Michael Segal
Where to hear and read about Michael Segal1 episode or test
Michael Segal is co-founder and CEO of Skylight, a bootstrapped, profitable maker of family calendar and photo frame devices with a subscription, and a former Bessemer VC. The thing most worth learning from him is to price and gate by how customers feel, not only by what the test spreadsheet says, and to choose growth over profit on purpose.
We look in the whites of customers' eyes and we try to figure out what feels relatively reasonable.
How Skylight Doubled Subscription Prices to $79 – Michael Segal & Mark Ungerer
Michael Segal: the story and the lessons
1How Michael Segal built Skylight around a subscription
Segal, a former Bessemer VC, wanted a subscription from the start. He held that consumer hardware without a subscription builds no enterprise value, with rare exceptions. Skylight launched in 2015 and added the subscription in 2018.
The founders decided it was time when the company had a couple million in revenue and two people. Their net promoter score was 'off the charts', so they expected goodwill to carry paid features.
Segal nearly killed calendar and starved it of resources. Calendar shipped in 2018, but a marketing push in 2021 and 2022 failed because the product was not ready. It took off in fall 2022.
The team had put resources into the frame business while it worked. Calendar was hard to build, and the frame sold with two ads a year while calendar needs a long emotional sale.
Segal says marketing is wasted below roughly 40 NPS. Below about 40 NPS, spend gives a false negative because you cannot tell whether the product or the marketing is failing.
"if your product isn't like a solid 40 NPS, I think marketing effort and dollars are kind of a waste of time"
Segal sees a market of 10 to 20 million households. Skylight is in about 2 million households, in a US with 100 million households, 30 million of them with kids.
He hopes Skylight is a tenth of the way there. Reaching 20 million households would likely need a lower hardware price, which he says Skylight cannot cut yet for 'complex reasons'.
2How Segal funds and runs a bootstrapped hardware company
Segal has paid back every order on day zero. Every order in Skylight's history has paid back on the day it was placed, because the device is sold up front. That removed the need for outside funding.
He says bootstrapping usually fails on three costs: engineers, working capital for inventory, and marketing. His advice to subscription apps is to turn a $5 monthly plan into a $60 to $70 upfront annual payment.
Segal funded inventory with lenders instead of selling equity. Cost of goods is about 50%, and stock must be bought weeks ahead. Some loans cost 15 to 20% annualized.
Early on he kept a weekly cash flow model, since cash could be near zero one day and millions the next on Black Friday. "I would trade off gross margins all day long instead of selling a piece of the company."
Segal earns on both hardware and software instead of running hardware at zero. Profit is about half from each, even though software is only 10 to 20% of revenue today.
Thin hardware margin gives room to hold prices through tariffs and rising memory costs. It also leaves room for a price cut later to reach more households.
Segal caps acquisition spend because returns fall off a cliff. Going from $130 to $180 or $200 per customer buys almost no one more.
He compares raising spend that way to lighting $50 million on fire.
Segal decides growth versus profit before choosing a payment path. Skylight keeps Apple in-app purchase because easy subscribing beats the fee it saves. Web billing saves closer to 20% than 30%.
He picks growth because Skylight is not preparing a sale and has no private equity owner chasing EBITDA. A company in that position might choose differently.
3How Segal prices and gates the Skylight subscription
Segal doubled the plus plan from $39 to $79 a year. The increase cost minimal retention and attach and was 'transformative' to the ARR curve. More than half of Calendar buyers already subscribed, so price was the lever left.
No retention, attach or revenue figures are given, so 'minimal' is the only measure of the loss.
Segal chose $79 over a $99 price that tested slightly better. Tests showed $99 would earn slightly more, but customer interviews put it near 'disgust' or 'not worth it yet' territory.
He says: 'we want good vibes to continue. We can always go up.' Skylight expects to revisit $99 as it adds more value to the subscription.
Segal tested subscription price and hardware discount together. The two trade off against each other in hardware conversion, margin per unit and subscription attach, so Skylight varied both and modeled the combined result.
The biggest mistake, he says, is skipping customer calls and trusting the numbers alone, because people say one thing and do another.
Segal kept existing customers on the old price. Existing subscribers moved to a 'legacy plus' plan. The increase was not announced, and anyone who had waited months and wrote in got $39 effectively indefinitely.
Squeezing legacy customers was worth 'a couple million dollars', but he chose to build for the next 10 million customers.
Segal draws the paywall line at what the customer came to buy. On the frame, sending photos to grandma is free and video is paid. On the calendar, events, tasks and chores are free, while recipes, meals and an AI engine are paid.
