Postiz Review (2026): The Open-Source Issue Tracker Tells You What No Review Site Will

Here is a strange fact about Postiz: the project has more than 36,000 stars on GitHub, and exactly five reviews on Trustpilot. If you are trying to decide whether to trust it with your publishing, the usual review sites have almost nothing to say.

So we read the place where Postiz users actually complain: its public issue tracker. The upside of open source is that the problems are public too. What you find there is more specific than any star rating, and in two cases, more surprising.

Why review sites can't answer "is Postiz good?"

On September 23, 2026 the Postiz Trustpilot page showed a score of 2.7 from five reviews: four 1-star and one 5-star. The complaints are real ("I've asked to cancel my subscription. It kept running for 3 months" wrote Erico Maz, July 26, 2026; "I paid for the highest plan and my openclaw agent can't post" wrote Rick R., June 20, 2026). But five reviews is an anecdote, not a sample.

Meanwhile the repository (gitroomhq/postiz-app) is licensed AGPL-3.0, shipped version 2.24.0 on September 22, 2026, and had 298 open issues and pull requests on the day we checked. That is not a sign of a broken project; busy open-source projects always carry a long queue. It is a sign that the real record of how Postiz behaves lives on GitHub. For the pricing side of the decision, see our separate Postiz pricing breakdown; this review is about whether the product holds up.

What Postiz gets right

Before the problems, the fair version. Postiz does a lot for a tool you can run yourself for free:

• Breadth. Its FAQ lists X, LinkedIn, Instagram, Facebook, TikTok, YouTube, Threads, Pinterest, Reddit, Discord, Bluesky, Mastodon "and more".
• Agent-friendly surfaces. A hosted MCP server, a CLI and an installable skill, so Claude, ChatGPT or Cursor can draft and schedule posts, with a human approving the plan.
• Two ways to own it. A hosted service with a 7-day trial, or self-hosting the open-source version on your own server.
• Honest small print. The pricing page says AI video clipping minutes "are given back" if a clip fails. Few vendors write that down.

If what you value is control over your data and code, those points carry real weight. None of the issues below cancel them out.

Two open issues every Postiz reviewer should mention

1. Cloud TikTok posts can fail at publish time, and it is not your fault. Issue 1854, opened August 7, 2026 and still open when we checked, describes a Postiz Cloud customer whose TikTok post was accepted and scheduled normally, then flipped to ERROR at the scheduled time. The quoted notification: "Daily active user quota reached, please try again later". The reporter explains why this matters: "this is an app-level quota associated with the Postiz Cloud TikTok developer application, an individual Postiz customer cannot resolve it from their connected TikTok account." The same video published fine to Instagram and YouTube in the same campaign.

Read that twice. On a hosted scheduler, your TikTok posts go out through the vendor's TikTok developer app. Its daily quota is shared by every customer. You paid for your plan, but your publishing slot sits in the same pool as everyone else's.

When you buy a hosted scheduler, you are also buying a seat in someone else's API quota. No pricing page lists that seat.

2. The analytics trend badge is a constant. Issue 1911, opened August 17, 2026 and still open, reports that the analytics dashboard shows "a ▲ 5.0% trend indicator on every card for Facebook, Instagram, Threads and LinkedIn Page channels. The value is not computed — it is a literal in the provider source." We checked the main branch ourselves on September 23: the value percentageChange: 5 is still present in the Facebook, Instagram, Threads and LinkedIn Page provider files (the line numbers have moved since the report). Other providers use 0, which hides the badge.

The underlying numbers on those cards may be fine; it is the green arrow that means nothing. If you send client reports from Postiz, crop that badge out until the issue closes.

The older pattern: scheduled posts that fail quietly

Issue 1854 is not isolated. Issue 1724 (July 14, 2026, open) describes Instagram posts created via the public API sitting in QUEUE "indefinitely" on some days, and links similar reports for TikTok and Pinterest. Earlier, issue 1259 (February 27, 2026) reported that on a self-hosted install, "No error log is shown when Postiz misses these scheduled posts"; it was closed in June 2026 with the status "not planned".

To be fair, this is not a Postiz disease. We have collected near-identical complaints about closed-source, paid schedulers: posts that silently do not go out, with no notification. The difference is that with Postiz you can read the exact failure path. That is the strongest argument for it, even in a review that lists its bugs.

If you self-host, one of those failure paths moves to you: the platforms' app approvals. Postiz's own FAQ says Facebook, Instagram, Threads and YouTube approvals "can usually take more than a month". Our piece on Postiz self-hosting and the credential wall goes through that step by step.

Who should use Postiz, and how to test it in a week

Postiz fits you if you value open source, want AI agents to drive your posting, and are comfortable reading a GitHub thread when something breaks. It fits less well if TikTok is your main channel on the hosted plan, or if you need analytics you can hand to a client without editing.

A one-week test plan:

1. Day 1: connect every channel you will really use; schedule one post per channel for the next morning.
2. Day 2: check each post on the platform itself, not in Postiz. Note any ERROR or QUEUE states and the time they appeared.
3. Days 3 to 5: schedule TikTok at different times of day. If the quota issue exists for you, it will show.
4. Day 6: open Analytics. If every Facebook or Instagram card shows ▲ 5.0%, you are looking at issue 1911.
5. Day 7: decide between hosted, self-hosted or something else. Our Postiz open source vs paid guide covers that fork, and Postiz vs Blotato compares it to the most common paid alternative.

A different route avoids the shared-quota problem by not using a vendor developer app at all: tools that act through your own logged-in browser session. NoobClaw works that way, with each account in its own local browser and no password handed over. That trade has costs of its own: the app runs on your machine, and browser-based automation carries platform risk that no vendor can remove, so keep human review and a human pace.

FAQ

Is Postiz good for TikTok?

It supports TikTok, but on Postiz Cloud there is an open report (issue 1854) of scheduled TikTok posts failing at publish time because the shared developer app hit TikTok's daily active-user quota. Test TikTok specifically during the 7-day trial before relying on it.

Is Postiz really free?

The open-source version is free to self-host under AGPL-3.0; the hosted service has a 7-day trial and paid plans. Self-hosting shifts costs to your server and to the platforms' app approvals, which Postiz says can take more than a month for some networks.

Are Postiz analytics accurate?

The underlying numbers come from the platforms, but the green trend percentage on Facebook, Instagram, Threads and LinkedIn Page cards is a hard-coded 5 per issue 1911, still present in the main branch on September 23, 2026. Ignore that badge.