What bootstrapped founders actually said about AI, pricing and getting acquired
Nick Francis
20 claims that held, each with the quote it rests on and the date it was true.
Seen in the corpus 1 times, 2026-06-30 to 2026-06-30. Claims that did not hold are not listed here: they were written by research agents rather than by Nick Francis, so they belong with the other refusals, grouped by how they failed.
runs
HeldNick Francis no longer runs Help Scout day to day.
And just to cut to the end of where Help Scout is today, you stepped down as CEO about eight months ago, I guess late 2025, and became chairman of the board.
I’ve started to work with some founders. I’ve started to work with other CEOs. I’ve built a CEO group that I’m really excited about. So I spend maybe a quarter of my time just trying to help other founders, and I just absolutely love that work.
Yeah, I am working on something.
HeldNick Francis is chairman of Help Scout's board.
And just to cut to the end of where Help Scout is today, you stepped down as CEO about eight months ago, I guess late 2025, and became chairman of the board.
I’ve started to work with some founders. I’ve started to work with other CEOs. I’ve built a CEO group that I’m really excited about. So I spend maybe a quarter of my time just trying to help other founders, and I just absolutely love that work.
Yeah, I am working on something.
HeldNick Francis spends about a quarter of his time coaching founders and CEOs.
And just to cut to the end of where Help Scout is today, you stepped down as CEO about eight months ago, I guess late 2025, and became chairman of the board.
I’ve started to work with some founders. I’ve started to work with other CEOs. I’ve built a CEO group that I’m really excited about. So I spend maybe a quarter of my time just trying to help other founders, and I just absolutely love that work.
Yeah, I am working on something.
HeldNick Francis has built a CEO group.
And just to cut to the end of where Help Scout is today, you stepped down as CEO about eight months ago, I guess late 2025, and became chairman of the board.
I’ve started to work with some founders. I’ve started to work with other CEOs. I’ve built a CEO group that I’m really excited about. So I spend maybe a quarter of my time just trying to help other founders, and I just absolutely love that work.
Yeah, I am working on something.
business model
HeldIn November 2024, Help Scout moved from per-seat pricing to fully usage-based per-contact pricing.
I have a note about a pricing overhaul in November of 2024, you guys moving from seat to per contact.
And so we tested and we iterated on three different variations of pricing and packaging over the course of 12 months. Long story short, it wasn’t there. Even when this new usage-based business model would benefit a customer, even when they would pay less, they just didn’t want to do it.
The problem is that our pricing per contact was actually 30% less variable than per seat.
So we ended up landing on a hybrid, which is kind of where the whole market has now landed, which is seats plus you pay for AI resolutions, which is what we wanted to avoid.
HeldHelp Scout tested three variations of the pricing over 12 months.
I have a note about a pricing overhaul in November of 2024, you guys moving from seat to per contact.
And so we tested and we iterated on three different variations of pricing and packaging over the course of 12 months. Long story short, it wasn’t there. Even when this new usage-based business model would benefit a customer, even when they would pay less, they just didn’t want to do it.
The problem is that our pricing per contact was actually 30% less variable than per seat.
So we ended up landing on a hybrid, which is kind of where the whole market has now landed, which is seats plus you pay for AI resolutions, which is what we wanted to avoid.
HeldHelp Scout reverted from the usage-based per-contact pricing.
I have a note about a pricing overhaul in November of 2024, you guys moving from seat to per contact.
And so we tested and we iterated on three different variations of pricing and packaging over the course of 12 months. Long story short, it wasn’t there. Even when this new usage-based business model would benefit a customer, even when they would pay less, they just didn’t want to do it.
The problem is that our pricing per contact was actually 30% less variable than per seat.
So we ended up landing on a hybrid, which is kind of where the whole market has now landed, which is seats plus you pay for AI resolutions, which is what we wanted to avoid.
HeldCustomers rejected the usage-based per-contact pricing even when it cost them less.
I have a note about a pricing overhaul in November of 2024, you guys moving from seat to per contact.
And so we tested and we iterated on three different variations of pricing and packaging over the course of 12 months. Long story short, it wasn’t there. Even when this new usage-based business model would benefit a customer, even when they would pay less, they just didn’t want to do it.
The problem is that our pricing per contact was actually 30% less variable than per seat.
So we ended up landing on a hybrid, which is kind of where the whole market has now landed, which is seats plus you pay for AI resolutions, which is what we wanted to avoid.
HeldPer-contact billing was 30% less variable than per-seat billing.
I have a note about a pricing overhaul in November of 2024, you guys moving from seat to per contact.
