Build your reference
library in an afternoon.
You already give the same six corrections every week, and one afternoon of filming turns them into twelve clips you can send between sessions.
Create a reference feed.
A reference library is one feed of correct execution and named errors, tagged so you can find a clip for an athlete you next see in a month.
Create the feed
- New feed, type set to Reference (reference feeds need a Pro account).
- Set the sport first, since the sport is what loads the right tag vocabulary.
- Name it organisation first, purpose second.
- Organisation | Reference
- sportsense | Mountain Biking Reference
One library per sport
- One library per sport, so mountain biking and skiing are separate feeds, but downhill, enduro and trail are one.
- Finish the clips before you invite athletes by link, email or QR code.
Record or import your clips.
Two ways into the library, record straight to the feed or import what you already have, and either way each clip lands as its own post.
Record in the app
- Open the sportsense camera and film the rep, and the green arrow in the bottom right chooses the feed it sends to, so point it at the reference feed.
- Say the skill out loud as you film, and sportsense starts the tagging for you.
Import what you already have
- In the reference feed, tap the plus in the bottom right to add clips from your camera roll or your sportsense library.
- Old footage counts, so the demonstration you filmed last season is a reference clip once it is titled and tagged.
How to film a clip.
A reference clip is the one you will compare every athlete against, so capture it as efficiently as possible.
- Four to eight seconds, entry through exit, one skill.
- 60fps minimum, 120fps if the phone has it, because these get slowed down and drawn on.
- Film with the subject centered, and keep them centered through the skill.
- Use a tripod when applicable.
Which clips to make.
The tag sheet is the menu, so work down it and decide how much coverage each skill actually needs.
- Foundational clips can be filmed with or without features.
- Where a feature is relevant, include it in the skill, so jumping is filmed on a jump.
- Cover the versions of a feature too, small, medium and large jumps, or a tabletop against a gap jump.
| Skill type | Examples | What to film |
|---|---|---|
| Terrain-neutral | Bunny hop, manual, wheelie, track stand | One clean version, with or without a feature. |
| Feature-dependent | Jumping, dropping, wall ride, pump | Film on the feature, and cover its versions (tabletop, gap, small to large). |
| Terrain-sensitive | Cornering, braking, descending, line choice | Film the surfaces that change the technique. |
Title and tag every clip.
Tags are how a clip is found, so an untagged clip is invisible.
The title convention
Level and clip role are not tags, so the title carries them.
- Skill first, then the phase or the named fault, then role and level.
- The Media grid truncates titles near 25 characters, so the first 25 have to identify the clip.
- Cornering | Initiation | Model L2
- Cornering | Late vision | Error L2
- Braking | Front loaded | Error L1
The tag stamp
- Tags are detected automatically as the clip uploads, so adjust them by hand and write the description in your own words.
- Actions, the one thing the clip is about, so a cornering clip is Corner or Berm or Flat corner, not all three.
- Terrain, the grade and the surface you filmed on.
- Camera angle on every clip (Front, Side, Behind, POV), plus Weather when it applies.
Add levels as you go
A suggestion that pays off: mark each clip with a level.
- L1 foundation, L2 developing, L3 advanced, L4 expert.
- The level is the standard the clip is for, not the ability of the rider in it.
Send a reference, not a correction.
You have built a compounding resource you can pull from at any time, anywhere in the app.
Compare against it
- Open an athlete's clip beside the reference and play them together, with overlays and angles when the difference needs numbers.
- The comparison does the explaining, and the athlete sees the gap instead of reading about it.
Send it anywhere
- Attach the reference in chat, or use Add Reference in the feedback composer.
- Every time you would type a correction you have typed before, send the reference instead.
Find it fast
- Filter the Media grid by tag and the right clip is in your hand in seconds, mid-conversation.
- Skill-first titles keep the grid readable, so the first 25 characters tell you which clip you are holding.
A library that compounds.
Every clip you add works for every athlete you coach after it.
The compounding effect
- Every explanation you catch yourself typing twice is the next clip, filmed once and reused for years.
- Terrain variants only where the surface changes the technique (cornering, braking, climbing, descending), because a bunny hop on green and one on black is the same clip twice.
Coach beyond the session
- Show a movement without riding it that day, because the model clip is the demonstration.
- Coach athletes between visits and at a distance, since the standard travels without you.
- That is the base of a remote offer, athletes keep progressing between sessions without adding hours to your week.
Keep it clean
- One role per clip, and if a clip needs two, it is two clips.
- If a value you need is missing, request it through Want something new? rather than inventing one in the title.
Coaching a team sport? The volleyball version of this guide covers filming a whole library in one gym session, and tagging against the skill names your sport already uses.
Team Sports Version
sportsense.
Reference clips work the same in every sport sportsense supports. The tag values change, the workflow does not.