AppGlance vs PostHog
This is the least even comparison on this site, and we would rather say so at the top than bury it. PostHog is an entire product-analytics platform, it is open source, and its free tier is ten times ours. If you are choosing on breadth or on price, stop reading and go use PostHog. This page is about the narrow case where a smaller tool is the better answer, and about being honest that it is narrow.
PostHog does far more than we do. Product analytics, session replay, feature flags, experiments, surveys, error tracking and a data warehouse, on a free tier of 1,000,000 events a month against our 100,000. Its SDKs are MIT and its server is open source.
AppGlance does one thing: your own apps, live, on your own phone, with a push the moment something happens. It is a 300 KB dependency-free SDK and a dashboard you can read in ten seconds. Pick it when that is the whole job, and pick PostHog when the job is bigger than that.
Who wrote this: we make AppGlance. PostHog is an independent company and we do not speak for it. Everything below was checked on 31 August 2026 against posthog.com/pricing, their self-hosting documentation, and the public posthog-ios source. Products move fast, and PostHog moves faster than most, so verify anything decision-critical with them. If we have got something wrong, tell us and we will fix it.
The free tier, which is not close
We are not going to dress this up. Here is what each free plan gives you every month, as published on 31 August 2026:
Both of us refuse to surprise you with a bill. PostHog stops usage at the free limits until you add a payment method; we block at the plan ceiling and never charge an account that has not chosen a paid plan. Neither product will quietly turn your hobby app into an invoice.
One line of ours does not appear in their table because it does not exist elsewhere: presence pings are never billed. The signal that powers "who is in the app right now" is the chattiest thing an analytics SDK can send, and we decided at the start that charging for it would make the product argue with itself.
A platform against a pulse
The honest way to describe the difference is not feature counting. It is who the tool is shaped around.
PostHog is shaped around a team improving a product. You define a funnel, you run an experiment behind a flag, you watch ten session replays to find out why step three leaks, you ship a survey to the people who dropped out, and you read the whole thing back in a warehouse query. That is a real workflow and PostHog is very good at it.
AppGlance is shaped around one developer with a few apps in the store. You want to know that the release is not on fire, that someone bought something this morning, and whether the people who installed last week came back. You want it to reach your phone without you asking. You do not want to define anything first.
If your question is "why did conversion drop", PostHog answers it and we do not. If your question is "is anything happening right now", we answer it in a glance and PostHog makes you build a dashboard first.
Which one to pick
Pick PostHog if…
- You will use more than 100,000 events a month and do not want to pay for it.
- You want session replay, feature flags, experiments, surveys or error tracking, none of which we have.
- Open source matters, or you want to read the server as well as the client.
- You have a web app as well as apps, and want one tool across all of it.
- Your data must live in the US, or you need a BAA.
- You still support iOS 13 to 15. Their SDK goes back to iOS 13; ours starts at iOS 16.
- You are a team. Almost everything above is a team workflow.
Pick AppGlance if…
- You want to open a page and see who is in your app right now, with no dashboard to build first.
- You want a push on your phone the moment someone buys, wearing the icon of the app it is about.
- You want the smallest honest dependency: 300 KB, no third-party packages, one line to configure.
- You want your App Store privacy answers generated rather than researched.
- You run several small apps and want them on one board without one project each.
- You would rather have one screen you actually read than twelve you meant to.