And so we tested and we iterated on three different variations of pricing and packaging over the course of 12 months. Long story short, it wasn’t there. Even when this new usage-based business model would benefit a customer, even when they would pay less, they just didn’t want to do it.
The problem is that our pricing per contact was actually 30% less variable than per seat.
So we ended up landing on a hybrid, which is kind of where the whole market has now landed, which is seats plus you pay for AI resolutions, which is what we wanted to avoid.
HeldHelp Scout landed on the seats-plus-AI-resolutions hybrid they had been trying to avoid.
I have a note about a pricing overhaul in November of 2024, you guys moving from seat to per contact.
And so we tested and we iterated on three different variations of pricing and packaging over the course of 12 months. Long story short, it wasn’t there. Even when this new usage-based business model would benefit a customer, even when they would pay less, they just didn’t want to do it.
The problem is that our pricing per contact was actually 30% less variable than per seat.
So we ended up landing on a hybrid, which is kind of where the whole market has now landed, which is seats plus you pay for AI resolutions, which is what we wanted to avoid.
distribution channel
HeldNick Francis required a phone number at signup.
And so I would literally just walk around with my laptop. I would be that guy that sort of tapped you on the shoulder and was like, hey, what are you guys doing for customer support?
And something else that we did was every time somebody signed up, we would require a phone number and I would call every single signup just to understand — I wasn’t there to sell.
ai adaptation
HeldNick Francis built six things in two weeks.
But basically when we first learned about ChatGPT, which I think was in late 2022, we did a hackathon, much like a lot of other companies, just to explore these tools and see what was possible. And so right away we built a tool to summarize conversations. We built a tool to draft responses, and then we ended up making an acquisition along those lines. But we built six things over the course of two weeks.
And ultimately less than 5% of our customers had enough data for a machine learning model to even be useful to them.
HeldNick Francis made an acquisition in the area of conversation summarisation and draft-response tools.
But basically when we first learned about ChatGPT, which I think was in late 2022, we did a hackathon, much like a lot of other companies, just to explore these tools and see what was possible. And so right away we built a tool to summarize conversations. We built a tool to draft responses, and then we ended up making an acquisition along those lines. But we built six things over the course of two weeks.
And ultimately less than 5% of our customers had enough data for a machine learning model to even be useful to them.
HeldNick Francis repriced the entire company because of AI.
And so very early on, as I mentioned, we were building a lot with AI. We realized, wow, we’re going to be able to create a lot of value for our customers that doesn’t tie back to a seat.
And generally you could sort of see: hey, if you fast-forward this ten years, my sense is that a lot of businesses that are per seat today are not really going to make a lot of sense per seat tomorrow.
So in summary, I would say we were too early.
And so had we been bootstrapped at the time, maybe a $20 million business, we wouldn’t have made that move.
HeldNick Francis says he would not have made the swing had the company been bootstrapped at ~$20M.
And so very early on, as I mentioned, we were building a lot with AI. We realized, wow, we’re going to be able to create a lot of value for our customers that doesn’t tie back to a seat.
And generally you could sort of see: hey, if you fast-forward this ten years, my sense is that a lot of businesses that are per seat today are not really going to make a lot of sense per seat tomorrow.
So in summary, I would say we were too early.
And so had we been bootstrapped at the time, maybe a $20 million business, we wouldn’t have made that move.
acquirer signal
HeldHelp Scout acquired a company.
And so right away we built a tool to summarize conversations. We built a tool to draft responses, and then we ended up making an acquisition along those lines.
HeldHelp Scout's acquisition was for AI conversation-summarisation / response-drafting capability.
And so right away we built a tool to summarize conversations. We built a tool to draft responses, and then we ended up making an acquisition along those lines.
HeldHelp Scout raised $28M total.
And so $18K, and we ended up returning about two and a half million dollars on that investment.
you raised $28 million in funding
But for me personally, looking back now on a 15-year journey, I would’ve done it differently. I wouldn’t have raised money.
I can promise you this, Rob: it’s going to be bootstrapped. No institutional capital.
HeldNick Francis says he would not raise capital again.
And so $18K, and we ended up returning about two and a half million dollars on that investment.
you raised $28 million in funding
But for me personally, looking back now on a 15-year journey, I would’ve done it differently. I wouldn’t have raised money.
I can promise you this, Rob: it’s going to be bootstrapped. No institutional capital.
HeldNick Francis says his next company will be bootstrapped with no institutional capital.
And so $18K, and we ended up returning about two and a half million dollars on that investment.
you raised $28 million in funding
But for me personally, looking back now on a 15-year journey, I would’ve done it differently. I wouldn’t have raised money.
I can promise you this, Rob: it’s going to be bootstrapped. No institutional capital.
Where they were quoted