> ## Documentation Index > Fetch the complete documentation index at: https://evalgate.com/docs/llms.txt > Use this file to discover all available pages before exploring further. # Review the results > What the scores, costs, and failures actually mean. # Review the results After a run, the results page tells you what happened and why. This guide explains each column and what to do with it. ## The results table A screenshot of the results table with columns for case, status, score, cost, and latency. | Column | What it means | What to do | | -------------- | ----------------------------------------------- | ---------------------------------------------------------------------------- | | **Status** | Pass or fail for this case | Start with the failures | | **Score** | The scorer's output, normalized where supported | A low or withheld score means the scorer could not decide — inspect the case | | **Cost** | The model-call cost for this row, in USD | Spikes often mean a runaway prompt or tool loop | | **Latency** | Model-call wall time | Compare against your latency budget | | **Provenance** | Complete or incomplete | Incomplete provenance cannot be used as release evidence | ## Drill into a row Click any row to open the detail. You will see the exact input sent to the model, the raw output, the tool calls, the scorer's reasoning, and the model-call reference. If the case was promoted from a trace, the source trace is linked. ## Group failures Use the **Group by** control to cluster failures by scorer, failure category, or model. A cluster of the same failure usually means one root cause, not many. ## Send failures to review Select failing rows and click **Send to review** to create review tasks for your team. Approved review items can be promoted into golden cases that gate future releases. ## Up next Turn a real failure into coverage with the [Logs explorer](../traces/explore-logs).