Side by side
| AppGlance | PostHog | |
|---|---|---|
| Free tier | 100,000 billable events a month, 20 apps, no card. Presence pings and identify never billed. |
1,000,000 events a month, plus 5,000 session replays, 1M feature-flag requests, 1,500 survey responses and 100,000 exceptions. No card. |
| Going over | Blocked, never billed. A paid plan is a choice you make. | Usage stops at the free limits until you add a payment method, so no surprise bill either. |
| Paid model | Flat plans from $4.99 a month. | Pay as you go with no platform fee. A card also unlocks 6 projects instead of 1 and 7-year retention instead of 1 year. |
| Live, right now | The product. Presence over a websocket, a pulse dot per session in flight. | Not the shape of the tool. Recent events and dashboards, not a presence count. |
| Push to your phone | A native iPhone app, free on every plan, per-signal rules, the alert wears the app's own icon. | No first-party mobile app for reading your own analytics. Alerts go to email, Slack and webhooks. |
| Session replay | No, and not planned. | Yes, mobile included, 5,000 a month free. |
| Feature flags, experiments, surveys | No. | Yes, all three, with generous free allowances. |
| Funnels and retention | Yes: any-event funnels, D1/D7/D30 retention, new against returning on any chart. | Yes, and deeper: cohorts, paths, correlation analysis, SQL over the warehouse. |
| Per-user stories | Yes. Any install opens as a day-grouped timeline of sessions and events. | Yes: person profiles with event history, plus replay of the actual session. |
| Open source | SDKs are MIT and public. The server is not. | Client and server both open source. posthog-ios is MIT. |
| Self-hosting | No. Hosted only. | Possible, but in their own words "officially unsupported", with all paid features Cloud-only and no support or CVE notifications. |
| Where the data lives | Ireland (AWS eu-west-1), EU only. | US or EU cloud, your choice. |
| Apple platform floor | iOS 16, macOS 13, tvOS 16, watchOS 9, visionOS 1. | iOS 13, macOS 10.15, tvOS 13, watchOS 6. |
| iOS SDK size | ~300 KB compiled, no third-party dependencies (measured 2026-08-31 at 1.2.4). | Not measured by us. We will not print a number for someone else's SDK that we have not built ourselves, and their SDK does considerably more than ours, so expect more. |
| Scope | App analytics. One thing. | Analytics, replay, flags, experiments, surveys, error tracking, data warehouse, web and mobile and backend. |
| Who it is shaped for | One developer with a handful of apps in the store. | A product team improving one product. |
Where PostHog is the better pick
- The free tier is ten times ours, and that is before you count replay, flags and surveys as things you would otherwise buy separately. For most people reading this, that alone settles it.
- It is open source, server included. Ours is not, and a privacy policy is not the same thing as source you can read.
- It answers "why", not just "what". Session replay and experiments are a different class of question from anything we can show you.
- One tool for web and app. If you have a marketing site, a web app and a mobile app, we cover one of those three.
- US residency, and a BAA if you need one. We are EU only.
- iOS 13 support. Ours starts at iOS 16, which rules us out of some existing apps entirely.
- It will still be there. PostHog is a funded company with a large team. We are an indie studio in Norway, and you should weigh that.
Where we think we are better
- The live glance. "Eleven people in your app right now" is not a dashboard you configure, it is the first thing on the screen. It is what the product is named after.
- The phone in your pocket. A native iPhone app with per-signal push, free on every plan including Free. Reading your own numbers should not require a laptop.
- Weight. 300 KB and no third-party packages. A platform SDK carries the surface area of the platform, and on mobile that is a real cost you pay in every build.
- Nothing to define first. One line and you have installs, sessions, retention, versions and countries. No funnel to configure before the tool says anything.
- The App Store questionnaire, answered. Your privacy label ticks generated from your own toggles, plus a free public tool.
- Several apps, one board. Twenty apps on the free plan, on one screen, rather than a project each.
Questions people ask
Is PostHog's free tier really bigger than AppGlance's?
Yes, by ten times on events: 1,000,000 a month against our 100,000, and that is before session replay, feature flags and surveys, which we do not offer at any price. If your decision is about free volume, PostHog wins it and we are not going to pretend otherwise.
Can PostHog show me who is using my app right now?
Not in the way AppGlance does. PostHog shows recent events and dashboards you build; AppGlance's presence count is a live figure over a websocket, with a pulse dot on each session in flight, and it is the first thing on the screen rather than something you configure. That single difference is most of why AppGlance exists.
Does PostHog have an iPhone app for reading my analytics?
PostHog has no first-party mobile app for reading your own product analytics; its alerts reach email, Slack and webhooks. AppGlance's iPhone app is the product's second half: per-signal push the moment something happens, free on every plan including Free.
Can I self-host PostHog instead of paying?
You can, but read their own wording first: self-hosted deployments are “officially unsupported”, every paid feature is Cloud-only, there is no customer support, and they do not publish CVEs for it. They say plainly that they have “literally never seen this math work out” against their cloud. AppGlance has no self-hosted option at all.
Can I use both PostHog and AppGlance?
Yes, and some people should. They answer different questions and neither blocks the other: PostHog for the funnel work and the experiments, AppGlance for the live pulse and the push on your phone. Both SDKs are small enough to coexist, though you will pay for the events twice in attention if not in money, so be deliberate about which events go where.
The fastest way to compare is to look
The demo is the real dashboard on sample data from four apps. No account, nothing to uninstall.
Also worth reading: all the comparisons, how this compares with Mixpanel, how this compares with Amplitude, how this compares with Aptabase, and what privacy-first analytics actually means.