Start free
Help others

Practice DAX by helping someone.

Browse open debug issues posted by the community. Fork into your sandbox, fix the measure, propose your fix back. The OP gets notified. Real data never in the room.

Browse open issues → What is this?
01 — Browse

Pick an issue that catches your eye.

Every open issue has structured context: failing measure, expected vs actual, the synthetic data model, optional screenshots. Filter by concept — userelationship, time intelligence, filter context — or by what looks most interesting to solve.

  • Concept-tagged for fast discovery
  • Sort by oldest unanswered or most recent
  • Free-text search across title, symptom, measure
SALES MODEL · 3D AGO
USERELATIONSHIP returning blank on fiscal calendar
Blank for every month in FY24. Other years correct. SUMX over Sales measuring …
USERELATIONSHIP AND INACTIVE PATHS
02 — Fork

Your own sandbox in one click.

Forking clones the synthetic model + the failing measure into your account. You can rewrite, run, test, and iterate in your own workspace without touching the OP’s session.

  • Fork without leaving the issue page
  • Full access to inspector + side-by-side compare
  • Daxie’s diagnose-first mode active by default
YOUR FORK
-- Active path was wrong
CALCULATE(
  [Total Sales],
  USERELATIONSHIP('Sales'[ShipDate],
                  'Date'[Date])
)
03 — Propose your fix

The OP gets notified. The loop closes.

Once you’ve got a fix working, click “Propose as fix”. Your fork becomes publicly viewable and appears on the original issue’s page as a proposed solution. The OP gets a notification in their DAX Solver inbox — no Reddit-monitoring required.

  • OP sees diagnosis + fixed measure side-by-side
  • OP clicks “Accept” to merge the fix into their workspace
  • Your fix lives as a permanent reference on the issue page
✓ ACCEPTED PROPOSED BY YOU
Active relationship was on OrderDate; switched to ShipDate
The OP accepted your fix.
FAQ

Common questions.

Do I need an account just to look?
No. Browsing open issues is fully anonymous — you can read the problem, the measure, the schema, and the diagnosis transcript without signing in. You only need an account to fork the issue into your own workspace or to propose a fix back to the OP.
Is helping just charity, or do I get something?
Practice, mostly. Every issue is a real DAX puzzle someone got stuck on — fixing them sharpens the kind of judgment you can’t train on toy problems. Reputation/credit features are on the roadmap once enough fixes are flowing.
What if the OP’s measure references real data I shouldn’t see?
They can’t — public shares are restricted to .pbit templates, which carry schema only. The synthetic data you query against is generated by us; it respects the OP’s relationships but contains none of their real numbers.
Can I report an issue that looks off?
Yes. Every share has a report button — off-topic, spam, sensitive data, inappropriate, or other. Reports go to our triage queue.

Open the feed. Pick a problem.

The fastest way to get better at DAX is to debug someone else’s code.

Open the feed → Or post your own issue