AppealFlow

Inspiration

Ban appeals are one of the clearest trust moments in a subreddit. A user is asking the team to reconsider a serious decision. The mod team needs context, consistency, and a clean way to close the loop.

Most communities handle that inside modmail. That works for conversation, but it is weak as a workflow. Appeals get mixed with everything else. Ownership is unclear. Follow-ups depend on memory. When a decision is overturned or upheld, there is rarely a structured record the team can learn from later.

AppealFlow was built around a simple idea: mods do not need another inbox. They need a lightweight appeal desk.

What it does

AppealFlow turns every ban appeal into a case with a state, an owner, a deadline, and a final decision.

A banned user starts with a structured form. Instead of sending a vague message, they answer focused prompts: what happened, which rule they think applies, why the decision should be reconsidered, and what they would do differently. The app checks eligibility, blocks duplicate open appeals, and applies cooldowns after upheld decisions.

On the moderator side, the dashboard shows open appeals, SLA urgency, search, filters, user answers, ban context, and prior signals. The action model is intentionally small: uphold, reduce, overturn, or escalate. Every outcome requires a moderator note, so the user receives a clear answer instead of silence or a one-line close.

How we built it

The app is built on Devvit Web with a Hono server layer and Redis-backed storage. Appeal records are scoped per subreddit, and the workflow is enforced through explicit state transitions rather than loose UI state.

There is also background SLA logic. Open appeals can trigger reminders, become stale, and escalate to the team when deadlines are missed. The analytics view gives teams a picture of response time, outcome breakdown, SLA compliance, rule distribution, and reviewer handling.

What was hard

The hardest part was keeping the product useful without turning it into over-automation. Ban appeals are sensitive; software should not pretend to be the judge. AppealFlow automates the structure around the decision, not the decision itself. Mods stay responsible for the outcome, while the app makes sure the process is visible, consistent, and finished.

The other challenge was keeping the interface practical. A mod dashboard can easily become a generic admin panel. The goal here was a desk that explains what needs attention, why it matters, and what action is safe to take next.

Why it matters

For busy communities, AppealFlow can save real moderator time by removing repeated context digging and response drafting. More importantly, it improves trust. Users may still receive an upheld decision, but they know the appeal was received, reviewed, and answered.

It also gives teams a feedback loop they usually do not have. If one rule creates most appeals, or a meaningful percentage of bans are reduced or overturned, that is a signal. The community can use that signal to clarify rules, improve enforcement consistency, and reduce future conflict.

Built With

• devvit
• devvit-web
• hono
• javascript
• node.js
• reddit-developer-platform
• redis