Published, on a schedule, maintained
Recurring intelligence built for one business: a brief, a dashboard, a model that arrives every week and is kept honest between issues. Designed around your basket and your decisions, not a generic category feed.
Ark Solutions builds decision-grade data products — the recurring briefs, cost models and dashboards a business runs on. Every figure we publish carries a date, a named source and a stated confidence, so you always know how much weight it can take.
Most of it starts the same way: someone is making a decision with a number they cannot fully vouch for.
Recurring intelligence built for one business: a brief, a dashboard, a model that arrives every week and is kept honest between issues. Designed around your basket and your decisions, not a generic category feed.
Cost models, scenario work, benchmarking a quote against what a market actually charges. Delivered as a specification you keep and can rerun — not a slide deck that goes stale the week it lands.
Sourcing, reconciliation and automated checks that sit between a data provider and whatever you publish. Including harnesses that stop a report from going out when the data behind it does not hold up.
Our standing product line, and the clearest example of how we work.
A published commodity price is a starting point. By the time a tonne clears a plant gate it has picked up ocean freight, insurance, duty charged on that freight, terminal handling and haulage — and none of it is in the number on the screen. The Procurement Brief prices the whole distance, for one buyer's actual basket and actual lanes.
It goes out weekly, with a monthly and an annual close, in two editions: a short chart-led read for the people who approve, and a fuller data-and-playbook edition for the people who buy.
We never present an estimate as an observation. Where a number is modelled, the page says so.
A, B or C — published by the body that sets the price, a real quotation, or modelled. The grade is printed next to the number, not buried in a methodology note nobody opens.
No percentage is published unless the current and prior readings came from the same series and can be recomputed. Where they cannot, we print the level and say so.
Validation runs before anything renders — staleness, arithmetic, basis and source checks. If a check fails, nothing publishes. It is not a review step that gets skipped under deadline.
Where no one publishes a price, we say no one publishes it, then model it and label the model. Single-source figures are named as single-source. Open questions stay open.
The useful first conversation is a specific one. Send the decision you are trying to make and the numbers you are having to trust, and we will come back with a straight answer on whether we can help.
We will reply to you shortly.
Open it in your mail client, or copy the text and send it to reports@ark-solutions.ch yourself.