The test is whether the buyer came, spent the money and got what they wanted. Customers who feel 'tricked and duped' are the risk.
Segal launches new features behind the paywall, never moves them in later. Moving a feature out of the paywall is easy. Moving one in is 'very painful'. Meal planning launched paid, and many such launches go untested.
The photo screensaver was paywalled with no rigor, and eight years on more customers say it should be free. The team is debating making it fully or partly free.
Segal treats paywall anger as a cost even when numbers rise. Some paywalled items lifted numbers but drew begrudging or angry reactions, so Skylight reversed course. The team watches reviews and interviews, not only A/B results.
He calls it 'not a get-rich-quick scheme' and says disgust is the emotion to avoid when building a multi-billion dollar brand.
4Segal's growth playbook and what he got wrong
Segal sells on his own site, Amazon and retail, and accepts the tradeoffs. On its own site Skylight turns the subscription on automatically and discounts the hardware, so attach is much higher than on Amazon or in stores.
Costco is a 'stamp of quality', and retailers usually take only one player, which makes it a moat. Retail buyers set up the calendar and never explore the rest, which his product chief is working on this year.
Segal wants new users to set up more features in the first days. Within weeks or even days of setup, families settle into habits, and persuading them to try new things drops to near zero.
Skylight is exploring conversational, AI-driven onboarding that could build a chore chart in the first 10 minutes by asking the kids' ages.
Segal tracks the share of customers who call it life-changing. About 20% of survey replies say life-changing, and he thinks those people bring in dozens of customers each. The goal for the chief product officer is 50%.
Growth runs on four levers: influencers, paid, retail and word of mouth. Word of mouth is the one he most wants to grow.
Segal's free trial test failed because growth was starved of resources. Skylight tried one version taking a credit card and one without. Trial takers were about the same people who would have subscribed anyway.
Growth work had about 5% of the roadmap and no growth PM. He says the first test fails and about eight more are needed, so the trial was a 'Hail Mary'. The company is hiring its first subscription growth PM.
Segal cut checkout friction and lifted conversion over 100%. Sending non-subscribers to Apple in-app checkout instead of a web checkout through a QR code lifted conversion over 100% for that flow.
Fewer people use the QR code, so the dollar gain is smaller than the price change. He still counts it, with the $79 plan, among the year's biggest wins.
Segal says neither hardware nor software is a moat. Competitors cloned Skylight's UI and sold it through Chinese OEMs on Amazon.
For now the edge is the best product and speed. Lasting defense needs network effects or partnerships that fit the product, and his aim is to be 'the Kleenex of this product'.
How Skylight Doubled Subscription Prices to $79 – Michael Segal & Mark Ungerer
Skylight, a bootstrapped family calendar and photo frame maker, explains how it prices and gates a subscription on top of a $300 to $700 device. It doubled the plus plan from $39 to $79 a year with minimal loss of retention or attach, and chose $79 over a revenue-maximizing $99 after talking to customers.
Segal, a former Bessemer VC, wanted a subscription from the start. He held that consumer hardware without a subscription builds no enterprise value, with rare exceptions. Skylight launched in 2015 and added the subscription in 2018.
The founders decided it was time when the company had a couple million in revenue and two people. Their net promoter score was 'off the charts', so they expected goodwill to carry paid features.
Segal nearly killed calendar and starved it of resources. Calendar shipped in 2018, but a marketing push in 2021 and 2022 failed because the product was not ready. It took off in fall 2022.
The team had put resources into the frame business while it worked. Calendar was hard to build, and the frame sold with two ads a year while calendar needs a long emotional sale.
Segal says marketing is wasted below roughly 40 NPS. Below about 40 NPS, spend gives a false negative because you cannot tell whether the product or the marketing is failing.
"if your product isn't like a solid 40 NPS, I think marketing effort and dollars are kind of a waste of time"
Segal sees a market of 10 to 20 million households. Skylight is in about 2 million households, in a US with 100 million households, 30 million of them with kids.
He hopes Skylight is a tenth of the way there. Reaching 20 million households would likely need a lower hardware price, which he says Skylight cannot cut yet for 'complex reasons'.
Segal has paid back every order on day zero. Every order in Skylight's history has paid back on the day it was placed, because the device is sold up front. That removed the need for outside funding.
He says bootstrapping usually fails on three costs: engineers, working capital for inventory, and marketing. His advice to subscription apps is to turn a $5 monthly plan into a $60 to $70 upfront annual payment.
Segal funded inventory with lenders instead of selling equity. Cost of goods is about 50%, and stock must be bought weeks ahead. Some loans cost 15 to 20% annualized.
