# Dan Kowalski, Lead Estimator at Buckeye Precision CNC — read of Quote-AI, May 10 2026

> 14 years estimating at a 9-person job shop outside Dayton. We run 3-axis and 4-axis, mostly aerospace and ag equipment. I do about 25 RFQs a week out of my inbox and a shared drive folder that nobody cleans up.

## How I got here

Googled "job shop quoting software" on a Tuesday after I spent three hours on a quote for a 14-part assembly that came back "thanks we went with another vendor." A guy in the NTMA Ohio chapter Facebook group had mentioned this in a comment. I bookmarked it two weeks ago and finally got to it this morning during my commute buffer (I park a few blocks away and sit in the car for 15 minutes before walking in).

## What I clicked first

The subhead: "Stop losing Tuesday afternoon to quotes that never close." That hit weirdly specific. Tuesday is actually the day I dread most because that's when the weekend RFQs pile up from the inbox. So that caught me. I kept reading.

The hero stat "2 min Avg time to first draft" I basically ignored. Every tool says something like that and it never means what you think it means once you're inside the software.

## Where I paused

"94% First-draft acceptance rate."

I stopped there for a while. I genuinely do not know what that means. Acceptance by who? Does that mean the estimator accepts the draft and sends it out? Does it mean the customer signs it? Our shop sends maybe 22 quotes a week and wins 6 or 7 of them on a good week. If a customer is signing 94% of first drafts then either the jobs are so simple that anyone could quote them, or the prices are too low, or this metric is measuring something totally different from what I assumed. I would want a one-sentence clarification of that number before I moved past it and I did not find one on the page.

## What I distrusted

The single testimonial. Rick Holvenstot, Midwest Precision Parts, Toledo. I do not doubt Rick exists. But one quote from one guy in Ohio is doing a lot of work on this page. "We went from four hours on quotes every Tuesday to maybe 25 minutes" is a great line but I want to know what kind of work Rick's shop runs. If he's quoting simple turned parts from clean PDFs all day, that's a different universe from what I deal with -- incomplete drawings, verbal spec changes, "quote it in 304 but we might switch to 316 depending on price."

The NTMA survey stat ("8 to 15 percent margin on the table per job") is cited but not linked. I looked for a footnote or a source link. It's not there. That's the kind of number a marketing person writes and a founder approves without pushing back on because it sounds authoritative.

Also "Every feature exists because a shop owner asked for it by name." That's a classic founder-voice line that means nothing. Every SaaS says that.

## What would convince me

Show me a real RFQ -- a messy one, scanned drawing with a handwritten note on it, missing material callout on one of the parts -- and show me the output Quote-AI generates. Not a polished demo with clean inputs. The hard part of quoting isn't the math. I can do the math. The hard part is what you do when the input is garbage, which it is about 40% of the time. Show me how the tool handles that.

And I want to see two or three testimonials from shops that do work similar to mine. Not "job shop" generically. CNC, multi-op, aerospace or similar tolerance requirements. What does their miss rate look like on the AI-drafted quotes before they send?

## What I'd ask in an email reply

1. When the RFQ is missing a tolerance callout or the drawing is ambiguous, what does Quote-AI do -- flag it, make an assumption, or ask me before generating the draft?

2. Your "94% first-draft acceptance rate" -- can you tell me exactly how that's measured? Is that the estimator accepting the AI draft, or the customer signing the quote?

3. We use Shoptech E2. The page says you integrate with it but doesn't say much about what that actually covers. Does accepted quote create a job in E2 with routing, or does it just push a header record?

## Verdict: on-the-fence

The page actually communicates what the product does, which puts it ahead of most tools I've looked at. But the 94% stat unresolved and the lone testimonial leave me skeptical enough that I wouldn't reply today. I'd probably come back if someone in my network vouched for it specifically.

---
*Memo by skeptic persona, generated 2026-05-10. Studio breaks own self-grading loop.*