Where the inputs come from
Every figure on this site is derived from mapping sessions run with real businesses. In a session, a business owner describes how one process runs today: what triggers it, who touches it, which tools are involved, how long each step takes and how often the whole thing runs.
Those descriptions are the raw material. We aggregate them across every business that has mapped the same process, strip anything identifying, and use the result to describe the process rather than any one company. Where a process has been mapped by only a handful of businesses, the figures are correspondingly rougher, and the mapped count shown alongside them is there so you can judge that for yourself.
How time identified is calculated
Each step in a mapped process carries an estimated number of minutes and a monthly frequency. Multiplying the two and converting to hours gives the manual time that step consumes in a month; adding the steps together gives the figure for the whole process.
On a process page, "manual time identified" is that monthly figure scaled by the number of businesses that have mapped the process, so it describes the total manual time we have observed going into that work rather than the load on a single business. The "companies have mapped" number beside it is the multiplier, shown so the basis is visible.
How cost is calculated
Cost is time multiplied by what that time costs. For each process we identify the role that actually does the work and apply a typical annual salary for that role, converted to an hourly figure on a standard full-time year.
Two consequences worth stating plainly. A process owned by an unpaid role has no salary cost, so its cost figure is genuinely empty rather than zero, and we show a dash instead of inventing an estimate. And because cost depends on both the time figure and the mapped count, it moves when either does.
What the process pain score measures
The pain score is a 0 to 10 indicator of how much friction a process tends to create: how repetitive it is, how many tools and hand-offs it spans, how often it produces errors or rework, and how much of it is waiting rather than working.
It is a comparative signal, not a measurement. It exists to help you rank one process against another when deciding where to start, and it is deliberately shown to one decimal place rather than as a precise-looking integer.
Where friction percentages come from
The friction points listed on a process page are the problems businesses raise most often when describing that specific work. The percentage beside each one is the share of mapping sessions for that process in which the problem came up, combined with published industry research where the sample is thin.
They are reported frequencies, not measured failure rates. A friction point at 60 per cent means roughly six in ten businesses mentioned it, not that the step fails six times in ten.
What "companies mapped" counts
It is the number of distinct businesses that have completed a mapping session for that process. It counts completed sessions, not page views, sign-ups or interest.
It is the honest denominator behind the other figures, which is why it appears next to them rather than in isolation.
What these numbers are, and are not
They are estimates built from aggregated, anonymized session data, and in places from published industry research. They describe what we have observed across the businesses that have mapped a process.
They are not guarantees, quotes, or audited measurements of your business. Your own figures will differ, sometimes substantially, because your volumes, tools, team and edge cases are your own. The point of a mapping session is to replace these aggregates with numbers specific to you, which is the output you actually take to a developer.
If a figure on this site looks wrong to you, that is worth telling us. The inputs improve as more businesses map the same work.
Questions about a specific figure? The FAQ covers how sessions work, or map a process and see your own numbers instead of ours.