# Brendan Reyes, Fractional CFO — read of Bookkeeper AI, June 20, 2026

> 14 years in corporate finance, went fractional 3 years ago, currently managing close cycles for 6 SMB clients ranging from $4M to $22M. I drive 40 minutes each way to a coworking space because I refuse to work from my kitchen table with a 4-year-old asking me about dinosaurs.

## How I got here

Typed "automate monthly close report multiple clients" into Google during a Tuesday afternoon when I had four clients expecting deliverables by Friday. A listicle mentioned "AI bookkeeping tools" and this was one of three links I opened in new tabs. Landed here around 3pm, had maybe four minutes of real attention.

## What I clicked first

"You're writing 28 monthly reports before your Friday deadline." That landed. Not "streamline your workflow" or "save time with AI." Twenty-eight reports. That's a number I recognize. I kept reading.

Then I hit this: "This product page is being finished." Okay, so I've been pulled in by a headline for a product that doesn't exist yet. That's a specific kind of disappointment.

## Where I paused

The scoring breakdown. "pain intensity: 10/10. buyer clarity: 10/10. distribution ease: 10/10." Then: "financial upside: 1/10." I stopped there. What does financial upside mean in context? Is that telling me the builder's upside is low, or is this saying the product won't make me money? I read around it twice and still couldn't tell. It's in a section called "How honest is this idea, really?" which implies these are scores the builder is assigning to their OWN idea. This is a product idea catalog, not a product.

## What I distrusted

I did not understand for almost two full scrolls that I was on a marketplace selling the concept of building this thing, not buying access to an actual tool. The "FOR INDEPENDENT BOOKKEEPERS" callout at the top made me think I was a potential user. The $5 unlock, the $99 "adopt the build," the "Operator partnership Custom" tier -- that's a completely different purchase than "here's software that solves your Friday problem." The page flips audiences without telling you it's flipping.

Also: "$5,897 Year-1 take-home (Fermi)" and "1 in 4 Meaningful-success odds." These are disclosed as Fermi estimates, which I respect. But Year-1 take-home under $6K on a product requiring $10K investment is a loss. The page says that clearly. It also says the mid-case ARR is $82K and take-home is $5,897, which means there's about $76K in costs baked in somewhere that I'd have to go read a $5 dossier to understand. That's not a red flag necessarily but it made me suspicious about what I was actually looking at.

## What would convince me

If I were being sold the concept of building this: one story from someone who actually went through the build process, what they shipped, what happened in the first 90 days. Not an estimate. A real number, even a bad one.

If I were being sold the tool itself: let me run one report against one of my actual clients' QuickBooks exports. No form to fill out, no sales call scheduled. Just a demo that works on real messy data. I've got a client whose chart of accounts looks like someone sneezed on it. If the tool survives that, I'll care.

## What I'd ask in an email reply

1. You say "shippable in 4 to 6 weeks" -- is that with a developer I hire separately, or is that what the Operator Partnership includes? Because those are completely different commitments.
2. The financial upside score of 1/10 -- can you explain that scoring axis? Is that telling me the addressable market is small, or that this is a commodity play, or something else?
3. Has anyone who bought the dossier actually shipped this? If yes, can I talk to them for 20 minutes?

## Verdict: on-the-fence

The pain-point framing is the clearest I've seen for this category and I almost stayed. But I needed to get to the third scroll to understand I was reading a pitch for a business idea, not a product I can use this Friday, and by then I'd already done most of the work of convincing myself.

---
*Memo by skeptic persona, generated 2026-06-20. Studio breaks own self-grading loop.*