12 companies, 7 exits, $75M raised, angel on the other side. I write about building, money, travel and whatever I am chewing on that week.
Joined February 2008
- Tweets56K
- Following1.3K
- Followers2.7K
- Likes2K
a lot of early founders set the price low so the first customers say yes. it is difficult to raise it later with the same customers so I think it is worth starting higher than feels comfortable
people can explain a layoff in their next interview. being told you were below the bar is much harder to explain and I'm not sure how hiring managers will read it
Brutal reality is Microsoft have just admitted they don't want 27% of their workforce.
“These 27%, we do not want them. This is a worse scarlet letter than a layoff.
This is 27% of the country they said do not meet our minimum, but we do not want you.
It is so damning that Microsoft does not want 27% of its people. That is the age of AI.” @jasonlk
Love to hear your thoughts @GergelyOrosz @NolanChurch @molly_g @Austen
I think per seat pricing is what dies. companies still pay for software that saves them time but they won't keep paying for 200 seats if a few people and some agents do the work
people take less money for a better title more often than you would think. status matters a lot to most people even if they don't say it
IT teams almost never rip out a system of record, so a company under 3 years old doing full replacements is a big deal. I think the old ITSM vendors should be worried
early customers often use the product for something a bit different from what it was built for. I think it is worth asking them why before building the next feature
$10K a month per dev is close to a second salary. big companies can pay that but I think small teams will have to be careful about who gets the expensive models and for what
Correct, NVIDIA has a $10K/dev/month limit, a dev a trading firm I talked to (who wants to keep it anonymous) said its $25K and mainly traders who get close
Full: newsletter.pragmaticengineer…
a lot of parents hold money back so the kids learn to work for it and I understand that. but if they end up resenting you I'm not sure what it was for
Charlie Munger’s no bullsh*t advice on giving money to your kids:
"You give them the money."
"Why?"
"They'll resent you if you don't. And what's the point of building a relationship with them and making all this money, and then at the end of your life your kids hate you?"
"Yeah, but won't that make them less ambitious?"
"Yes. Some of them have a few hundred million and they're not as ambitious. So what?”
ADT already has millions of customers paying every month, so the AI only has to make the service cheaper to run. I think we will see a lot more buyouts like this
a lot of what gets called ARR now is a paid pilot or one good month times 12 so I think investors should ask how much of it actually renewed
I think a lot of the satisfaction came from being stuck for a while and then working it out. with agents you skip the stuck part so it makes sense the people still inventing things are the ones who still enjoy it
when there is a lot of money the easy answer to every problem is to hire someone, and then nobody on the team really understands the problem
Revolut CEO Nik Storonsky says easy money makes soft founders:
“A lot of founders think money is free, which it almost is in California because you can raise so much money at ridiculous valuations and then just spend.”
“There is a different culture of trying to solve problems with money rather than with your brains.”
cofounders should talk about salary and equity and who has the final say before they start. it is a much easier conversation when there is nothing to split yet
I think it is easier to keep going through a bad year when you like the work itself. the money alone is not enough for most people
consumption revenue can drop fast if customers move to a cheaper model but the compute contracts stay the same. I think that is what public investors will ask about first
I don't think everything matters the same. if you treat every small thing as big you are tired by the time something actually big shows up
saying it early is the one I care about. I can work with a bad quarter but if I hear it from someone else first it is hard to trust the numbers after that
