AI insights
How 86’d’s AI suggestions use your restaurant data—and where they stop
Exactly which records Ask 86’d reads, how permissions limit it, and why every suggestion needs a manager’s judgment.
By 86operations · Published · 6 min read
AI is useful in a restaurant when it starts from your own records and shows its work. It’s risky when it sounds confident about things it can’t see.
What Ask 86’d reads
- Daily sales for the last 8 weeks (only for users with financial access).
- Open tasks, recent manager logs and handoffs, open maintenance.
- Guest issues and incidents from the last 30 days.
- Events in the next 14 days.
- Waste totals for the last 30 days and products below par.
- Top product-mix items for the last 30 days (financial access only).
- Your team roster and your Restaurant Builder profile and targets.
What limits it
- Permissions: anything a user can’t see in the app is excluded and listed as not visible.
- Missing data: if you haven’t recorded labor or inventory, it says so rather than guessing.
- Owner control: AI features can be turned off.
- Training: 86’d doesn’t use your data to train AI models.
How to read an answer
- Facts: what the records show, with the dates and source.
- Possible causes: hypotheses to check—not confirmed explanations.
- Suggested actions: concrete next steps.
- Follow-up: what to compare next period.
Hypothetical example
“Why was labor high last week?” Facts: labor % was above target on Thursday and Friday (Daily Sales, Sep 22–28). Possible causes: the log mentions a training shift and a late private event. Suggested actions: confirm the training hours were planned, and review the event staffing on the BEO. Follow-up: compare the same nights next week.
Illustrative only—not a customer result.
AI suggestions don’t guarantee savings or results. Check the cited records before acting.