Case study · Fintech, Series C

240 dashboards, one six-figure licence, and no answers

Six-week engagement, fixed scope and fixed price. A six-figure BI licence replaced with a tuned embedded Superset stack.

What came out of it

Money figures were signed off by the customer's own Finance team, and the rest come from the customer's own systems.

$72kSaved in year oneA six-figure BI licence replaced by a tuned embedded Superset stack.
240 to 8DashboardsNot eight of the old boards kept. Eight built around the decisions the business makes.
15s to 3sLoad timeFast enough to open mid-conversation, which is what drives adoption.
6 weeksEngagementFixed scope, fixed price.

What 240 dashboards actually looked like

Each dot is one board on the old licence, and the lit ones are what the estate became.

240 dashboards on the old licence8 boards after the rebuild
The eight are not survivors. They were built new, around the decisions the business makes, and they answer more than the 240 ever did.

How an estate reaches 240 boards

Nobody planned this, it accumulated one reasonable request at a time.

  1. Ask

    A question arrives

    Someone needs a number no existing board answers, so a new dashboard gets built for it.

  2. Keep

    Nothing is retired

    The old board stays where it is, because retiring it is nobody's job.

  3. Grow

    The licence grows with it

    The estate grows by accretion until 240 boards sit on a six-figure licence.

  4. Fail

    The board settles nothing

    Opening one rarely ends the argument, so people go back to asking an analyst.

  5. End

    Back to Excel

    The analyst answers in a spreadsheet, and the exec still cannot say why a number moved.

How a board earned its place in the eight

Week 1 produced a test, and every board on the old licence was put through it.

  • It carries a decision

    Each of the eight was shaped backwards from a decision the business actually makes.

  • It has a named owner

    Somebody is accountable for the board and for the decision behind it.

  • It is not a duplicate

    Boards that repeated another with a different filter were merged rather than rebuilt.

  • It is still opened

    Boards nobody had opened in months were retired, not migrated.

Usage was never the real problem. The problem was that opening a board rarely ended the argument.

Six weeks, week by week

Fixed scope and fixed price, including the part that is unglamorous.

  1. Week 1, Audit

    We traced which boards carried a decision, which were duplicates with a different filter, and which had not been opened in months.

  2. Weeks 2 to 4, Rebuild around decisions

    Eight new boards on an embedded Superset stack, tuned properly rather than installed and left.

  3. Weeks 5 to 6, Cut over

    Migration off the old licence, with load time brought from 15s to 3s.

Speed is an adoption feature

Under a few seconds people check a number mid-conversation, above ten they stop bothering.

Before15s
After3s
Median dashboard load in seconds, same queries and same data volumes.

What the six weeks replaced

Left is what the business was paying for, right is what it works from now.

  1. 240 dashboards8 decision boards
  2. A six-figure BI licenceEmbedded Superset, tuned
  3. 15-second loads3-second loads
  4. Ask an analyst, then waitAnswer inside the meeting
$72k saved in year one, on figures signed off by the customer's own Finance team.

Before you ask

Why is the customer not named?

Anonymised where we have to be, specific everywhere we can be. Reference calls are possible after the audit, subject to the customer agreeing.

What did this cost?

Fixed scope, fixed fee, agreed before Week 1. We give you the number on the audit call once we know the shape of the work, rather than publishing a band that would not apply to you.

Is our situation close enough to this one?

That is exactly what the Week 1 audit answers. Two calls, fixed fee, and you keep a written one-pager whether or not we work together.

Is your shape close to this one?

Two calls, a fixed fee, and a written one-pager you keep either way. We will tell you on the call if we are not the right fit.