Walkthrough Full case study

Reporting 2.0 · Sendible · 2026

The report that couldn't explain itself

A two-minute walkthrough of the rebuild, one beat at a time.

arrow keys, tap, or swipe to move

It started with a note. Four years later I was asked to answer it.

I'd prefer to stay with Sendible, the price is perfect and the interface is easier to manage. But I haven't felt much confidence in the reports lately.

An agency customer, 2022

Reporting had grown into two systems that didn't know about each other.

The legacy Sendible Facebook report. 1 Seven networks, seven separate reports. No way to combine them.

You could rearrange what you were given. You couldn't change it.

The legacy report, its module drag handle marked. 1 Seven networks, seven separate reports. No way to combine them. 2 Reorder the modules, yes. Add, remove or redraw one, no.

And nothing on the screen belonged to the report.

The legacy report, its five clashing palettes marked. 1 Seven networks, seven separate reports. No way to combine them. 2 Reorder the modules, yes. Add, remove or redraw one, no. 3 Five palettes on one screen, none of them the report's own.

The seams showed up as small bugs customers read as lies.

A report filtered to February logging only four Facebook posts, when there had been many more.

So before proposing anything, I sat with the three teams who absorb it, then read the tickets myself.

  • Customer success
  • Onboarding
  • Support
Accounting firm · NPS 0

Reporting just isn't what is needed to manage anything at scale.

Agency, legal sector

Despite the date filter set to Feb 1st to 29th, the widget is still showing 4th Jan to 1st Feb.

Marketing agency

September saw higher engagement than August, but the percentage shows a decrease.

Nonprofit

I need January to December. Engagement doesn't give me all the information.

Media monitoring, APAC

Why are YouTube and TikTok not included in the engagement report? It only shows Facebook, Instagram and LinkedIn.

Influencer agency

Modules always get added to the top, and dragging them to the bottom is tedious.

The central decision: stop treating quick and custom reports as different things.

The rebuilt report canvas.

The real work was designing for the moment a report has nothing to say.

Two rules came out of it, and they look like opposites. First: never hide capability, name the barrier.

Second: show only what the decision needs. The template cards briefly carried a module count. I liked it. I took it out.

The reports list shipped smaller than it started, by reusing a list component the product already had.

Telling someone whether they were viewing or editing took nine iterations. The fix wasn't a louder notice, it was almost none.

And view mode was never a separate screen. It's the same report with about a dozen things switched off.

The first release shipped two networks, and looked finished doing it.

The dashboard scoped to two networks: Facebook and TikTok templates plus start from scratch, above two saved reports.

No launch metric. Instead, a precise picture of what it was costing.

Five cards on an orange field: Retention, Expansion, Plan choice, Support load and Pricing, the five places weak reporting cost money.

The state model is what I'd carry forward. What a report says when it has nothing to say won't change.

1 / 16