Skip to content Need urgent help?
Pre-launch

There are no case studies.
Here is what ours will contain.

SHELY has not launched. No hospital, lab, pharmacy or community health team has completed a pilot with us, so there is no result here to read. What is here instead is the standard we are holding ourselves to before we have any results: the measures we intend to be judged on, how a pilot works, and what gets published when one goes badly.

This page will never carry an illustrative example, a composite customer, or a hypothetical hospital with a plausible-sounding name. If a story appears here, it happened, and the partner agreed to it in writing.

What a SHELY case study will measure

Committing to the measures before we have any is the only part of this we can do honestly today. Each one is defined so that a partner could recalculate it without us.

  • Lab

    Sample to report, in hours

    Counted from the barcode scan at collection to the moment the report is available to the patient — not to the moment the lab signed it off. Reported as a median and a 90th percentile, because an average hides the samples that get stuck.

  • Lab + Vault

    Results collected without a phone call

    The share of patients who receive a report without anyone ringing the lab to chase it. Checked against the lab's own inbound call log, not against our app analytics.

  • Rx

    Batch and expiry captured at the counter

    The share of dispensed items with a batch number and expiry date recorded at the point of sale. This one is on the list precisely because anyone auditing the stock ledger can check it.

  • Hospital

    Check-in to seen, in minutes

    From the check-in timestamp to the start of the consultation. We will report the queue in its worst hour of the week, not its quietest.

  • Doctor

    Prescriptions readable before the patient leaves

    The share of consultations that end with a prescription the patient can open on their own phone, in their own language, before they walk out of the building.

  • Assist

    Follow-ups closed inside the window

    For community health teams: how many scheduled follow-up visits were actually completed within the intended window — and how many were missed.

  • Every app

    Staff still on the old system in month three

    The measure most likely to embarrass us. If the paper register stays open next to the screen, the pilot has not worked, and we would rather report that than count logins.

Every one of these can move the wrong way. That is what makes them worth writing down in advance.

Who we are looking for first

The honest shape of an early pilot. We would rather be genuinely useful in a few places than thinly present in many.

A single lab, or a small chain

Enough daily volume that turnaround time means something, and small enough that we can sit next to the people doing the work and change the software the same week.

One or two pharmacy counters

Dispensing and batch-level inventory are the parts of Rx furthest along. A counter currently running on a register and a spreadsheet is the sharpest test we can ask for.

A clinic or a district health team

Somewhere genuinely multilingual, where the interface being in Tamil or Bengali is not a nice-to-have. We build in ten languages and we need to find out whether that survives a real shift.

We are based in Chennai, so the first pilots will most likely be in Tamil Nadu, for the unglamorous reason that we can be in the room. That is a constraint, not a strategy.

What a pilot actually involves

Nobody has run one of these yet, so treat it as our opening offer rather than a settled process. It is written down so you can argue with it.

What SHELY provides

  • Setup, migration from whatever you run today, and staff training — at no cost for the length of the pilot.
  • A named person on our side you can phone, rather than a ticket queue.
  • Fixes and changes shipped during the pilot, not filed for a later release.

What you provide

  • Staff time. Realistically the first fortnight is slower, not faster, and it helps if everyone knows that going in.
  • Two weeks of baseline measurement before you switch, so the numbers afterwards have something honest to sit against.
  • Blunt feedback, including the parts that make us look bad.

Length, data, and walking away

  • We propose twelve weeks: two measuring, two switching over, eight running properly.
  • Your data stays yours. You can export it in an open format on any day of the pilot, including the last one.
  • If you stop, you keep the export and owe us nothing. There is no clause that makes leaving expensive.
  • You are not named anywhere — site, deck, or press — without written consent, and consent can be withdrawn afterwards.

What we will publish, and what we will not

A case study is only worth reading if the rules were set before the result was known. These are ours.

We will publish

  • The method next to the number: what was counted, over what period, against which baseline.
  • Results that went the wrong way. A measure that got worse is published beside the ones that improved.
  • Pilots that were abandoned, and the reason they were abandoned.
  • The partner’s own words, unedited, if they want to write them.

We will not publish

  • Patient data. Not identified, not de-identified, not dressed up as an anonymous vignette.
  • A partner’s name, logo, or quote without written consent on file.
  • A percentage without the denominator printed underneath it.
  • A comparison against a named competitor’s product.
Get Started

Ready to write your success story?

Somebody has to be first, and the first pilot gets more of our attention than any partner after it ever will. If that is a trade you want to make, tell us what you run and what is currently going wrong with it.

No pilot has started yet. When one has, this page changes.