Early on he kept a weekly cash flow model, since cash could be near zero one day and millions the next on Black Friday. "I would trade off gross margins all day long instead of selling a piece of the company."
Segal earns on both hardware and software instead of running hardware at zero. Profit is about half from each, even though software is only 10 to 20% of revenue today.
Thin hardware margin gives room to hold prices through tariffs and rising memory costs. It also leaves room for a price cut later to reach more households.
Segal caps acquisition spend because returns fall off a cliff. Going from $130 to $180 or $200 per customer buys almost no one more.
He compares raising spend that way to lighting $50 million on fire.
Segal decides growth versus profit before choosing a payment path. Skylight keeps Apple in-app purchase because easy subscribing beats the fee it saves. Web billing saves closer to 20% than 30%.
He picks growth because Skylight is not preparing a sale and has no private equity owner chasing EBITDA. A company in that position might choose differently.
Segal doubled the plus plan from $39 to $79 a year. The increase cost minimal retention and attach and was 'transformative' to the ARR curve. More than half of Calendar buyers already subscribed, so price was the lever left.
No retention, attach or revenue figures are given, so 'minimal' is the only measure of the loss.
Segal chose $79 over a $99 price that tested slightly better. Tests showed $99 would earn slightly more, but customer interviews put it near 'disgust' or 'not worth it yet' territory.
He says: 'we want good vibes to continue. We can always go up.' Skylight expects to revisit $99 as it adds more value to the subscription.
Segal tested subscription price and hardware discount together. The two trade off against each other in hardware conversion, margin per unit and subscription attach, so Skylight varied both and modeled the combined result.
The biggest mistake, he says, is skipping customer calls and trusting the numbers alone, because people say one thing and do another.
Segal kept existing customers on the old price. Existing subscribers moved to a 'legacy plus' plan. The increase was not announced, and anyone who had waited months and wrote in got $39 effectively indefinitely.
Squeezing legacy customers was worth 'a couple million dollars', but he chose to build for the next 10 million customers.
Segal draws the paywall line at what the customer came to buy. On the frame, sending photos to grandma is free and video is paid. On the calendar, events, tasks and chores are free, while recipes, meals and an AI engine are paid.
The test is whether the buyer came, spent the money and got what they wanted. Customers who feel 'tricked and duped' are the risk.
Segal launches new features behind the paywall, never moves them in later. Moving a feature out of the paywall is easy. Moving one in is 'very painful'. Meal planning launched paid, and many such launches go untested.
The photo screensaver was paywalled with no rigor, and eight years on more customers say it should be free. The team is debating making it fully or partly free.
Segal treats paywall anger as a cost even when numbers rise. Some paywalled items lifted numbers but drew begrudging or angry reactions, so Skylight reversed course. The team watches reviews and interviews, not only A/B results.
He calls it 'not a get-rich-quick scheme' and says disgust is the emotion to avoid when building a multi-billion dollar brand.
Segal sells on his own site, Amazon and retail, and accepts the tradeoffs. On its own site Skylight turns the subscription on automatically and discounts the hardware, so attach is much higher than on Amazon or in stores.
Costco is a 'stamp of quality', and retailers usually take only one player, which makes it a moat. Retail buyers set up the calendar and never explore the rest, which his product chief is working on this year.
Segal wants new users to set up more features in the first days. Within weeks or even days of setup, families settle into habits, and persuading them to try new things drops to near zero.
Skylight is exploring conversational, AI-driven onboarding that could build a chore chart in the first 10 minutes by asking the kids' ages.
Segal tracks the share of customers who call it life-changing. About 20% of survey replies say life-changing, and he thinks those people bring in dozens of customers each. The goal for the chief product officer is 50%.
Growth runs on four levers: influencers, paid, retail and word of mouth. Word of mouth is the one he most wants to grow.
Segal's free trial test failed because growth was starved of resources. Skylight tried one version taking a credit card and one without. Trial takers were about the same people who would have subscribed anyway.
Growth work had about 5% of the roadmap and no growth PM. He says the first test fails and about eight more are needed, so the trial was a 'Hail Mary'. The company is hiring its first subscription growth PM.
Segal cut checkout friction and lifted conversion over 100%. Sending non-subscribers to Apple in-app checkout instead of a web checkout through a QR code lifted conversion over 100% for that flow.
Fewer people use the QR code, so the dollar gain is smaller than the price change. He still counts it, with the $79 plan, among the year's biggest wins.
Segal says neither hardware nor software is a moat. Competitors cloned Skylight's UI and sold it through Chinese OEMs on Amazon.
For now the edge is the best product and speed. Lasting defense needs network effects or partnerships that fit the product, and his aim is to be 'the Kleenex of this product'.