Product Manager for Success Academy's public website and family/teacher/internal mobile apps. Greenfield work, hybrid in Manhattan. Five rounds: recruiter, hiring manager, three-back-to-back onsite panel, final with the CTO. I built a prototype to ground my thinking before the onsite. Did not advance. The "feedback" was boilerplate.
No formal case study. Each round was conversational.
None of the rounds were a take-home. Five conversations across multiple stakeholders without a deliverable to anchor on is the kind of process that drifts toward "did we click?" rather than "is the work good?" I wasn't asked to make anything. I built a prototype anyway, mostly so my own thinking had a place to live.
Then I used it. At the onsite, at the end of each of the three back-to-back conversations, I pulled it up and walked the interviewer through the relevant section so they could see how I think about a problem space they were testing me on without giving me a take-home. Three different angles for three different roles in the room. Same artifact.
Lives at middleton.io/prototypes/sa/. It walks through what a refreshed family app and public web experience could look like, what data flows through Salesforce today, and where I'd start a roadmap.
Success Academy is mission-driven and culture-forward. I leaned into the consumer-facing PM experience that maps cleanly: HVAC.com (about two million annual visits), Rocket Lawyer (5% mobile conversion lift on the highest-revenue product line, 11.5M users via co-branded partner sites), Lever HRIS Sync as proof of greenfield platform work. What drew me to this role was proximity to the user. The value wouldn't be theoretical or pulled from a dashboard. I could measure outcomes AND hear about them from real people. That's the part of the role that interested me most. Families navigating their kids' education and teachers trying to do their jobs well are specific, real users.
The artifact is the live prototype. Embedded below for a quick look. Open in a new tab to actually click through it.
Each round felt good in the moment. Onsite at 120 Wall Street was a strong day. The final with the CTO was 30 minutes that ran clean. Eight days later, the boilerplate landed.
I asked for the call.
The reply the next day:
"Move forward with another candidate who more closely aligns with our current needs" is the same sentence twice, with the second one written to look like a real reply. It isn't. It's the boilerplate restated. After five rounds, including a final with the CTO, that's the part that doesn't sit right.
I'm not owed a reason. But the explicit invitation in the first email ("if you would like to discuss this further, please do not hesitate to reach out, and we can schedule a call") and the actual response when I took them up on it are different things. The mismatch tells you what kind of feedback culture you're looking at, which is itself useful information for a candidate considering whether the role would have been a fit in the first place.
For what it's worth: by round four I had questions about the team structure and the security-and-privacy listing on the JD that I didn't fully resolve. Five rounds for a single PM hire, where the team was newly formed and the scope was still being defined, was probably a signal too. I'd take this conversation again. I'd ask harder questions earlier next time.
Share the prototype with the hiring manager before the onsite. I waited until the onsite to walk people through it. The hiring manager round was a week before. If I'd shown him the prototype then, framed as "here's how I'd think about the first 90 days in this role," the onsite panel would have walked in already primed by him, and I'd have had at least one round of "here's how I'd refine this." Surfacing it later cost me a round of compounding signal.
Ask about the team and timeline earlier. The team was newly formed. The scope was still being defined. The "Security & Privacy" department listing on the JD was unresolved by my last round. Asking those questions in round two instead of waiting for round three would have given me data to use, not data to wonder about.
Push for a real debrief, not just a call request. "Hey can we have a quick call?" is too easy to deflect. "I'd value 15 minutes to understand what would have made my candidacy stronger, specifically on [the thing I'm worried I missed]" is harder to brush off. If they're going to give boilerplate, at least make them write it specifically.
The artifact is the live prototype.