Impact Mapping questions the business value of your Epics. It forces you to understand where you have assumptions and gaps in your subject matter expertise that must be filled to avoid building an irrelevant or needlessly limited product. And takes your product squad one step further into aligning their frames of reference, which is critical to avoiding subsequent analysis paralysis.
The way product roadmaps are usually specified often create market-fit and frame problems.
- Product owners and stakeholders may regard creative intuition as more useful than product analytics. They may be right about that too – customers seldom know what they want until they see a good design for it.
- On the other hand, business cases that don’t optimise product-market fit analysis generate products that don’t fit the market. Building the wrong thing, no matter how excellent the quality and productivity of delivery, will result in negative ROI. Maintaining the wrong thing will reduce the ROROI too. Relying on intuition may also give rise to the All or Nothing Anti-pattern.
- Many analysts do not think “Breadth First” (may need explanation/not sure), and thereby generate requirements that constitute frame problems, half solutions and local maxima in the landscape of business value. That can be avoided with a decision making method called Whole Board Thinking. It isn’t commonly taught to analysts and therefore most people don’t test their frame of reference frequently enough to prevent these problems.
Impact Mapping is a comprehensive way of clearing out these problems.
- Convert each existing business case / Business Requirements Document (BRD) / Lean Canvas / back of the napkin into Epic Normal Form.
- For each epic, quantify the assumptions that link the Why to the Who, the Who to the How, and the How to the What. If you can’t figure out how to attach a metric to an assumption, find a subject matter expert who can figure out what numbers are underpinning your intuition.
- Starting with the Why to Who assumption, test these assumptions against real analytic data that describes real market behaviors. If you don’t have that data, build a spike and gather some, or find a subject matter expert you trust to provide it. If you don’t have budget just Google for some. Remember, you’re testing your assumptions so you don’t waste a lot more budget building the wrong things.
- Winnow assumptions as you go. If the link from Why to Who is broken, no amount of Who to How or How to What is going to get you market share. This way you avoid analysis paralysis by not analyzing things for no good reason.
- Where your assumptions simply don’t hold up, or where one is numerically inferior to a competing assumption about which Who must change its behavior to satisfy your Why, winnow the epic landscape so you don’t waste further analysis and delivery bandwidth when you could be producing delighters instead.
- Map the epic landscape of all your product lines and your competitors’ products too. If you don’t know who your competitors are or why their products are winning market share, find out. If you have overlap in the epics of different products, figure out how much ROROI you’d gain by fixing it. If the numbers hold up, put a fix in your release map.
- Determine critical numbers for each quantified assumption. Use those as acceptance criteria when you get down to “Behavior Driven Analysis” (may need explanation/not sure).