Dr. House Rules for Product Problem-Solving

Mart Objartel
The Startup
Published in
4 min readJan 21, 2021

Dr House is a misanthropic, ingenious, and unsociable diagnostician in a fictional TV show. He ignores rules and clashes with his colleagues, comes up with controversial hypotheses, and saves his patients' lives more often than not. If you have not seen the series yet, you have roughly 124 hours worth of episodes to binge-watch.

It’s difficult to imagine a more toxic workplace than what Dr House has created for his team. His miserable ass would be fired from any normal organization; it's even difficult to imagine how he would get hired. Building a culture of trust and sustainability are not his forte. But he gets results (at least in this fictional universe), so maybe there is something to learn...

Everybody lies

This is a reoccurring theme in the series. Patients lie and screw up the diagnosis. It’s easier to break into their homes, search trough their belongings and read their private diaries to get necessary supporting data.

The same principle works in UX testing.

“Don’t listen to what people say; watch what they do.”

― Steven D. Levitt, Think Like a Freak

People want to be nice, and doing problem discovery interviews is a science to get it right. Don't ask them what they want, or would they buy your product. This info is mostly useless. Better ask them when was the last time they did… If possible, observe their behaviour, try to get them to really make a buying decision, to take out their wallet. You can always give their money back and say sorry. But you have learned a more valuable lesson about your idea.

Brainstorm but expect that most of the ideas are idiotic

https://house.fandom.com/wiki/No_More_Mr._Nice_Guy

House’s problem-solving method is whiteboard brainstorming where everyone shouts out ideas that House dismisses (and sometimes fires for obviously stupid ideas) until there is one that fits.

“You’re an idiot!”

- Dr House

It’s not nice to call you colleagues idiots when you have an ideation or design thinking workshop. But you should, at some point, focus and discard the ideas that most probably do not work. There is never enough time to work on all crazy ideas, pick a few.

Good ideas are just hypotheses and wrong treatment kills

When you irradiate your patient for cancer, but they have an infection, you kill them. Or if you cut out the wrong part of their brain, it's also not good. Mostly wrong treatment creates new, bigger and now time-critical problems such as cardiac arrest, insult etc.

“We need a crash cart here!” https://www.tvfanatic.com/gallery/chase-house/

This is where Dr House lacks a bit. They don't have a good process of how and when to validate ideas. They usually resolve it by pitching the idea to Dr Cuddy, Dr House’s boss, who then tells them to prove the hypothesis before going on with an unconventional solution.

Product management analogy would be to ask your department head whether you can start building or do you need to validate your ideas. It would be best if you validated without asking. You should also test all your ideas and hypothesis, and it would be best if you had a process in place so that this would be a part of the way of working. Solving a wrong problem or right problem in a wrong way can kill your product if you have a short runway or waste everyone's time and money. If your manager knew all the answers, she would not need you.

There are no rules on how to experiment, the only thing that matters are the results

Experiments are what they sound like. Experiments. There are standard diagnostic tests such as MRI, bloodwork, etc., but sometimes it's not enough, and you have to go creative and experiment. Give your patient a growth hormone that makes cancer grow faster, to prove that its cancer, so you can find it and remove the problem. Induce a seizure, if that is what is missing piece in the puzzle.

It's similar in product development. We have a set of tools, AB testing, fake door experiments etc. but sometimes that’s not enough. Any experiment that will prove or decrease the risk of building a wrong thing will help. Build a prototype from paper and glue, remove a feature to learn if that induces reaction, fake AI with the back office, do things that don't scale. Do whatever it takes to learn and reduce the risk of investing in the wrong direction. Results are what matter.

--

--

Mart Objartel
The Startup

Product manager at Klausapp.com. Hands-on experience working in scaled agile (teams-of-teams) and startup environments.