# Dmitri Okafor, Senior Backend Engineer at Fintelo (210 people) — read of unified-email-api, June 5 2026

> 8 years writing backend services, currently on the payments team at a Series B fintech, been kicking the tires on indie SaaS ideas for about 14 months. Two failed side projects behind me.

## How I got here

I was searching "unified email API sendgrid mailchimp alternative" because at work we're duct-taping together SendGrid for transactional and Klaviyo for marketing and I keep re-writing suppression logic. I wanted to know if someone already solved this. Google surfaced this page, probably from a blog or link somewhere. I clicked assuming it was an actual product I could evaluate.

## What I clicked first

The hero got me. "One API. All your email infrastructure." is exactly the problem statement I typed into Google. The spec table is clean: Campaign API, Event Tracking, List Management, Unified Billing. I thought I was reading a product landing page for something I could sign up for.

Then I hit the scoring panel and got confused. "64/100 Adoptability" and "The Wishdeal Factory scores every idea against 10 Adoptability axes." I had to re-read the whole page. This isn't a product. This is a packaged pitch for building one.

## Where I paused

The Fermi math. "$-15,576 Year-1 take-home" sitting right there in the open, presented as a feature. That stopped me cold. Most idea marketplaces would bury or round that number or skip it entirely. The fact that it's front and center, negative, and called a "Fermi estimate" is either genuinely unusual or a very deliberate trust move to get me to overlook it. I sat on that for a minute. I still don't know which one it is.

## What I distrusted

"We shipped the strategy package; you ship the customer conversations." That line is doing a lot of work. It's graceful phrasing for "we have no proof this works, that's your problem now." The "Honest disclosure: we don't have live customers on this idea yet" is at least direct, but it comes right after credibility scores of 9/10. Credibility of what? The idea? Credibility is supposed to be about customer trust in a real product. Scoring an unbuilt product 9/10 on credibility felt like the scoring model is grading the idea quality, not market evidence.

Also "uniqueness: 9/10" on a unified email API. I've seen Nylas, Courier, Loops, Postmark, Resend. The space is not wide open. That score made me distrust the scoring rubric.

## What would convince me

A single operator case study. Not hypothetical. Someone who bought the $99 dossier, actually built the thing, and got to even 10 paying customers. Revenue numbers optional, but a named person or company and a timeline. Right now the page asks me to take a $99 bet on their judgment with zero evidence that their judgment has worked before. One example flips that.

Also: what does the working code starter actually look like? Does it scaffold a real multi-provider abstraction or is it a README and a Stripe webhook template? That distinction matters enormously to a backend engineer.

## What I'd ask in an email reply

1. The $-15,576 year-1 estimate: is that net of the $99 purchase, and what assumptions drive the revenue side? Because a unified billing API targeting developers could have wildly different TAM depending on who you're selling to.

2. "Implementation upsell: 9/10" is the top-scored axis. That implies the play is consulting or custom work, not SaaS subscription. Is the real model here a productized-service business, not a pure API product?

3. Has any adopter of ANY idea in the Wishdeal catalog gotten to meaningful revenue? Not this specific idea, any of them. I want to know if this studio has a track record before I evaluate the individual idea.

## Verdict: on-the-fence

The transparency is genuinely unusual and I respect it. The negative Fermi, the no-customers disclosure, the honest scoring with visible concerns listed. But I landed here looking for a product to use, not an idea to build, and the page never resolved that confusion cleanly. I'd read a reply email if the founder could point me to one real operator story.

---
*Memo by skeptic persona, generated 2026-06-05. Studio breaks own self-grading loop.*