Skip to content
MH
All projects
AutomationData EngineeringProfessional / Private

Weekly Agent Performance Reporting

Scheduled ETL / Analytics / Slack

A scheduled ETL workflow that pulls paginated activity data, computes weekly metrics, upserts the results to Airtable, and posts a formatted report into Slack.

Year
2026
Role
Generative AI Engineer, Tekvix
Source
Not public
Weekly Agent Performance Reporting — concept diagram

Technologies

  • n8n
  • Domo
  • Airtable
  • Slack API
  • ETL
  • Pagination

The problem

Weekly performance reporting is the kind of task that quietly consumes a person: pull the data, page through it, recompute the same metrics, paste the result somewhere people will read it.

The solution

Put the whole thing on a schedule. The workflow handles pagination properly, computes the metrics the same way every week, stores them where they can be queried later, and delivers the summary where the team already is.

Architecture

How it is put together.

01

Schedule

The workflow runs on a fixed weekly cadence.

02

Paginated pull

Activity data is retrieved across pages until the full period is covered.

03

Metric computation

Weekly metrics are computed consistently from the complete dataset.

04

Airtable upsert

Results are upserted so history accumulates without duplicates.

05

Slack report

A formatted report is posted to Slack for the team.

Key features

What it actually does.

  • Fully scheduled — no manual trigger
  • Correct pagination across the full reporting period
  • Consistent week-over-week metric definitions
  • Airtable upsert preserving history
  • Formatted Slack delivery

Engineering challenges

The parts that were hard.

Pagination is where reports go wrong

A report that silently reads only the first page looks fine and is wrong. Completeness had to be guaranteed, not assumed.

Outcomes

What came out of it.

Qualitative outcomes only — no invented benchmarks or metrics.

01

Weekly reporting runs without anyone remembering to do it

02

Metric history accumulates in a queryable store

03

The team reads the report where they already work

Voice AI Assistant — Real-Time Voice / LLM / Docker

Next project

Voice AI Assistant

Real-time speech-to-text, LLM reasoning and text-to-speech in one deployed loop.

Let’s build

Want something like this for your business?

Tell me the problem and I will come back with an architecture, a scope and a timeline.