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.
intermediate
8:05
Filter a noisy trigger before it floods Slack
Two hundred and forty messages in a day means nobody read any of them. Three places to fix that, ranked by how much they save: a narrower trigger, a filter step that runs before the action, and hourly batching with a count on top.
Video placeholder
A channel column on the left crammed with near identical notification rows, fading out. On the right the same column with four spaced messages and a small digest header. A funnel glyph between the two.
Chapters
From the transcript
This channel got two hundred and forty messages yesterday, and I promise you nobody read them. That is a filter problem, not a Slack problem. There are three places to solve it and they are not equal. The first is the trigger itself. If you are watching a HubSpot property, watch one property, not the whole record. The second is a filter step, which is what most people reach for. Amount over a threshold, stage equals one specific stage, owner is not empty. Filters run before the action, so a record that gets filtered out costs you nothing against your monthly tasks. The third is the one people forget: batching. Instead of a message per record, collect them and post once an hour with a count at the top. I will build all three on the same flow and show you the numbers afterwards. We go from two hundred and forty down to eleven, and the eleven are the ones somebody actually acts on.
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.