I built this for myself first.

I'm Myke Mueller. I run Community Tap & Pizza in Fort Dodge, Iowa. I've been an operator longer than I've been a founder. Still am — the floor on Friday night, the books on Saturday morning, payroll on Sunday afternoon.

I started building Never 86'd because nobody was making the screen I actually wanted. Every restaurant tech vendor either sold me a dashboard I had to interpret myself, or charged me enterprise prices for software built for the office, not the line. The screen I wanted answered one question: what costs me money this week, and what's the name attached to it.

A single HTML file at 11pm

The first version was a single HTML file on my laptop. I was trying to figure out why food cost drifted four points one week and nobody could explain it. I wrote the math, fed in my own Z-reports, and it pointed at the right station within the hour.

Then it pointed at the next leak. Then the next. By the end of the month I had a tool I trusted more than any of the dashboards I'd been paying for. I kept it on my own restaurant for nine months, refined it, and then a chef I trust asked if I could run it on his 16-unit group.

The first number was wrong

I ran the model on his data and produced a recovery surface of $8.3M / year. It was on the screen. It came out of the math. It was internally consistent. It was also wrong.

The signal was a sales rollup that physically couldn't be real — the model was reading a doubled view of the network total and confidently extrapolating from it. I caught it the next day. The honest number, after I de-duplicated the rollup, came out at $1.81M — about 22% of what I'd reported.

Every other vendor in this category would have left the $8.3M number in the deck and never spoken of it again. I walked it back in writing to the design partner who'd already seen the original figure.

That's the moment I knew the discipline of correcting your own number down is the product.

It's why he stayed. It's why the next number we shipped — $15.72M reconciled across 545,677 orders — landed without anyone needing to verify it twice. The case is the public version of that walk-back.

Why source-tagging is the moat

Every figure that comes out of Never 86'd ships tagged. Verified means we can re-pull it from a primary source and defend it to the penny. Estimated means we've modeled it from a benchmark; we name the assumption next to the number. Unverified means the source isn't wired yet; the figure is illustrative.

No competitor in this category does this publicly. None disclose model error. We checked. The closest thing is one vendor quoting “15% labor forecast accuracy gains” — a brag stat, not a per-figure disclosure.

What I'm building now

Eight agents. Sales, labor, voids, 3P fees, tips, catering, rate-card audit, shift sentiment. Each one reads a slice and tells you the one thing to fix. Per store. Per name. Every figure source-tagged.

60-minute free trial. Drop a CSV, see the read, no card. Pricing from a solo operator (free) to enterprise (custom). Same source-tag discipline at every tier. The price scales; the rule doesn't.

The direct line

If something on the screen doesn't make sense, you can email me directly. I read everything. I respond personally. myke@n86.app.

If we're wrong about a number on your data, we'll walk it back in writing — same rule we ran on the first $8.3M.

— Myke Mueller · Operator · Fort Dodge, Iowa