We don’t need no stinkin’ test plan!
A weekend spent as a volunteer on a construction project has resonances with testing for Dave Whalen. Who always has a plan!
My wife’s boss bought a small two-bedroom house just west of Denver. He wanted to add to the house and recruited volunteers from his employees, their spouses and kids. A ‘weekend project’ I was told. So one gorgeous, cool, crisp, Saturday morning in July, we drove up to his house in the foothills to help out. Since I had some experience with construction in a prior life, I was made project foreman. We were in serious trouble!
It was a bit more than an addition. He had added a four-car Garage to the existing house and wanted to build a new house on top of it. So I asked the boss where his plans or blueprints were. He pulled out a spiral notebook and a window catalogue and began to draw his vision.
Soon a parade of trucks arrived and dropped off loads of construction materials and lumber. How he knew what to order and how much is beyond me. We began to build and raise walls with rough openings for the windows (which had not been ordered yet). As we broke for lunch another truck arrived carrying pre-made roof trusses. A few minutes later, a crane arrived. We finished raising the walls and had the crane operator hoist the trusses to the top of the structure. As we were nailing them in place, the boss looked out at the magnificent view of Denver below us and declared: “We need another floor!” Construction came to a halt, and we packed up with plans to return the next weekend to build another floor.
Next weekend the scenario replayed itself. We built the third floor using the same process, with only a drawing on a piece of notepaper to guide us. We put on the roof this time. Needless to say the project was a nightmare. Most of us had no idea what we were doing. We ran short of supplies, and had to stop to wait for them to be delivered or go get them ourselves.
We had too much of some items, which ended up being wasted. Some things were built incorrectly only to be torn down and rebuilt. When the code inspector arrived, we failed miserably (did I mention we didn’t have a building permit). We had to tear off the roof, order new trusses, and completely rebuild it. It had to have cost a fortune, even with volunteer labour!
As Dr Covey says in his ‘Seven Habits’: “Begin with the end in mind!” The same thing applies to software testing. Take the time to write a test plan, Identify the resources you need, the time you have, what tools you need, etc. It can be an exhausting mental process, but it’s a valuable one. It ultimately saves you time and money and the more you do it, the easier it becomes.
Sadly, most managers will take a wrong approach to test planning – they throw a bunch of people at it, typically in the last couple of weeks before release (which is then cut to a week because “development is a little behind schedule”) then figure it will all work out. Never has, never will!
Take the time to think it through and write a comprehensive test plan. I’m always asked: “Should I have it approved and signed?” I personally like to, I like to have all the group leaders sign it: project managers, development managers, business analysts, etc; basically, anyone with a stake in the outcome.
I view the test plan as my contract to the project team as to what we will test, how we will test it, how long it will take, how many people we will need, tools, etc. Will they read it? Probably not! But, when things don’t go according to plan, it’s nice to pull it out and point out the signatures. I always keep a signed copy handy. It should be a working document, continually updated as the project changes.
A word on templates – I like them, but they are just guidelines. Don’t just cut-and-paste your project name into an existing template. Use it as a guide to write your own project-specific plan. Personally, I’m a great collector of test plans. I save them and refer to them often, but always write one from scratch for each project. I may steal a little from this one, or a paragraph from that one, but at the end of the day, I have a unique, project-specific plan. A roadmap to the future; my vision of the end.
