FestQuest dashboard, live example
FloydQuest
FloydFest 26 “Daydream” · Jul 22 to 26
How players moved between checkpoints
Each arc leaves the checkpoint it starts from and takes that checkpoint's colour, matching the route list below. A to B and B to A count separately. Scans more than 4 hours apart, or repeats of the same sign, are not movements.42 distinct routes observed across 7 checkpoints.

Busiest routesTRIPS counts every walk along this route, including the same person doing it more than once. PEOPLE counts how many different players walked it at all. They match when nobody doubled back, which is normal at a small event where each checkpoint is scanned once. When trips runs ahead of people, that route is being re-walked: at FloydFest 16 of 56 routes were, one of them 5 extra times. That gap is the interesting part, because it shows a path people return to rather than pass through once.
Exploration
Player location was never tracked. Everything in this section is worked out from which signs somebody scanned and when, so it describes their SCANNING rather than their movement.- Someone who scanned a single checkpoint has no journey to measure, so they are left out of how far they walked rather than counted as a zero, which would drag the number down. They ARE counted in coverage and in days played, where one scan is a real answer.
- The Entry Poster is not counted as a defined location as the same QR was used for both the Entry Poster itself + around 1500 Table Stickers, so the origin scan could be anywhere in the park.
How much of the map they saw
Checkpoint coverage: how many different checkpoints each player scanned, out of the total on site. Cumulative, so each bar counts everyone who reached that many OR MORE. It can only fall from left to right, and the size of each step down is how many people stopped there.Players who found at least this many of the 8 checkpoints.
Steepest drop is 1 to 2: 122 players stopped at 1. 331 players counted.
How many days they played
Distinct festival days each player scanned on, counted in the festival's own timezone. Everyone with a scan played at least one day, so unlike the other charts here nobody is left out.Whether the quest brought people back, or was a one-time thing on the way in.
154 of 331 players came back on a second day or more.
How far they walked
Distance roamed, a lower bound: measured as straight lines between the checkpoints a player scanned in order. Player GPS was never on, so this is inferred from scans rather than tracked. Real paths curve and wander, and any walking between scans is invisible to us, so the true figure is always higher.At least this far. Straight lines between scans, so real walking was further.
walked the least0.88 mi1.72 mi
walked the most
- Lower quarterA quarter of players are at or below this. The quieter end of the range.
- 0.43 mi
- Upper quarterThree quarters of players are at or below this, so a quarter are above it. The keener end.
- 0.88 mi
- AverageEvery player's value added up and divided by how many players there are. A few unusual players can pull it away from the typical.
- 0.67 mi
- Players countedHow many players this particular number was worked out from. 47 others are left out: measuring needs two checkpoints we can place on the map.
- 203
The shape of a festival day
Every scan, including repeats, by the hour it happened in festival time, added up across all five days. Midnight at the top, running clockwise. Further from the centre means busier.When people were out scanning, across the whole festival.
Busiest hour is 5 PM, 291 scans (14% of all of them). The three busiest hours carry 35%. Outer ring is 300 scans.
Tier funnel: players who reached each tier
Each step is a SUBSET of the one above it, so the bars can only shrink. What matters is the size of each drop: that is how many people stopped there.134 did not go on from their first scan
16 did not go on from Hiker
29 did not go on from Wanderer
14 did not go on from Wayfinder
Players who found each checkpoint
People, not scans: each player counts once per checkpoint however many times they scanned it. The flow map above counts every scan including repeats, which is why its numbers are larger.What happened to every reward earned
Each tier's bar is everyone who qualified for it, split by what they actually received. The outcomes are exclusive, so the segments always add up to the whole.Stopped being handed out Sun 7:21 PM · everyone after that got +1 giveaway entry
Stopped being handed out Sun 5:20 PM · everyone after that got +1 giveaway entry
Stopped being handed out Sun 12:26 PM · everyone after that got 2X giveaway entries
279 items were handed over at the booth. 208 players got a Plan B substitute instead, because the item had run out and they had never collected one. 181 earned the coupon, which is fulfilled by email rather than at the booth and so can never show as a handout.
