Team & operations

Weekly KPI digest in Slack

A weekly KPI digest gathers a handful of agreed figures from the tools you already use and posts them to Slack at the same time each week. Each line shows the number, how it changed from the week before and where it came from.

An example of how the studio would build it. No live services are connected.

Start every week looking at the same numbers.

  1. Step 01

    Monday arrives

    At 8am on Monday, in your own time zone, the digest starts for the week that has just closed.

  2. Step 02

    Figures collected

    Pull last week’s visits from Google Analytics 4, payments from Stripe and new deals from HubSpot.

  3. Step 03

    Weeks compared

    Compare each figure with the week before, using the same Monday-to-Sunday window for every source.

  4. Step 04

    Digest written

    Turn the results into five plain lines, each with the figure, the change and a link to its source.

  5. Step 05

    Posted to Slack

    Post the digest to the agreed channel, with a clear note if any source failed to respond.

The result

Everyone starts the week with the same picture.

Every figure uses an agreed definition and the same date range. If a source fails, the digest says so instead of posting a misleading zero.

Before we connect it.

Where this is useful

Useful when the team watches different dashboards, or none, and Monday conversations start from different numbers. A sensible first version reports three to five figures people already care about, such as website visits, revenue and new deals, before adding anything more detailed.

What we need to connect

  1. 01Read-only access to Google Analytics 4, Stripe and HubSpot, or whichever tools hold the figures.
  2. 02An agreed definition, date range and time zone for every figure in the digest.
  3. 03The Slack channel, the posting time and a person to tell when a source fails.

What needs careful handling

Each source counts differently. Analytics data can take a while to finish processing, Stripe separates payments from refunds and payouts, and a CRM total shifts when someone edits a deal. Agree each definition up front, report a missing source as missing rather than zero, and make sure a retry can’t post the digest twice.

Questions people ask.

Can it explain why a number went up or down?

The digest reports what changed, not why. It can add context, such as a campaign or bank holiday that week, when that is recorded somewhere it can read. Interpreting the figures stays with the team.

Will everyone in Slack see our revenue?

Only the figures you choose are posted, and only to the channel you choose. Sensitive figures can go to a private channel or a direct message, and the connection to each tool is read-only.

All automation examples