What happens after a run goes wrong, and what happens when you point a flow at eleven thousand old rows. Which errors get retried and which never will, how a queue looks different from a failure, and the backfill settings worth checking twice.
beginner
6:24
Send a Slack alert when a payment fails
A declined charge is worth four minutes of somebody's attention and usually gets none. Build the alert that puts it in the channel your team already reads, write the message so it can be understood in one glance, and keep it out of the wrong channel.
Video placeholder
A single Slack message card in the centre of the frame, four short lines: customer email, amount, decline reason, link. A small red dot beside it. Muted background showing a blurred flow canvas.
Chapters
From the transcript
A failed payment is worth about four minutes of somebody's attention and usually gets none, because nobody sits in the payments dashboard all day. So we are going to put it in the channel where people already look. Trigger is Stripe, payment failed. Stripe pushes this one to us, so there is no polling and no waiting: a decline in the morning is a message a few seconds later. Action is Slack, post a message, and here is the part I care about. Do not dump the raw object into the channel. Write a sentence. Customer email, then the amount with the currency spelled out, then the decline reason in plain words, then a link to the charge. Four lines, one glance, somebody knows whether to pick up the phone. I add a filter for amounts over fifty dollars so the channel stays readable, and I send it to a private channel, because this message names customers. Test it with a card number that always declines.
More videos
One trigger, three actions, and the order decides whether the result is usable. A first payment creates the contact, moves the deal and opens the onboarding card, each step handing its output to the next, with a clean stop when the first one fails.
The flow is rarely broken. One record is failing over and over, and the history names it in about thirty seconds. Sort by failed, read the two columns beside the error rather than the error itself, fix the mapping, replay the backlog in one pass.