Tuesday, March 29, 2011

Lean Problem Interviewing

Lean Coffee was hosted today by mad scientists in white lab coats (ie. the Big Bang Technology crew).

We talked about the Problem Interview and various tips & techniques, such as:

  • Keep a "perimeter" -- don't go too far outside the topic
  • Who/Where/What/When/Why questions (from journalism)
  • Why validate?
    • Are there any customers?
    • Assumption: Your plan A won't work
    • No validation = no repeatable sales process
  • Beware of confirmation bias: Tendency to seek the answers you want to hear
  • Problem interview vs. Solution interview (very different)
  • Remember: Can pivot around the business model, solve different problems
  • Do you need a script? Helps keep things on track - vs. map or outline
    • But more open-ended discussion lets you discover a lot more
  • Customers are like kids -- easy to lead them, and have them agree with things they don't actually want
  • Should be a conversation, not a formal interview
  • They'll tell you the symptoms -- you need to determine the problem
  • Keep doing the interviews until you know what the answers are going to be
  • Problem: Looking for what's "broken"
  • Important: What people say vs. what they do is often very different
  • Methodology doesn't prevent you from still missing the mark on a "must have" problem
  • Failed products lead to the next one -- always learning
  • Answers change in different contexts -- anonymously vs in a group (observer effect)
  • Maximize the comfort level to get honest answers
  • Warm them up: Ask about themselves, make environment safe
  • Technique: Parrot back responses
    • Lets them confirm, helps them feel more comfortable
  • Discover people's motives
Another great session!

Next week I'm going to be in Orlando on business (I know, poor me) so hopefully someone else can take a few notes.

Tuesday, March 22, 2011

Problems Worth Solving

Today at Lean Coffee Toronto, we discussed the topic of "Problems Worth Solving".

We focused primarily on Customer Development and validation with some great discussion points:

  • Relatively few products ever see "the light of day"
  • Very few businesses (10%?) make it past the 2-year mark
  • There's often a difference between end user and the person who's going to pay, ie. the customer
  • Value chain: Need to talk to everyone, but be careful about focusing only on one piece at a time
  • Developing a product inside a service business
    • Risk of being introverted when you develop in-house
    • Risk of entrenching vision and ability to pivot
    • Often need to take a step back and talk to people outside
  • Customer development
    • Validate your problem first
    • Then validate your sales process and ecosystem
  • Is it acceptable to sell a product that isn't 100% finished yet?
    • Yes. Find out if the idea sells. If yes, go to next step.
  • Lean: Is there a kisk of killing bad ideas?
    • Yes, but very small compared to the risk of building something nobody cares about
  • Most devs are not good salespeople naturally
    • Need to develop your sales skill set
    • Session idea: Lean sales?
  • Danger of "sales people" do customer validation -- might get overly optimistic early results
  • Using Google ads, etc. to test hypothesis / validate customers
    • A few people had tried this
  • Kickstarter is essentially the Lean model (but need a US bank account)
Thanks to everyone for their participation, and to Mark at BNOTIONS for hosting and moderating.

Thursday, March 17, 2011

Lean Metrics

Chris Eben hosted this morning's Thursday session of Lean Coffee Toronto at The Working Group.

We chatted about Metrics, including:
  • What are metrics? Defining the things that are most important to your business
  • Don't base decisions on metrics until you know they're the right metrics
  • Start at revenue and work backwards to find out the actions that lead to revenue
  • Be careful about which numbers you're basing decisions on:
    • Does your sales funnel indicate whether your pricing is wrong?
    • ...or if you're trying to sell to the wrong people?
  • Try to see where the bottleneck is -- eg: What are they key indicators of dropoff?
  • Don't make decisions on the "marketing/vanity" metrics
  • Another thing aggregate metrics often don't tell you:
    • One user might be 10x more valuable than another
    • How do you know which ones?
  • "Stop listening to" and "start watching" your customers
  • Product-market fit: 40% of your users would care if you disappeared
  • Can't apply metrics from one customer segment to another
    • ie. early adopters vs. majority
  • Track at least one key metric in each part of the AARRR model
  • Practical example:
    • If you notice that conversions suck after 1 month, could be "honeymoon period"
    • Trim eval period back to 10 days and improve conversions
  • One of the best tools: Another set of eyes, second opinion
  • OKCupid demographics data set overviews -- generally useless but entertaining
  • Need to change your key metrics you have as you move into new business phases

We also spoke briefly about my recent software project management survey, how survey design is difficult, and how we could spend an entire session talking about surveys. I'd be happy to plan and moderate such a session, but would need a location.

