The Scenario
You are three months into the Product Manager role at Outgo, DAT's fintech product unit serving independent carriers across North America. Outgo offers invoicing, factoring, and a banking product. It operates as a startup — your team moves fast, decisions are yours to make — but you have a significant structural advantage: DAT's carrier network, which processed 400 million freight postings last year and holds $150 billion in annual shipment transaction data.
Your first 90 days have surfaced three things:
- Invoicing adoption among newly onboarded carriers drops sharply after the first invoice. Carriers get set up, submit one invoice, and then go quiet. Your ops team calls it 'one-and-done' and it's been a known problem for eight months with no clear owner.
- The factoring product has a meaningful approval rate gap — roughly 30% of factoring applications from smaller carriers are declined by the underwriting model, and a significant portion of those declines are for carriers who are, by all observable signals from DAT's freight data, creditworthy and active. The risk team believes the model is conservative by design. You're not sure that's the right call.
- A competitor fintech targeting carriers has started running ads directly against Outgo's factoring product, positioning themselves as the 'carrier-first' option with faster approvals and no hidden fees. Your GTM team flagged this two weeks ago. No one has responded yet.
Your VP of Product has asked you to come to next week's quarterly planning meeting with a clear recommendation: which one of these problems should Outgo prioritize first, and what should we actually do about it? She has also asked you to be explicit about what you are not doing and why — the team has a history of trying to run three initiatives at once and landing none of them.
You have access to the following data:
- Invoicing activation data: first-invoice-to-second-invoice conversion rates by carrier segment, cohort, and onboarding channel
- Factoring application data: approval rates, decline reason codes, and carrier freight activity signals from the DAT network (load volume, on-time delivery rate, lane consistency)
- Support ticket data: top 10 support categories for invoicing and factoring by volume and resolution time
- Competitive intelligence: the competitor's public pricing page and three carrier forum threads discussing Outgo vs. the competitor
You do not have access to a full data analyst. You have two hours of engineering time available before the meeting. You are the only PM on Outgo right now.
YOUR TASK — THREE DELIVERABLES
Deliverable 1 — Outgo Growth Brief
Produce a short, decision-ready brief (no longer than two pages) that your VP of Product can read in five minutes and use to run the quarterly planning conversation. It must include:
- Your prioritization decision: which problem you are recommending Outgo address first, and a clear rationale for why this one and not the others
- Your proposed intervention: what specifically you would ship or change, what metric moves as a result, and how you would know it worked within 60 days
- Your attribution approach: how you would isolate the impact of your initiative from other variables so the team can evaluate whether it actually worked
- What you are explicitly not doing: one or two sentences on each deprioritised problem, explaining why it is not the right first move — not that it doesn't matter, but why it is not the right first move now
Deliverable 2 — README (Sections A, B, and C)
Section A — Discovery & Prioritization Rationale
- Walk through how you would use the available data sources to make your prioritization decision. Which data matters most? What would you look for in each source? What is the key question each data source answers for you?
- Identify the most important assumption in your recommendation — the thing that, if wrong, would change your answer. How would you test it?
Section B — Stakeholder & Execution Plan
- Name the internal stakeholders whose buy-in you need before this initiative can move (engineering, risk/underwriting, ops, GTM — be specific about what you need from each). How do you get alignment with two hours of engineering time available?
- What is the single highest-risk thing that could go wrong in the first 30 days of execution? How would you get ahead of it?
Section C — AI Usage Log
- List each AI tool you used during this challenge.
- For each tool: what did you ask it to do, what did it produce, and what did you change, reject, or redirect before including it in your submission?
- If you used no AI tools, state that explicitly.
Deliverable 3 — Video Walkthrough (8–10 minutes)
Record your walkthrough as an MP4 or MOV file and upload it directly on the Provn platform as a separate file.
Cover these four points in your video:
- Walk us through your prioritization decision. What was the moment you committed to this choice over the others — what data or reasoning closed it for you?
- Explain your attribution approach out loud. How would you actually measure whether your initiative worked, and how would you handle the 'but maybe it was something else' objection from your VP?
- Tell us how you would handle the competitive threat in the background — not as a separate initiative, but in the context of whatever you're already doing. What's your posture on it?
- Mandatory AI question: Walk me through one moment in this challenge where you redirected the AI. What did you ask it to do, what did it give you, what was wrong or incomplete about that, and what did you do next?
CONSTRAINTS
- You have two hours of engineering time available before the meeting. Your brief and plan must work within this constraint — proposals that require significant eng scoping before the meeting are not actionable.
- Outgo's underwriting model is maintained by the risk team, not engineering. You cannot change the model directly. Any initiative that touches underwriting approval rates must go through risk team alignment first.
- DAT's carrier network data is available to you as signal, but it is not yet integrated into Outgo's product infrastructure. Using it in your initiative requires a data pipeline that does not exist yet — scoping that pipeline is part of what you're working with.
- The brief must be decision-ready for a non-technical VP. It cannot contain unresolved ambiguity, options without a recommendation, or metric definitions that require explanation.
- You are the only PM. There is no one to delegate the research or writing to. Your process choices — including how you use AI — are visible in your submission.
