Planning doesnt have to suck!

Geeks hate to plan. I must admit, I do not always like doing it either. We just want to “do real work” which means designing and coding. Planning also got a bad rap, deserved I believe, from the 6 to 9-month requirement and design phases of waterfall. Unfortunately for us geeks it is planning that enables us to

• Improve time to market.
• Increase value realization.
• Become more predictable.
• Deliver higher quality.
• Reduce stress and avoid emergencies.

So how do we plan in a way that does not suck? My claim is to do it all the time, do it thoroughly and do it quickly. Sorry, Army story follows!!

Many of you have heard the military phrase of “no plan survives first contact.” This might leave many to ask, “why plan then?”. It is this idea that drove the Army to create a planning process that makes the Army the most agile organization I have ever experienced.

Consider one of the simplest operations to plan and execute. A company sized operation would typically include 15-18 teams with 10 soldiers each (How many of your projects include a team this size?). Now add in the amount of change that happens during an operation. Often, we would have at least 2-3 changes occur within an hour. Each of those changes requires all those squads to understand the change and its impact on them. Each of those impacts must be understood and communicated quickly, often within minutes to ensure success.

This means leaders need to be good at planning and planning fast! This means everyone must know what it takes to create a plan and continually look for ways to improve it. It also means it must become a way of thinking and operating.

Below are my top considerations for what makes a good technology plan. If you are not doing one of these, think about how it could fit into your process.

1. Know what success means and prioritize. Understand how the business benefits from the work you are doing and do the work that provides the biggest benefit first! Don’t work on anything you can’t tie to a business benefit remember 66% of all features are rarely to never used!
2. Know what the current state is. What is the technology stack in use, how does the business use that technology, what is working for them and what is not. Without this understanding, you will not know where your plan will take you and how it will impact that success we defined above. I am often surprised to how little most organizations understand their entire process, most only understand their single step and putting it together can truly drive the most critical changes.
3. Create clear targeted features. Ensure that each feature is clearly articulated and can stand on its own and delivers an impact against your success criteria. This will provide the team with clear, concise requirements and provide the business with faster benefits. This includes decomposing the business requests and architecting your solution to enable incremental delivery.
4. Set Team Expectations. Know what you and your team are doing for the next two weeks. Are you working on top priority items and do you know what it takes to complete them? Setting expectations is the fastest way to reduce stress and avoid emergencies in all work!
5. Create a roadmap. Use story points, hours, dog points, or whatever method you want to estimate work. Remember you only need enough details to decide how big it will be, and the decisions needed to implement it. Your estimate is wrong, but its still closer than no estimate! Do not underestimate how good you and your team are at estimating work! Try planning for the next 6 weeks and identify the milestones, dependencies and decisions required in that time. This will give you and your business great context to make all decisions on priority, scope, and timeline!

With practice, a good team spends 8-10 hours every two weeks on this. That time will return significant benefits as you improve time to market, increase value realization, become more predictable, deliver higher quality, and reduce team stress and emergencies!