BEHIND THE BENCH / A PRACTICAL GUIDE
Friction Finder
Turn a recurring work frustration into a starting point for a useful conversation.
When it helps
You know what keeps going wrong, but not what tool or service to ask for.
What to bring
One recent, non-confidential example: who needed what, where work slowed down, and what happened next.
What happens
Choose a few symptoms, add where they occur and their impact, then describe one incident. Find a starting point sends those answers to AI, which proposes possible patterns and questions worth exploring.
What you leave with
A working hypothesis, a possible next step and discussion questions. Copy the starting point for your own notes. Sending it with your context to Brandon is a separate, optional human-review form.
Illustrative result excerpt · fictional situation
A short example
Input. Fictional example: Every Friday, two people copy job updates between spreadsheets. Last week an old delivery date was copied and the team had to check it again.
Result. Possible pattern: repeat-work loop. Worth exploring: follow one update from its source to the final spreadsheet and identify the first repeated check.
Example only. AI wording can vary; this is not an interactive result.
Why it works this way
A recent example is easier to examine than a broad complaint. The three short steps connect what happens, who is affected and what better would look like before suggesting a next conversation.
Your judgment stays in the loop
Does the hypothesis fit the actual incident? Correct the context before treating a suggested pattern as the cause. You decide whether to request human review.
Current practical limits
This is not a diagnosis, estimate or approved project. If live AI fails, a clearly labeled local pattern preview may appear. Drafts are saved in the current tab; Start over clears them. No system access is granted by using the tool.