Checkpoints found per day
Same measure as the Checkpoints found tile, so these add up to it. A repeat scan of a sign somebody already had is not counted.Networks that carried the fest
From 328 players with a usable address. A block we cannot identify is shown as such rather than folded into a guess. 37 used a privacy relay at some point, which hides them from any carrier or location reporting, including this one.Each player counted once, under whichever network carried most of their scans.
Apple routes iCloud Private Relay traffic out through Fastly, Cloudflare and Akamai. These players are invisible to any location or carrier reporting, including ours.
The local Floyd, Virginia co-operative, and almost certainly the site’s own wired network: thirty players on its main block shared just three addresses between them. More players touched it than count here, since this row is the network that carried MOST of a player’s scans.
An Alaskan provider, which is where that lone Eagle River reading came from.
Where those networks are registered
77 of these players were given more than one city over the weekend without moving anywhere.The registered location of each address block. NOT where players live.
316 of 331 players had a location on their address block. 26 in all, scroll for the rest.
What changed while the gates were open
In the order it happened.
- Wed 8:50 AM
Per-scan tracking went live
Shipped on opening morning, forty minutes before the first player scanned.
- Thu 9:45 AM
Fixed a map bug on Android
Tapping the site map could open the wrong stage schedule. Found and fixed live on day two.
- Thu 3:12 PM
The Wanderer reward changed
FloydFest merch could no longer cover the fourth-checkpoint tier, so it became a 10% coupon for The Exchange Music Hall, emailed after the fest rather than collected at the booth. 181 players ultimately earned it.
- Thu 4:00 PM
Fixed sign-in losing your place
Signing in with Apple mid-scan returned players to the home page instead of the checkpoint they had just scanned.
- Fri 6:12 PM
The growler ran out, and the swap needed a deploy
The top reward was replaced with double giveaway entries by editing the app and shipping it. Growlers came back the next afternoon, indicating a need for a "Booth stock switch". Great lesson.
- Sat 12:00 PM
More growlers and drink rings arrived
This is why the units left at the end are not a leftover from an opening count, and why what was on hand at the start cannot be worked out from them.
- Sat 3:42 PM
Booth stock became a live switch
Prompted by the growler. From this point a reward could be flipped to its backup from a phone at the booth in seconds, with no deploy. Everything that ran out afterwards was handled this way.
- Sun 12:26 PM
Vintage FloydFest growler ran out
The last one went at the booth. 90 players had earned this tier and never collected one, so they received 2X giveaway entries instead.
- Sun 4:45 PM
Live inventory counts
New feature added: The booth could enter how many of each item were left, and the app switched to the backup reward automatically on reaching zero. Two of the three Sunday stock-outs were handled with nobody having to press anything.
- Sun 5:20 PM
2025 pint cup ran out
The booth count reached zero. 45 players had earned this tier and never collected one, so they received +1 giveaway entry instead.
- Sun 7:21 PM
Booth closed for the 2025 pint ring + carabiner
Closed about forty minutes early: the crowd had thinned and the other vendors had already packed up, with roughly 50 rings and carabiners still in the box. 73 players had earned this tier and never collected one, so they received +1 giveaway entry instead.
What we would do differently next time
The point of the timeline above.
Lock the reward ladder before gates open
Two of the four rewards changed mid-festival, and the top one changed twice inside a day. Nobody did anything wrong, the merch simply was not there.
Count the stock on the way in
Every stock-out was a surprise on the day. Inventory counts arrived ninety minutes before the last item ran dry. Knowing the numbers on Wednesday would have turned three surprises into a forecast.
Assume anything printed will end up somewhere else
The entry poster QR also went onto 1,500-plus table stickers, which was a good idea and drove the single biggest scan count of the fest. It also means that checkpoint has no usable location, so it is left out of every distance measure here. Next time a code that will be duplicated should be issued as its own code from the start.
Give the backup reward a bigger margin
All three physical rewards ran out on the final afternoon, within seven hours of each other, and 208 players received a substitute rather than the item. The substitutes worked, but the ladder was sized for a smaller finishing crowd than turned up.
Brainstorm on how to incentivize player to get to the second scan
The steepest drop in player fall-off is from Scan 1 to Scan 2: 119 players stopped at 1.
Want one of these for your event? Read the FestQuest pitch.