Proving a trail worked once the funding has gone.

The Heritage Fund pays the last 20% of a grant only, in its own words, "when you have finished your project and sent us a final completion report, project evaluation and proof you have acknowledged your grant". For most trail projects that evaluation comes down to sixty clipboard responses and a sentence about positive anecdotal feedback, because a physical trail produces almost no evidence on its own.

The Fund knows this. Its guidance for grants of £10,000 to £250,000 says "we recommend you build in evaluation from the beginning of your project", and adds that "the more carefully projects budget for their evaluation, the higher the quality of the final report". Evaluating heritage projects is still mostly left to the last month, which is why the reports read the way they do.

What a physical trail can tell you

Very little, and none of it about the thing you were funded for.

Boards can't count. A path counter can, but it counts passes at one point, so the dog walker doing the same loop every morning counts the same as the visitor who read every board, and neither tells you where anyone stopped or turned back. Leaflet uptake tells you how many leaflets left the rack. A survey tells you about the people who stop for surveys. Families with tired children, people in a hurry and anyone who would rather not talk to a stranger are invisible to it, and they are a large part of who walks a trail.

So the final report says that the trail was well received, quotes three visitors, and gives a footfall figure from a counter that also counts the postman. Assessors read that paragraph in report after report. It satisfies nobody, including the people who wrote it.

What a digital route can measure

A route that people follow on their phones can record what happened at each stop without asking anyone anything. That changes what is possible, but each measure has limits, and a report that ignores them is the report that falls apart at the first question.

Completion rate is the number everyone leads with and the softest one there is. It tells you what share of the people who started reached the end. A short route will always beat a long one on completion, and a route where people can tap through the last few stops from the café will flatter itself. Use it to spot a route that is badly too long. Don't lead with it.

Stop-level drop-off is the useful one. It shows where people give up, and when a route loses a quarter of its walkers at the same stop every week, that stop is telling you something specific. Usually the text is too long, the stop is in the wrong place, or there is nowhere comfortable to stand and read. It is the only measure that tells you where the writing failed, which makes it the one that leads to a better trail rather than a better chart.

Dwell time per stop is good evidence of attention, with one caveat. It measures how long a phone stayed still near a stop, which is also what happens when someone takes a call. Across a season the noise evens out. On any single day, don't trust it.

Repeat visits are the strongest signal and the slowest to arrive. Someone who walks your second route six months after the first is evidence a funder cares about, because it says the first was worth coming back for. Measuring it means recognising a returning phone, which has a privacy cost some organisations rightly won't pay, and it takes a full season or more to accumulate. Most evaluation reports are due before then. Say so in the report, and say why it is worth tracking anyway.

Two limits apply to all four. A phone is not a person, so a family walking on one phone counts once, and a route popular with groups will always undercount. And a stop with no signal records nothing unless the platform stores check-ins and sends them later. Describe the numbers precisely, as recorded check-ins rather than visitors, and the report will be believed.

Report against the framework you already have

Funders and boards each have a framework, and the evidence lands better under its headings than under a new one of your own.

For a Heritage Fund project that means the four investment principles: saving heritage, protecting the environment, inclusion, access and participation, and organisational sustainability. Drop-off by stop and changes made because of it belong under organisational sustainability. Who walked the route, and when, belongs under inclusion, access and participation.

Destination organisations mostly report visitor numbers and spend through STEAM, which models the visitor economy from accommodation and attraction data. Route data sits beside a STEAM figure as something measured rather than modelled, and should be labelled that way so nobody adds the two together. For a Business Improvement District the audience is levy payers at a renewal ballot, and the claims have to be narrower still, which is the subject of evidence a BID can take to a ballot. Museums and galleries reporting on learning often use the Generic Learning Outcomes from Arts Council England's Inspiring Learning for All, and dwell time at a stop is a reasonable supporting measure for the ones about engagement.

Set it up before you build

Evaluation designed after launch can only describe what happened to have been recorded. Designed before, it shapes what gets built.

Decide what you will measure before the route is written, and take a baseline before it opens: visitor numbers at the site, leaflet uptake, whatever you already count. Decide what counts as a start, too. Someone who opens the route from a poster in the car park and someone who walks to the first stop are different people, and a report that counts page views as walkers will be found out by the first person who asks how the number was made. Decide which stops matter most, so you know which drop-off figures to watch. Budget for the evaluation under its own line, which the Heritage Fund asks for anyway in its "Evaluation" cost category. And plan when you will look at the numbers mid-project, because the most persuasive thing a report can contain is a change you made because of them.

That last point is why stop-level data is worth more than any headline figure. "We rewrote the stop where people were giving up, and fewer gave up" is a sentence that shows a project being managed. It is also the sentence boards remember.

Work backwards from the sentence

Start with the sentence you want to write in the final report, and build the evaluation that would let you write it honestly. Something with this shape: in its first season, [so many] phones started the route and [so many] reached the end; most of those who stopped early stopped at the same stop, where the text was the longest on the route; after we cut it, the drop there fell by [so much].

Every bracket in that sentence is a decision you make before launch: what counts as starting, what counts as finishing, which stop you are watching, and when you will look. Make them early and the final 20% of the grant is a formality. Leave them to the last month and you are back to the clipboard. If the funding itself is still being worked out, where the money for a trail actually comes from covers that side, including why evaluation belongs in the budget from the start.

If you have an evaluation plan due and want it read against what a route can actually measure, send it over and we'll mark it up.