Tuesday, January 15, 2013

Product Management: What do I do to be a PM of my own idea? (Part 3)

This post is the continuation of my earlier post about how to start out as a product manager.  If you did not read the first two, I recommend you start from the first two articles which are available here and here.

I started out by laying out 7 steps to become a product manager for your own product idea.  Here were 7 steps:

1. Pick an expensive problem to solve in your life.
2. Come up with a product idea that you can use yourself.
3. Write down how you are solving the problem without the product today.
4. Spell out how you want to interact with your product.
5. Fake the product to your potential customer.
6. Remove an interaction one at a time until you cannot remove any more without becoming useless.
7. Write down your finding.

We were discussing Step #5 earlier.

5. Fake the product to your potential customer.

Ok.  You have written down your product idea, and programmed yourself to be the product.  Now you are ready to sell it to your potential customer (it could be your friend, parents, brothers, sisters, or whomever).

I'm assuming that you've already done some homework that this potential customer that you are about to talk to has similar problem that you are set out to solve in Step #1.  If not, you are probably asking a wrong person.  Do some educated guessing and talk to people who are likely to have the problem that you are solving.

When you are ready to speak to the person, see if you can get a permission to record the conversation.  This makes it little more formal, and also give you a conversation transcript that you can go back to at later time.  It is important because how she reacts to your problem statement and simulation are important indicator to how close you are to solving a relevant problem for the right users (in other words, whether your product has a market or not).

First, state what problem you are trying to solve.  Pay attention to how she's reacting.  Did you have to explain to her twice?  Did she get it immediately?  Was she interested?

Second, if she shows sufficient interest, then see if you can simulate the product for her.  Ask her the questions and explain to her what the product will do for her.  In other words, act out what the product will do so that she can understand what information she has to provide and steps it takes to solving her problem.  Make notes of kind of questions she is asking.  Take notes of her body language.  These are all going to be incredibly useful feedback to you to simplify your end user experience and focus on the problem.

6. Remove an interaction one at a time until you cannot remove any more without becoming useless.

Now that you've spoken to a potential customer or two, you have some idea as to what your potential customers are thinking of.  I'm assuming that you were able to find a problem that you can get someone else excited and wanting to see it become a real product. (It is not an easy task to identify such a problem, by the way.)

Congratulation.  You now have version 0.1 of a product.  In order to make it into version 1.0 and launch the product, you have to now wear a pragmatist hat and figure out how to do this simpler yet solve enough of the problem so that users will still use.  You do this by taking things out from the interaction.

Remember that you listed out the steps of how the product will interact with users?  Now go back to the list and see if you can remove each line and still solve the problem.  You may not be able to solve the same problem by removing each step, but you will still be able to solve close enough problem.

What you are doing here is coming up with what product managers call minimum viable product.  The core product idea that provides 80% of the value without 20% of the extra.  Your goal is to figure out whether there is a way to simplify your instruction so that you can make your product cheaper and faster to build.

The main driver is this.  Although you've talked to a couple of potential customers, you have no idea whether your product will be useful in the real world.  You don't know what kind of competitors there are (many customers may choose not to use your product but stick to what they have been using earlier), you don't know how market will change until you release version 1.0 of your product, and the most importantly you don't have money to invest, wait and see if your idea will actually generate revenue.  All of these unknowns add risk to your business.

So what you are doing is to get a simple product out to the market fast, and see if it attracts buyers.  If it doesn't, you haven't lost nearly as much as building the entire product.  And you still have a chance to tweak your product and launch it to solve a slightly different problem.

The more you can simplify and get the essence of product, the smaller the risk is with launching your business.  This practice is called Lean Startup.

Going back to our example of content discovery product, we had several interactions:

  • Ask for registration, if first time user.
  • Upon registration, write down the user name and password on the notepad.
  • Ask for login, if already registered.
  • When user logs on, match the user name and password and if correct, display the next screen.  If not, display the login page with an error message.
  • Ask user to connect with Twitter account.
  • When user connects Twitter account, do the think that other websites do to get access to the user's Twitter account.
  • Now the product is in action.
    • Go fetch the Twitter accounts that the user follows.
    • Find out the most influential authors (by seeing its Klout scores) out of those who the user follows.
    • Get the yesterday tweets with links from these authors.
    • Display the articles referenced by the links to the user.

What's really necessary out of these?  Is personalized discovery what I'm after?  If I take out the personalization component, user does not have to create an account and log in.  This removes Step #1, #2, #3, #4, #5 and #6.  By removing the personalization component, I have reduced the more than the half of the steps.

Then how would I get the interesting content for the user?  How about focusing on the users who share the similar interest as me, such as social media, startup and leadership?  In fact, by focusing on those people who share similar interest as I do, I don't need to automatically curate the content.  I can manually curate those content that piqued my interest, and share them to like-minded audience.  In that scenario, I am literally being the engine of the product.  (Much like the guy sitting inside the ATM box!)

I can start doing it immediately.  I don't need any code or automation.  I can start finding interesting articles and share them.  In fact, that's how I ended up simplifying my product idea and why I've been doing one interesting blogger a week project.

7. Write down your finding.

If you've run through all the steps that I've just outlined, then bravo to you.  You have now completed what a product manager goes through to launch a new product.

Make sure you document each step and publish your finding on your own blog.  Once it's available in the public domain, you can list this experience on your resume as your own product launch.

Has this been helpful?  Any question on any of steps?

If you found this useful, please follow me on Twitter: @jaeho9kim.

UPDATE (1/15/2013): The part 1 and part 2 are online.  Please help me improve by providing comment.  Thank you!


  1. I absolutely love this post. I've followed it since part 1 and I'm learning some great new stuff. Thanks for always updating this.

    1. Thank you for following them, Pippa. Feel free to share with others, and if you have other help you need with launching PM career or getting better at it, I would love to know.