How it works
How it works, and how we count.
One engine runs every Scavvi trail. Here's how an experience is made and checked, how it reaches people, and what every number means.
a carving becomes a clue
A reason to look closer.
A square becomes a story. A nearby business becomes an optional next stop. Design the journey with local people so it suits the place, and gives people a clear choice about time and cost.
Explore together
Short missions invite people to look at the place, talk and make discoveries at their own pace.
Connect local partners
Keep the place's own café, shop and booking journey in the plan. Optional onward stops can introduce nearby places after the main visit.
Learn what to improve
Agree a useful question before collecting anything. Review participation, feedback and staff effort alongside any commercial evidence.
Ways in
Three ways to reach people.
A link or QR code
Your own web experience, in your colours. People scan the code at the door, or tap a link before they come. No download.
In developmentYour own channels
Inside your website, app, emails or tickets, or under your own name and web address.
In developmentThe Scavvi app
Where it suits families who already play Scavvi, with your name on it.
The app itself is live on iPhone and Android.
Partner listings in development
Each way needs its own content, set-up, permissions and support, agreed before launch. App events don't show the physical path people took, or what they spent.
Questions first
Start with a question.
Every report starts from the questions in the brief. What we can see, what people tell us and what would need a comparison are different kinds of evidence, so we keep them apart.
| Question | What we can observe | What people tell us | What needs a comparison |
|---|---|---|---|
| Did people enjoy it? | How many finished, where people stopped, and repeat plays. | Optional ratings and short comments. | Whether it made the visit better than it would otherwise have been. |
| Could everyone take part? | Use of the short, step-free and camera-free options. | Access feedback and requests for help. | Who didn't start, and why. |
| Which stops worked? | Stops reached, skipped or retried, and time to find. | Comments about particular stops. | Whether a change to a stop made it better. |
| Did it change visiting? | Plays during the dates it ran. | An optional question, such as whether they'd have come anyway. | Extra visits, return visits or longer stays need a comparison, or your own records. |
| Was it worth the work? | Time spent on content, set-up and support. | Notes from your staff. | Whether the results justify the cost, against other things you could have done. |
Finishing alone shows neither enjoyment nor an extra visit.
Definitions
Know what the numbers mean.
A pilot should separate what people do in the experience from what happens across the wider place. These are measurement definitions to agree and check, not a claim that every field is collected today.
| Measure | Interpretation | What it can tell you |
|---|---|---|
| Participation | Agree the unit | Starts and finishes describe participation once a start and a finish are defined. Player records, device sessions and real groups of people are different counts. |
| Checked finds | Participating players only | Location and photo checks can support a recorded find at a stop. They can fail or be inaccurate, and they don't count everyone visiting the place. If the photo check can't run, the app still counts the find at fewer points, so a report must keep checked and unchecked finds apart. |
| Order of finds | Recorded sequence | Timestamps can show the order of recorded finds. They don't show the physical path taken between them. |
| Time to find | Elapsed interval | The interval can include walking, queues, breaks or time away from the trail. It is not time spent at a place. |
| Who took part | An optional question | A player count doesn't show who was in a group. We don't ask about age or family by default. Any question about who took part is optional, needs an agreed purpose, and is reported with its response rate. |
| Enjoyment and accessibility | Participant feedback | Agree short questions and report the answers alongside the response rate. Feedback from the people who answer doesn't represent every visitor. |
| Referrals and purchases | Separate evidence | A link click shows interest. A redemption or booking record can show an attributed transaction where collection is agreed. Neither alone proves additional spending. |
| Wider impact on a place | Requires a comparison | More visitors, extra nights or extra local spending need separate data and a credible comparison. Activity in an experience can't show them on its own. |
Three rules for every report
- Start and finish will be defined and agreed before launch.
- Groups under 5 will be hidden in every breakdown.
- Every chart will show how many people it covers.
A concept
A reporting concept to discuss.
This sample shows how agreed measures might be presented, for a fictional example town. All figures are fictional. Definitions, collection, exports and privacy controls must be checked before a real report is used.
