The easiest version of a location scavenger hunt is the one where a human builds it: someone walks a neighbourhood, photographs a dozen spots, writes the clues, and publishes a route. That works, and for a city with an active local scene it is a lovely format.
It does not scale. Nobody is going to hand-build hunts for every neighbourhood in every city, and the ones that exist tend to concentrate in places where somebody was already willing to do the work.
So we went the other way with Ping Hunt: the hunt is generated from street-level imagery around the player. No author, no route to design, no catalogue to maintain. If you are standing in a street, you can start a hunt.
That sounds like it removes the hard part. It moves it, and the problems it creates are more interesting than the ones it solves.
"What is a spot?" is the entire problem
A player matches the scene with their camera to unlock the next location. For that to work, the thing in the photo has to be identifiable from the image alone - and a surprising number of candidates are not.
A red door is a red door. So is a café table, a bollard, a statue, a row of parked bikes, a bin. If the reference image is "a red door" then the matching task collapses into "find any red thing", and the player will unlock the wrong spot within thirty seconds. Worse, the game becomes trivial: if any red thing counts, the player is really just looking for red.
The fix was to get much more specific about what makes a good target:
- Prefer geometry. Corners, staircases, shopfronts, arches, steps and distinct rooflines are identifiable because of their shape, not their colour.
- Reject flat surfaces. Grass, asphalt, painted road lines, sky and blank walls are the most common things in street imagery and almost none of them are unique.
- Require local distinctiveness. If two candidate spots on the same street look the same, only one of them is usable.
- Keep the clue honest. The clue text should describe what the player will actually see, not what the reference image happens to contain.
This single change - designing around what can be recognised rather than what can be photographed - did more for the game feel than anything else we tried.
Coverage is a product feature, not an error state
A generated hunt is only playable where the underlying imagery exists, and only good where the imagery is recent and the streets are legible.
We treat coverage as a property of the map, tile by tile, and we are honest about it before the player commits. If there is not enough usable material near you, the right answer is to say so and suggest somewhere else - not to generate a hunt that becomes frustrating halfway through.
That sounds obvious, and it took us a while to learn. The failure mode is seductive: you always find something within a radius, so the system feels generous. But "we found four spots for you" and "these four spots are actually distinguishable and findable" are different claims, and only the second one produces a game people finish.
Stale imagery is a feature of the real world
Street-level imagery ages. A shopfront in the reference photo may be a different shopfront now. A bus stop may have moved. A mural may have been painted over. A tree may have grown.
Two consequences:
Match tolerance has to be generous. If the reference image is from six years ago, insisting on a near-pixel match guarantees failure. We score similarity and accept a range, with the threshold set by how old the imagery is.
The confirmation message has to teach. When a player matches a spot, the useful thing to show is not "correct" but "that is the corner with the blue awning" - a description that survives the shopfront changing colour.
The honest framing is that the game is played against a memory of the street, not the street itself. Players accept that readily, because they are used to it from every other map.
A hunt has to be walkable
A list of coordinates is not a route. The spots have to be sequenced so that the walk makes sense.
Our rules, which are less sophisticated than a proper route planner but far more predictable:
- Cap the radius. Everything must be within a walk of the start - roughly 800 metres for an urban hunt. Beyond that it stops being a scavenger hunt and becomes an errand.
- Sequence to avoid backtracking. Order the spots to reduce zigzag, because a route that crosses itself twice feels broken even when each individual step is short.
- Budget the total. A hunt should be about fifteen to twenty minutes of walking for a casual player. A two-hour hunt is a commitment the game has not earned yet.
- Prefer interesting streets. Spots on a main road with traffic noise are worse than a side street, even if the side street is a few metres further.
Hints should degrade, not fail
Binary hints - you either know where it is or you do not - make the game frustrating the moment a player is lost.
We escalate instead: first the neighbourhood, then the street, then something more specific about the spot itself. Each hint is weaker and more specific than the last, so a stuck player always has a next step and can decide whether to take it.
The detail that makes this work: a hint should narrow the search area, not just describe it. "It's near a canal" is a description. "It is on the canal side of the street" is a constraint that shrinks the space the player has to search.
The fallback ladder matters more than the generator
When the system cannot build a good hunt, it should have a graceful sequence of less-good options rather than failing outright:
- Widen the radius a little.
- Widen it further, accepting a longer walk.
- Lower the difficulty - choose less distinctive spots.
- Finally, say plainly that there is not enough coverage here and point somewhere else.
Each rung is worse than the last, and the player should feel the quality drop before they commit. A hunt that is slightly generic but playable beats a hunt that is perfect in theory and impossible in practice.
Testing this without going outside
The nice thing about a system driven by imagery and coordinates is that both are files. We can test the generator without walking anywhere.
Take a recorded GPS trace from a real walk, replay it against the generator at a hundred different start offsets, and count how many hunts come out playable: how many spots are distinguishable, how far the walk actually is, whether the sequence backtracks, how much of the route is on the right side of the street. Coverage holes show up immediately as regions where the generator produces hunts with fewer than the minimum number of usable spots.
Then you still have to go outside, because the thing the tests cannot tell you is whether a person can find the spot while holding a phone, in the rain, with one hand.
What I would tell someone starting this
- Design around recognisability, not around photography. The constraint shapes the whole content pipeline, so decide it first.
- Treat coverage as a first-class property. Know where your good areas are, and refuse to play where you do not have good material.
- Cap the scope of the hunt. Short, complete, repeatable is the target. Not ambitious.
- Make hints a ladder, not a wall.
- Build the fallback path before the generator. The moment you cannot produce a good hunt, you need to say something useful.
The promise of generated hunts is that any street becomes playable. The work is entirely in the cases where the street does not want to cooperate - and in being honest about those instead of shipping them.
If you are building something similar, I am happy to compare notes; Ping Hunt is out, and the awkward parts are mostly written down.
Top comments (0)