Today I’d like to introduce you to a podcast that is derived directly from an analysis performed by the platform I’ve developed - called Venture Proof. Shortly, I’ll be publishing a demo overview of the platform using this exact case, so it’s worth listening to even if you aren’t a professional in healthcare, or HealthTech / Medtech. I could have easily produced this — or helped you produce a 100%validated study — for whatever industry you happen to work in.

Innovation Unpacked is for people who are truly interested in making innovation more predictable. You can support me simply by subscribing for free, and sharing this with your colleagues.

The following is the initial problem framing I started with. I elaborated the problem statement further based on the data generated using the prompts at the end of this post:

EHR Semantic Interoperability — Siloed vendor systems cause data loss, redundant tests, dangerous medication errors when patients transfer.

Solution Hypothesis: Vendor-agnostic architectures achieving true semantic (not just file) interoperability

There are a couple caveats to this analysis:

Human-in-the-Loop

There is extensive Human-in-the-Loop (HITL) built into this workflow. Since this is not a client study, I opted to accelerate through some of them. There are a number of things right up front that I defaulted:

* Research: While the system performs deep research to capture facts and assumptions, users can also upload their own data. Alternately, they can perform deep research on a topic and upload that as well. I performed deep research using a prompting system and will show you the prompt at the end.

* Current Costs and Theoretical Minimums: The system auto-generates these costs based on the research and some LLM inquires. The calculations are all performed deterministically using Python. However, the user has full autonomy to add, edit, or delete any of these inputs if they have data that conflicts with it, or expands it. I just went with the defaults.

* Initial Friction Validation: This is the part the replaces bias-prone JTBD interviews. More on that at another time. There are several ways this system can accommodate this decision-gate:

* You can use the interview guide it generates (designed to validate / invalidate friction) and interview (and record) several job executors. 6-8, 8-10, or whatever you feel comfortable with; or as your budget allows. You can upload the transcripts to be evaluated

* You can take the interview guide and perform deep research designed to source observable facts that support the friction hypothesis. The prompt is included in the system

* You can also generate a comprehensive playbook for this gate that shows you exactly what data you need to capture, and where to get it. Who to interview and what ask them. And what you should attempt to observe and what that process looks like.

You can upload the unstructured results for one of these or all of these. Venture Proof don’t care!

* Survey: This section is under development but gives you a lot of options. In fact, this step is 100% optional now. Most the options are much shorter than an Outcome-Driven Innovation survey this platform doesn’t waste time and money exploring for a problem. It has already found the problem, quantified and mapped the friction (inefficiency gap) to the job map. A survey — if needed — is designed to validate friction at the metric level. That might only be 12-15 rating points. There is no segmentation needed.

* Minimum Viable Prototype (MVPr): This section has a much more extensive playbook generator that guides you through a comprehensive Wizard-of-Oz experiment. Once again, the system will accept whatever data you develop from this, in whatever format. This step is critical before going to your investment committee for funding the factory. I skipped this step 😜and you should be aware of that.

What this Podcast is Derived From

There are a lot of outputs from this platform to support you when you have to defend your investment request. One of them is a 25k word textual report — which no one in their right mind would read (except you Joe!). This is why I use NotebookLM and a custom prompt to generate a podcast (highly flattering to me, of course!) that tells the entire story. It has a beginning, middle, and end.

The other stuff — like external customer question defense, internal stakeholder question defenses, private equity question defense, and venture capital question defenses, will help you sell a fully-validated research package.

All I did was feed the report into NotebookLM. 🤷‍♂️

The 30 Year-Old Incumbents

You will never get an analysis like this from:

* Switch Interviews

* Other general JTBD sprints

* or even Outcome-Driven Innovation

No offense, but none of them are designed for delivering an outcome — the ultimate investment decision outcome. They all require more work to be done. This gets it all done for you. Well, with a little HITL assistance to make everyone feel warm and fuzzy.

No transfer of wealth needed.

Here are the prompts I promised; nothing glamorous.

Deep Research Prompt Generator

Create a system prompt I can use for deep research on [industry or topic]. It needs to collect hard numbers (observable facts), assumptions in the industry (educated guesses), and hunches that are floating around (wild-assed guesses or bias). Include cost basis for all hardware/software resources, labor, licensing, etc. required to get the job done. This must include sizing estimates for TAM and SAM and also projected CAGR%. Do not use graphics in the research output, only tables. If user enters nothing, prompt them to enter an industry, concept, or topic.

Interview Guide Deep Research

%% The goal of this prompt is to attempt to replace interviews with Job executors to find and validate facts that answer the questions and probes %%

**Role & Objective** You are an expert industry analyst and technical researcher. Your objective is to conduct deep, fact-based research based on the attached qualitative interview guide.

The provided guide contains structured questions designed to uncover operational friction points, bottlenecks, and technical challenges within a specific industry. Your task is to transition these questions from qualitative inquiry to empirical, evidence-based research. For each question and its associated follow-ups, you must find grounded, factual answers, industry benchmarks, and technological realities that explain _why_ these friction points exist and _how_ the industry currently addresses them.

**Instructions for Analysis** Please process the attached interview guide and output a comprehensive research report following these exact steps for **each** of the questions (Q1, Q2, etc.):

**1. Core Constraint Identification:** > Distill the main “Question” and “Goal” into the fundamental constraint at play. Is the friction caused by physics/chemistry, technological limitations (e.g., sensor latency), or organizational/human factors?

**2. Empirical Baselines & Benchmarks:** > Answer the main question and follow-ups using current industry data, scientific literature, or recognized engineering standards. For example, if a question asks “how long does it typically take,” provide the documented industry average or range (e.g., “Industry benchmarks indicate X to Y days”).

**3. Root Cause of Friction:** > Based on factual research, explain exactly _why_ this step carries the designated “Friction Level.” What are the documented points of failure?

**4. State-of-the-Art Interventions:** > Identify the current best-in-class technologies, methodologies, or software solutions that the industry is using to solve or mitigate this specific friction point. Separate established, proven solutions from emerging/hyped technologies.

**Output Formatting Requirements** Structure your report logically. Use the following format for each question analyzed:

- **### [Question Number]: [Brief Topic Summary]**

- **The Empirical Reality:** (A factual, data-driven answer to the core question).

- **Addressing the Follow-ups:** (Direct, researched answers to the specific sub-questions).

- **Industry Benchmarks:** (Hard numbers, timelines, or success/failure rates).

- **Current Technological Solutions:** (What the market currently offers to solve this).

**Strict Constraints:**

- Do not hallucinate data. If specific benchmarks or timelines are highly variable or undocumented in public literature, explicitly state: “Extensive variability prevents a standard benchmark; however, case studies show...”

- Ground your research in reality. Avoid marketing fluff from vendors; focus on physics-based realities, independent white papers, and operational case studies.

**Input Data** Here is the interview guide to analyze (or it’s attached):

Are you interested in innovation, or do your prefer to look busy and just call it innovation. I like to work with people who are serious about the subject and are willing to challenge the current paradigm. Is that you? (my availability is limited)Book an appointment: https://pjtbd.com/book-mike

Email me: [email protected]

Call me: +1 678-824-2789

Join the community: https://pjtbd.com/join

Follow me on 𝕏: https://x.com/mikeboysen

Articles - jtbd.one - De-Risk Your Next Big Idea

Always attack…Never defend

This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit www.jtbd.one/subscribe