Skip to content

PODB 5 ​

Generating Options ​

Environment Scanning ​

  • Political, regulatory change.
  • Economic change.
  • Social change.
  • Technological change.

Analogies ​

  • Lessons from adjacent industries.
  • Lessons from industries with similar cost and revenue.
  • Best practices in your industry.
  • Reverse benchmarking.
  • etc...

Creativity Stimulators ​

  • Perfect World (around a specific need)
  • Opposites (looks whats's there and not)
  • Bouncing (between people)
  • Forced Comparison (to any object)
  • What if (a most unusual possibility works)
  • What else (could the product look like)
  • Different Chair (for each team member)
  • Dialetics (thesis, anti-thesis, synthesis)

Business Theories ​

  • Customer needs
  • Unmet needs
  • Disruption theory
  • Jobs to be done
  • Customer journeys
  • Integrative Thinking
  • Core Capabilities
  • Etc...

Generating Options ​

You can enhance your ideas by looking beyond the pure product.

Ten Types of Innovation:

  • Profit Model
  • Network
  • Structure
  • Process
  • Product Performance
  • Product System
  • Service
  • Channel Brand
  • Customer Engagement

Be sure to think through a handful of relevant ideas by describing then as a set of choices.

Main Strategy Choices:

  • Where to play?
    • products or services
    • customer segments
    • channels
  • How to win?
    • value proposition
    • defensible advantages
  • What capabilities ?
    • Knowledge
    • Tech
    • People
    • Customers
    • key partners
    • access to funding

Evaluating Options ​

Because the future is chaotic, startups cannot plan ahead: they have to go through a three stage process and continuously refine their choices.

Three stages for launching a startup:

null

For any business to work three key things must be true. These things need to be tested.

AttractivenessCustomers valueCapabilities
Market potential is largeCustomers like our product or serviceWe offer lower costs or more value than competitors
Market is structurally profitableChannels like our product or serviceCompetitors will not retaliate

Stage 1 ​

"Quick and Dirty" tests to start with.

Choice structuring for problem solution fit

reqsattractivenessCustomers valueCapabilities
What needs to be true?The market size is large, and its structure is attractive, There is a commercial model that appears financially attractive.The problem is real, potential customers have shown initial interestThe concept is technically feasible
How to test this?Gather existing data to estimate the size of the potential market, check for regulatory and other showstoppersUndertake quick interviews with friends and fam to get a appeal measureCreate sketches, flow charts and wire frames of the idea and discuss with experts and friends, google competitors

Examples:

  • driver tree: rough estimate of market.
  • five forces model: check if market is structurally profitable.
  • benefit ladder (product attributes => functional benefits => emotional benefits): helps filter interviews.

Stage 2 ​

Choice structuring for problem market fit

reqsattractivenessCustomers valueCapabilities
What needs to be true?Customers profitability looks positive, a conservative financial model shows the venture to be commercially viableEarly users "Alpha Testers" responded positivelyThe functional prototype works, competitor response is limited
How to test this ?Build a basic financial model from a profit driver tree, with assumptions ready for testing, sense check the projections with others to ensure that the opportunity is still attractiveconduct small real world experiments to test key assumptions using digital tools ad prototypes (are customers giving time, money etc. or just because it is free?)Demonstrate the MVP to experts/friends to get feedback, Wargame or think through competitor response

Examples

  • HiFi Prototype

Stage 3 ​

When things get clearer you can switch "engineering mode" and continuously improve by measuring and analyzing key metrics.

Choice structuring for pilot scale up

reqsattractivenessCustomers valueCapabilities
What needs to be true?Metric targets can be achieved, the financial model still appears atCustomers "beta testers" are repeatedly willing to spend money on the solution, customers have recommended the solution to othersAll capabilities needed to scale and win remain available source-able
How to test this ?Check the weekly metrics to determine areas of underperformance (and take remedial action)Run targeted trial and A/B tests to continously improve sales and service elementsGather feedback for fixes and additional features and use these to make an improved version available for testing