---
title: Raising subscription prices (Skylight $39 → $79)
slug: subscription-price-increase
summary: How Skylight doubled its subscription from $39 to $79/year with minimal loss of attach or retention — price tests (incl. hardware discount), choosing below the ARPU-max $99, grandfathering existing subscribers, honouring the old price on request.
domain: mobile-apps
tags: [monetization, pricing, price-increase, grandfathering, experiment]
related: [setting-initial-subscription-price, what-to-gate-behind-the-paywall, tiered-pricing-and-consumables, weekly-vs-yearly-plans-and-trial-placement]
sources: [yt:AilakPuHdf4, yt:gqN6Z5WAeiI]
updated: 2026-10-24
---

# Raising subscription prices

## TL;DR
- Two ways to raise ARPU: more payers or higher price. With >50% of buyers already subscribing, **price was Skylight's biggest lever**.
- Skylight tested several price points (and, for hardware, the bundled device discount); **$99 penciled out slightly better on ARPU, but they shipped $79** to keep goodwill and leave room to go up.
- Result: **$39 → $79/year with minimal loss of attach rate or retention** — called "transformative" for ARR.
- **Grandfather existing subscribers** (Skylight's "Legacy Plus"); don't announce; quietly honour the old price for anyone who'd been on the fence and asks.
- Triangulate quant tests with **qualitative** interviews — they'll tell you when a price tips into "not worth it" territory.
- Consider announcing an increase in advance as a buying trigger (untested idea from David Barnard).

## Details
**Why raise.** Skylight's calendar subscription had launched sharing the same SKU and $39 price as the photo-frame sub (David: $39/yr is a "magic number where people don't really care"). Asking "what would we charge if we were a calendar-only company?" and seeing that more than half of calendar buyers already subscribe, Mark Ungerer concluded price was the lever and they were leaving value on the table.[^1][^2]

**How they tested.** They ran variants on both the subscription price and the hardware discount given with the subscription — for hardware those interact (unit conversion, unit margin, attach). The spreadsheet said $99 would make slightly more; they picked $79 because doubling already felt like a lot, they wanted "good vibes", and "we can always go up."[^3][^4] Michael Segal: interviews showed $99 was getting close to "disgust / not worth it yet"; they'll revisit as the bundle gets richer. The biggest mistake is skipping qualitative research and trusting numbers alone — people say one thing and do another, so do both.[^4][^5]

**Rollout.** No announcement; existing subscribers were grandfathered into a legacy plan. Because buyers research for months, anyone who reached out saying they'd planned to buy at $39 got the old price. They knew squeezing existing customers was worth "a couple million dollars" and chose not to: be nice to existing customers, build for the next 10 million. Michael notes a PE-owned company needing to double EBITDA would decide differently.[^6][^7]

**Outcome.** Their biggest win of the year: $39 → $79 with minimal loss in retention or attach.[^8]

**Advance announcement idea.** David hasn't seen an app pre-announce a price rise, but a home-gym brand announcing tariff-driven increases made him buy early; he suggests someone test it on a large freemium base.[^9]

**Other data point.** Natal raised prices while still growing ~20% in Q1 (David cites the growth; Nancy adds they raised prices over that period).[^10]

## Disagreements / caveats
- Skylight's subscribers bought $300–700 hardware and use it daily, which makes them unusually price-tolerant; app-only products may see more churn from a doubling.
- Grandfathering forgoes revenue; with a smaller or less loyal base, a staged increase for existing users might be justified — not covered by these sources.

## See also
- [[setting-initial-subscription-price]]
- [[people/michael-segal]], [[people/mark-ungerer]]

[^1]: yt:AilakPuHdf4 @ 17:44 — Michael Segal
[^2]: yt:AilakPuHdf4 @ 50:27 — Mark Ungerer
[^3]: yt:AilakPuHdf4 @ 51:17 — Mark Ungerer
[^4]: yt:AilakPuHdf4 @ 52:03 — Mark Ungerer, Michael Segal
[^5]: yt:AilakPuHdf4 @ 52:49 — Michael Segal
[^6]: yt:AilakPuHdf4 @ 53:37 — Michael Segal
[^7]: yt:AilakPuHdf4 @ 54:26 — Michael Segal
[^8]: yt:AilakPuHdf4 @ 55:15 — Michael Segal
[^9]: yt:AilakPuHdf4 @ 54:26 — David Barnard
[^10]: yt:gqN6Z5WAeiI @ 28:07 — David Barnard, Nancy Anderson