Evidence your results framework can use.

What is recorded, and what it can feed

Operations software produces one thing a results framework needs and rarely gets: a dated record of work, entered by the person who did it, at the moment it was done.

• Customer and citizen requests — volume, category, district, channel, time to first response, time to closure. Feeds service-responsiveness and customer-interface indicators.
• Maintenance — preventive tasks due and done, corrective work orders, time to repair, the intervention history of each asset. Feeds operation-and-maintenance capacity indicators.
• Assets — what exists, where it is, what state it is in and when it was last touched. Feeds asset-management and sustainability indicators.
• Meter readings, where they are taken through the platform — reading coverage, and the share estimated rather than read. An input to commercial performance and to the commercial side of non-revenue water; not a measurement of either.
• Training — which procedure was followed, attached to the work it belongs to. Completion recorded for each agent, exportable. Feeds capacity-building indicators.

What it will not claim

The platform records work. It does not produce the hydraulic or the commercial outcome, and it will not tell you that an indicator moved because of it.

Non-revenue water falls for several reasons at once — a main replaced, a tariff revised, a wet season, a repair crew that got faster. What the record shows is which interventions happened, where, when and by whom, with the evidence attached. Turning that into attribution is your evaluation’s work, not our software’s.

We would rather write that sentence into the programme document than be asked for it at the mid-term review.

How a figure is produced

An indicator is worth what its lineage is worth. Four things make the difference, and none of them is a dashboard.

Entered once, where the work happens

By a named agent signed in through your own directory, on the phone, at the moment of the job — dated, with photos, and with the asset identified by its tag rather than by memory. Not retyped at the end of the quarter from a paper register: a figure reconstructed after the fact is a figure nobody can defend.

Work performed, equipment status, dates and evidence build a reliable operational record over time.

Computed without a model

Core reporting — KPIs, response times, trends, comparisons — runs without any generative model.

Where a model is used at all, it is for a defined task, on limited data, with a controlled output and a logged call. No reported figure depends on one.

Recomputable by someone else

The underlying records export in an open format at any time, so an evaluator or an auditor can rebuild the figure instead of taking our word for it. That is the difference between a reported number and an evidenced one.

Measured against a baseline recorded the same way

An indicator only means something against a baseline collected by the same method. That is worth setting up in the first weeks of a programme, rather than reconstructing at mid-term from whatever can still be found.

Reporting

Clear daily, weekly and monthly reports from your platform data and external sources — ready for the board, the regulator or the funder. Analytics first; AI only where it adds value.

The calendar is yours, not ours. If the programme reports quarterly against a fixed template, that is the shape the digest takes, and the operator’s own teams receive the same figures rather than a separate version prepared for the funder.

When the funding stops

Every results framework carries a sustainability line, and it is usually the weakest page in the document. It is better answered with mechanics than with intent.

What the utility keeps, whatever happens

Your data is exported in an open format, and the training videos stay usable with no time limit: at the end of the contract, whatever the reason, the utility keeps the video files.

The record belongs to the utility, not to the programme. When the programme closes, the operational history does not leave with it.

What it then has to pay for

Pricing has two components: an initial deployment and a recurring operating fee. There is no public price list: each engagement is quoted on its scope.

Designed to stay affordable when the project funding ends, so the utility can keep running it on its own budget.

Only the second component continues. It is quoted against the scope and written into the programme document with a figure, not an adjective, so the first year after the funding is costed before the programme starts rather than discovered when it ends.

What the fee never depends on

Not the number of users. Field teams are never charged per agent, so putting the application in every hand does not increase the bill.

This matters more to a programme than to anyone else. Adoption is usually a programme objective; it should not also be a cost curve.

Where the data lives

On-premise or in-country. Encrypted, backed up and exportable at any time. No proprietary format. No lock-in. A simpler architecture also means fewer exposed components.

See how Sparkoper is built

The dossier a financing asks for

Scope, deliverables, indicator mapping, cost structure, continuity plan and risk register, in the format the financing requires rather than in ours. We have written one for a commercial bank, a national export agency and an export credit insurer.

It is not published. We send it on request, with the operator’s agreement.

What we need from you

• The results framework or logframe, with the indicator definitions as they will be audited — not as they are summarised.
• The reporting calendar and the template the programme reports against.
• The baseline, if one exists, and how it was collected.
• The data-protection and hosting requirements that apply to the programme.
• The languages the interface and the training content have to run in.
• The end date of the funding, and who is expected to pay after it.

Talk to us early

The cheapest moment to fix an indicator is before the baseline is collected. The most expensive is the mid-term review.

If you are an operator or a consortium bidding for the programme rather than funding it, the companion page is For PPP bidders.