm. You can switch back to Big 4 later if needed
VCs and boards need to accelerate the death of more software companies.
Founders feel a moral obligation to keep going, but everyone kinda knows it’s not going anywhere…
Free up investor capital and founder time to build something else for the AI world.
A lot of financial metrics only start to matter when you shift from being valued on a multiple of ARR (and terminal value) to FCF.
And if that shift comes too soon then you become irrelevant anyways.
Pay attention to them, but…maybe not too much