Thursday, March 10, 2011

Inside The Lean Startup

This evening I attended a great panel discussion event at the MaRS Discovery District entitled "Inside The Lean Startup".

There were some great points brought up by the panel of @ashmaurya, @skanwar, @leilaboujnane, and @davidcrow of StartupNorth etc.

My takeaways:
  • What do people misunderstand about Lean?
    • not a step-by-step guide -- it's a set of philosophies
    • not "cheap" or "bootstrapping" -- process efficiency
  • Potential mistakes:
    • Relying on the wrong set of tactics for the stage they're at
    • Over complication in systems (ie. continuous integration)
  • Should be about changing behaviour -- getting out and talking to customers
  • All about efficiency -- get the most for the least
  • Key Metric: Money -- Are people paying you for your product?
  • Key Question: How disappointed would you be if this product would be taken away tomorrow? 
    • 40%+ indicates not a fluke
  • Key Points:
    • Reduce assumptions
    • Identify riskiest parts first and evaluate
    • Don't afraid to be embarased
    • Find the shortest route to get in front of customer and get feedback
  • Lean applies to: unknown problems / unknown solutions
  • Hypothesis test: pull the plug and see if anyone cares!
  • When do you start charging? From day 1. ("free" is a customer acquisition tactic, not a business model)
  • Starting out: Is this a problem worth solving?
  • Make sure a hypothesis is falsifiable -- the scientific method
  • "Life is too short to build products nobody cares about." -- Ash
  • Adding features:
    • Unused features are waste
    • Creates technical debt
    • Start with "no"
  • How do you get feedback?
    • Initial hypothesis: pain is so great that they want to be involved in the process
    • Find your early adopters -- as visionary as you are (rare breed, hard to find)
    • People are always open to telling you about their problems ("tell me about your pain")
    • If you're not getting feedback, find out why -- not reachable? don't like forms, method of communication, type of question
And some miscellaneous notes:
Valuable stuff. Looking forward to more events like this one!

Tuesday, March 8, 2011

Lean Marketing

I attended the brand shiny new Tuesday session of Lean Coffee Toronto today at the awesome BNOTIONS headquarters at Yonge and Bloor.

My summary follows:
  • Thursday's session of "How to spend $5,000" apparently wasn't the right approach to the discussion
    • (good thing we got a second try :)
  • Different parts: Researching vs. Positioning vs. Building Demand
  • Can you describe the product in a few short sentences? (ie. the length of a Google Ad)
  • Very important to have consistent messaging
  • Creating the UVP (unique value proposition) is the hardest part.
  • Technique: "It's like X for Y"
    • Example: It's like Dropbox for Development Environments
    • Pros: You can use all of the marketing efforts from X for free!
    • Cons: Helps describe the What, but not necessarily the Why
  • 6 steps: (iterative)
    • Problem
    • Values
    • Market
    • Ideal consumer
    • Competition
    • Positioning (USP)
  • Technique: 3rd-party re-explanation:
    • Explain to someone, then have that person explain it to someone else new
  • Tip: Don't try to deliver too many different concepts in your message
  • How far along do you need to get to get proper feedback? Idea? Mockups? Prototype?
    • Answer: All of the above. Start with an idea and move up from there as you get validation
  • Pivoting: Do you change the market or the product? (or both?)

And some Meetup-related stuff:
  • Next meetup: Same topic both days again
  • Perhaps some attempt to:
    • Post reading material prior to meetup (and have people actually read it)
    • Keep the discussion on topic via the moderator (a difficult task, to be sure)
    • Look at the San Fran Group for topic ideas and literature
  • Check out http://dooo.sh/it/ from Big Bang (blog post)
Please comment on this post or email any corrections, additions or updates.

Until next time!