Three shifts · Two review layers · One taxonomy — yours
How we work.
We run that layer instead. Our analysts work inside your software, follow your taxonomy and deliver against your quality bar. The output is yours. The operation is ours.
How a game gets tagged
- 01
Tracker tags
An analyst watches the fixture live or on video and tags every event against your taxonomy, frame by frame.
- 02
Moderator reviews
A moderator sitting above the trackers watches the same feed, catches missed events and corrects classification in-shift.
- 03
QC samples
QC pulls a structured sample against your rubric, scores it, and feeds the result back into the analyst's next assignment.
- 04
Client receives
The file lands in your platform in your format — period-by-period for live fixtures, inside the agreed window for post-game.
The operation
Coverage runs on three rotating eight-hour shifts out of Nashik. A fixed core of admins, moderators, QC and support holds the structure; the trained analyst pool flexes against the fixture list.
Every tracker has a moderator above them and a QC layer above that. Nothing reaches a client having been looked at once.
Where it usually breaks
Coverage fails at shift handover, not mid-shift. A fixture that starts at 21:40 IST is tagged by one shift and finishes under another. We overlap the boundary by thirty minutes, hand over per-fixture rather than per-desk, and the outgoing moderator stays on any game already in progress until it ends.
24-hour rotation · Nashik · IST
- Shift 0106:00–14:00 ISTEuropean daytime fixtures
- Shift 0214:00–22:00 ISTEuropean evening fixtures
- Shift 0322:00–06:00 ISTNorth American evening fixtures
Questions buyers actually ask.
- How do you handle security and NDAs?
- NDA-first. We sign before any footage or credentials change hands, and we work inside your environment rather than exporting data to ours. Analysts are assigned to a single client's workflow, access is scoped per project, and footage is never retained beyond the engagement.
- How much timezone overlap do we actually get?
- Nashik runs IST (UTC+5:30). Our three shifts cover the full 24 hours, so European daytime and evening fixtures are handled live, and North American evening games fall inside our night shift and are processed overnight — usually waiting for you at the start of your business day.
- What happens if an analyst doesn't show up?
- The pool exists for exactly this. Assignment is per-fixture, not per-person, so an absence is reallocated by the shift moderator before the fixture starts rather than discovered afterwards. The fixed core — admins, moderators, QC — is staffed independently of the analyst pool so the supervisory layer never thins out.
- Can you scale for a playoff push or a tournament window?
- Yes, and it is the normal case rather than an exception. Surges are absorbed by widening allocation from the trained pool rather than by hiring, so the constraint is scheduling rather than a recruitment cycle. Give us the fixture list and we will tell you honestly what we can hold.
- What formats do you deliver in?
- Whatever your platform already reads. We work inside SportsCode, Catapult, Hudl and proprietary tools, and deliver to your existing schema. There is no migration and no intermediate format of ours to learn.
- What happens when you get something wrong?
- Errors are caught in three places: the moderator in-shift, the QC sample after, and you. Anything you flag is corrected by the shift on duty and fed back into that analyst's next assignment rather than logged and forgotten. Systematic errors trigger a taxonomy review with your team.
- What does a pilot involve?
- Your games, your platform, your scoring criteria, at no cost. We tag a batch, you score it against your own quality bar, and you decide. There is no commitment until you have seen the output.