Twilio and https://t.co/1x9tUdcebi both grew 22%. One is up 74%, the other is down 36%.
That spread tells me revenue growth is now a weak proxy for how AI changes a SaaS business.
There are 3 operating questions to look at:
1/ Does AI expand the billing unit? More agents create more queries, logs and security exposure, so Snowflake, Datadog and CrowdStrike earn as usage rises.
2/ Who pays for inference? Seat-priced products can face fewer paid users and higher delivery costs at the same time.
3/ Is the new revenue durable? https://t.co/1x9tUdcebi moving AI to consumption pricing did not earn a re-rating because the market wants proof in forward revenue.
The strongest model aligns price with customer value, converts AI adoption into committed revenue and protects gross margin.
Eight months ago the market decided that AI was going to eat B2B software. A rough, rough start to the year. Roughly $2 trillion of market cap came out of the group.
Then it came back. Just not ... evenly. https://t.co/bL6CopWtiC
didn’t realise that when descriptions are far too long and when you add too many skills, codex starts shortening their descriptions to fit 🤯🤯
so your accuracy eventually reduces despite using relevant skills.