Simplifying includes breaking concepts down (user friendly means more than a smiley face on screen) with analysis. It works by re-assembling concepts into understandable high quality summaries (models) with synthesis. More than ONE synthesis will be needed. One is for people who will read and follow the complexity. One is for people who won’t be able to pretend they are following unless the words are put into polygon shapes. “I’m visual” is the refrain, but more than four or five polygon shapes quickly reveal that the shapes have “words” anyway, and confusion reigns among the “visual”.
For productivity the experienced analyst understands that working in text can be faster than working in diagrams (at least for analysts that like working in text). An ideal situation is one where a technical writer engages the stakeholders in diagramming simple text while the analyst produces complex text.
Many diagrams are just fancy vague lists. Perfect examples are Visio and PowerPoint “flat” diagrams). As a productivity technique, drawings are easily outpaced by anyone who can create or follow an indented outline.
Here is a sample text “analysis” using the puzzle from the last blog (Everything We Needed to Know We Learned on Sesame Street):
- Business Requirements
- Business Goals / Objectives
- We want to increase sales
- We want happy customers
- We want to get rid of old technology because…
- Business Needs
- Capability Gaps
- Solution Approaches
- We want to buy this in ONE software package
- We want to outsource all software configuration and maintenance
- Solution Scope
- Business Case (what matters, how it matters and why it matters)
- Business Goals / Objectives
- Requirements [STATED]
- Every want is stated only, not modeled, verified or validated.
- All are vague – unspecific – need improvement
- Requirements [ANALYZED / MODELED]
- No want has been analyzed / modeled
- Even business objectives are unclear –
- Business Analysis Plan
- Solution Requirements
- We want users to have the freedom to override business rules
- We want the system to pick the best approach
- We want to capture name and address and contact info everywhere
- We don’t want managers interfering with our work (micro-management)
- We want to reduce data entry errors
- We want direct access to the database
- We want users to write their own reports and not wait for IT
- We want easier, better scheduling with fewer disruptions
- We want large monitors, aluminum cases and wireless keyboards on our PCs
- Non-Functional / Qualities
- We want easy to use
- We want a consistent high quality customer experience
- We want reliability, maintainability, scalability and no irritability
- We will know it when we see it but no sooner
- Transition Requirements
- We must have everything built before release
- We don’t want to change the work or conditions
We may not agree on whether the above is “correct”. I think we CAN agree that it is not useful, complete or free of conflict. There are unspoken relationships (e.g., relation of “solution requirements” to goals / objectives). There are potential conflicts right next to each other. One example is when stakeholders want direct access to the database vs. fewer data errors.
What can fix this? SYNTHESIS! Are you a business analyst only, or a business synthesist* as well? Creativity isn’t just gums bumping – talk is cheap, watcha got that a project team could follow, that could make sense to those who cannot code ambiguity, confusion or wishful thinking? Most importantly – which statements deserve more focus, and which less? Which statements represent the highest requirements risks? What is missing for effective objectives? What would your process model begin to suggest?
Can any of my readers write a single paragraph describing a workable approach? Can anyone resolve the apparent conflicts by addressing high-level business issues? What do the stakeholders mean and can it be built to the joy of all? Hint: Start by synthesizing “bottom feeding” requirements into “top priority” business needs related to goals and objectives.
And remember – you can’t skip any steps in BA world for true enterprise systems. Pay it now or fail it later. Think of all the BA work (yes, its hard, that’s why stakeholders don’t do it) as leading to a quality groomed backlog. Good process, followed by good interface design followed by quality development is unbeatable.
When we “do what stakeholders want” we are not serving their goals. Not unless they happen to be information systems experts. You might as well try to build a skyscraper for a toddler – “I want the 50th floor to be ALL ice cream” they scream.
Remember to say yes, and to place ice cream fountains everywhere on 50. Make it the centerpiece of the project, so no one will forget which stakeholders insisted. Allow all the stakeholders to pick flavors for prioritization. Above all synthesize the other 99 floors with FAST elevators to 50.
Mo’ later, thanks for reading, don’t get lost in the trees OR the forest.
* Do I have to spell it out for you :) ?
Don't forget to leave your comments below.