Skip to main content

Client Overview

Pick a client to see their full pipeline health on one page: a KPI summary, the latest ThriveStack job run with a 30-day error history, warehouse-vs-platform checksums, and the daily data-refresh SLA.

How to Use​

  1. Select a client from the dropdown (type to search). Your selection is saved in the page URL, so you can bookmark or share a link to a specific client.
  2. Read the header pill for the at-a-glance verdict:
    • 🟢 All healthy: no problems detected across the sections that have data.
    • 🟡 Minor issues: something is off (jobs still running or unsynced, metrics slightly off-range, SLA attainment below goal) but nothing is hard-failing.
    • 🔴 Needs attention: at least one job failed, a checksum is off-range, or a source missed its refresh check.
  3. Scan the KPI row: jobs failing, checksums in range, refresh SLA, and the time the client's data was ready (or Delayed if a job failed).
  4. ThriveStack jobs: expand a data source to see each pipeline's schedule, last run, duration, error days, and status. Toggle Grouped / Timeline to switch between the table and a run-time gantt view. The sidebar shows which of the last 30 days had job errors and when the data was ready.
  5. Checksums: expand a source to see each metric's 30-day status strip; click a metric to open its ratio chart with the expected range band. Statuses use the same in-range / off-range scale as the Metric Checksums page.
  6. Data refresh SLA: each source shows a 30-day strip graded by that day's pass-rate (accounts met ÷ accounts with data), so a partly-fresh day reads as an intermediate color rather than a flat fail; the source % is the average of those daily pass-rates. Expand to see individual accounts (still shown as plain pass/fail). The attainment card tracks the daily percentage of accounts that were fresh, against the 98% goal.
  7. Drill in via Open full dashboard → in any section header. It opens that feed's dedicated dashboard already filtered to the selected client.
note

The 30-day window is anchored to the latest data day from the pipeline (which runs on Pacific time), so "Latest day" can be yesterday if today's run hasn't landed yet. The As of timestamp in the header shows when the feeds were last exported. A grey day in any strip means no data was recorded that day, not a failure. If a feed is temporarily unavailable, only its section shows an error and the header notes it.