Which single cut earns its place every week?
- If only one breakdown ran every Monday, which one would change what somebody does?
- Could the others rotate — channel one week, programs the next?
- Is a breakdown that never surprises anyone worth a slide?
DCHS Insights CX Human-Centered Design @ DCHS PROTOTYPE no real data DCHS Insights CX How We Spotlight — How the week breaks down
The same week, cut three ways
Volume split by channel, by program, and by movement in rank — the closest neighbour to the headline numbers, and the group where it is easiest to publish three slides that say one thing.
What this is
Each option divides the same total in a different way. Limit the group to breakdowns that answer a distinct operational question.
The prototype currently carries five of these. That is almost certainly too many, and choosing between them is a genuine editorial decision rather than a technical one.
What could go in it
Options to react to, not a recommendation. Strike, keep, and add — reacting to a concrete list is far more productive than starting from a blank page.
-
By channel
Phone, in person, online — and within phone, what happened to the call. This is the cut that tells staff where the pressure actually landed, and the one most likely to change how a team schedules its week.
-
By program, sized to scale
A treemap where the area does the talking. It is the only slide that shows the whole department at once, and it makes relative size obvious in a way a ranked list never does.
-
By movement, not by size
Who overtook whom over five weeks. More interesting than a static ranking and much easier to over-read: a program can climb three places on twelve extra visits.
-
By where people came from
Referral source, first visit versus returning, and walk-in versus appointment could support operational decisions, but the required data may not be readily available.
-
Digital next to the doors
Portal, online booking and the eligibility checker shown in the same breath as foot traffic, so the web is not treated as somebody else's report.
Questions to settle
These are the decisions that have to be made before anything here can run every Monday. Most of them are editorial rather than technical.
Program, center, or service?
- Staff recognise one of these; the data is probably organised around another.
- Where the two disagree, which one do we publish under?
- A client does not experience a program — they experience a desk. Does that matter here?
Rank or volume?
- Rank hides scale and invites over-reading. Volume is honest and less interesting.
- Do we publish the underlying counts alongside the ranking?
- Is there a minimum change below which we simply do not report movement?
Do small programs get named?
- A treemap with a small box labelled “Other” hides somebody's whole week of work.
- Is there a floor below which we aggregate — and does being aggregated read as being unimportant?
- The alternative is a chart with fourteen unreadable boxes. Both options cost something.
Does the split include the people we lost?
- Abandoned calls, walk-outs, applications started online and never finished.
- A denominator of only the people we served is a flattering denominator.
- Is there an appetite to publish the ones that got away, and what happens when that number is bad?
What it would take
What each slide needs, and the question that has to be answered before we know whether we can produce it. The systems are named loosely on purpose — which system holds what is exactly what the working group should nail down.
| Metric | Where it would come from | What we would need | The open question |
|---|---|---|---|
| Calls by outcome | The phone platform | Weekly counts of answered, voicemail and transferred, on a definition that does not change | Is a transfer counted once, or in both the sending and receiving queue? Do internal calls land in the same total? Who can answer this: ______________ |
| Visits by program | The queue system, plus whatever records service delivery | A weekly count per program against a single shared program list | Do the queue's service categories and the case system's program names agree — and which is authoritative when they do not? Who can answer this: ______________ |
| Rank movement | Derived from the per-program series | Five consecutive weeks of the same series | Nothing new, if the per-program series is stable. This is the cheapest slide in the deck to produce and the easiest to make misleading. Who can answer this: ______________ |
| Channel mix over twelve months | Queue system, phone platform and web analytics together | Three series on a common weekly calendar and comparable definitions | Can a visit, a call and a session honestly sit on the same axis? If a call and a visit are the same event by two routes, the total double-counts. Who can answer this: ______________ |
| Digital traffic | Web analytics, plus whatever runs the portal, booking and eligibility tools | Weekly sessions and completions per tool | Are those tools instrumented consistently, and does “completion” mean the same thing across all three? Who can answer this: ______________ |