Example town: Learn report
Summer edition, six weeks (fictional)
- Started
- 1,247Groups that began the trail
- Finished
- 89372% of those who started
- Optional ratings
- 41246% of those who finished
- Median time to find
- 6 minIncludes walking and breaks
Which stops worked?
Groups that recorded a find at each stop. This shows how far people got, not how many people visited each place.
Fictional sample: 1,247 starts. Checked and unchecked finds would be shown apart.
Did people enjoy it?
The optional rating at the end. People who answer are not everyone, so the response rate sits next to it.
Fictional sample: 412 of 893 who finished answered.
Could everyone take part?
Which route people chose at the start. Who didn't start at all can't be seen here.
Fictional sample: 1,247 starts.
When do people play?
Starts by day and time. Any cell under 5 is hidden, so no small group can be picked out.
Fictional sample: 1,247 starts.
| 8am | 10am | 12pm | 2pm | 4pm | 6pm | |
|---|---|---|---|---|---|---|
| Mon | hidden, fewer than 5 | 22 | 34 | 31 | 18 | hidden, fewer than 5 |
| Tue | hidden, fewer than 5 | 20 | 31 | 29 | 17 | hidden, fewer than 5 |
| Wed | hidden, fewer than 5 | 24 | 36 | 33 | 21 | 6 |
| Thu | 6 | 27 | 40 | 37 | 24 | 8 |
| Fri | 9 | 35 | 49 | 46 | 33 | 12 |
| Sat | 14 | 62 | 88 | 80 | 52 | 16 |
| Sun | 12 | 56 | 78 | 70 | 44 | 11 |
If you need it
Test the commercial value.
Where partners want a commercial test, agree tagged referrals, booking records or offer redemptions. Record the place's own café and shop results alongside participating businesses, discounts and staff time. A comparison period, or a suitable comparison group, helps assess whether activity is additional.
A missing tracked purchase is not proof that someone did not spend. Transaction matching needs partner agreement and suitable coverage. This concept promises no payment integration and no extra revenue.
A first pilot
Shape a first pilot with us.
This is how we'd shape a first one with you, one small, useful experiment at a time.
Choose the question
Pick a place, a local owner and a visitor problem worth solving. Decide who the first audience should be.
Agree the scope
Set the route, the way in, everyone's duties, the budget and the measures. Keep the first build small.
Check it locally
Walk-test the route, review access and content, check the data collection and support, then agree a launch window.
Make the next decision
Review what you found, and its limits, together. Decide what to change, whether to repeat it, and what a paid programme would need.
FAQ
Questions partners ask.
- What would a first pilot include?
- Start with one place or local route, a named person on your side and a question worth testing. Agree content, the way in, reporting, support and any fee in writing. If we've already sent you a written offer, we'll honour it within its agreed scope.
- What is available now?
- The Scavvi app is live on iPhone and Android, with 25 city trails in 20 cities and a photo check on every find. The studio, your own web experience, partner listings in the app and Learn reports are in development. Screens and figures on this page are fictional examples.
- How do the three ways in differ?
- A listing in the Scavvi app reaches families who already play. Your own web experience opens from a link or QR code, with no download. Under your own name, it sits in your own channels or your own app. The extra depth comes from the content, set-up and support, not from branding alone.
- Who looks after content?
- We work with someone local who checks the route, the stories, permissions and practical information. You approve the content, decide who can pause it, and help visitors find it. Who looks after it after launch is agreed before launch.
- Who can access the data?
- Data responsibilities, permitted uses, retention, hosting and access belong in the pilot agreement. Partner reports will use grouped totals. Small groups will be hidden, and reporting definitions checked, before any real participant data is shown.
- Who is it for, and how are children protected?
- An experience can be for families, adults or mixed groups. Where children may take part, extra rules will switch on: no photographing people, invite-only groups, no ads and no chat with strangers. Partner reports will never show an individual child, their photos or their route. We begin with adult-led testing and short, optional feedback.
Talk through a first pilot.
Scavvi began with Harper wanting real places to feel like games. The next step is to work with local people and learn where that idea helps.
Talk through a first pilotOr write to us at hello@scavvi.com