In 2007 I was hired as Director of Engineering by PowerWise Group; a startup manufacturer of energy conservation equipment.
PowerWise had some very impressive technology, but the investors were becoming unhappy after the engineering department had missed several critical deadlines. They replaced the management team, and the new CEO asked me to address the problems, based upon my reputation dealing with similar situations. I am based in Los Angeles and was not willing to move to Dallas, but I agreed to rent a room at an extended stay hotel and stay on until engineering was back on track and I had hired and trained my replacement.
When I analysed the reasons why they had missed so many deadlines, I found that the previous management had implemented policies and bought software that was well suited for managing hundreds of workers in multiple cities, but a huge oveerkill for a startup with three engineers and one technician. All of this collapsed when the management was replaced, leaving everyone to try to figure out what to do next. Compounding the problem was the fact that the engineers had mostly big-company experience and didn't know how startups work. The good news was that the technology was sound, the inventor who held the patents was glad to offer technical assistance, and - most importantly - everyone knew that they had to find a better way of doing things.
I decided that a low-tech approach was best for this situation. I got a big box full of 8 inch by 6 inch Super Sticky Post-it Notes, and took over a conference room as our new war room. Then I sat down with the entire crew and together we worked out a plan and posted it on the walls. A typical note would say in large print something like "modify 100A prototype for lower-cost 20A model" or "translate all menus into French and Spanish." We went through the entire project, covering the walls with everything that needed to be done. As I expected, this identified multiple tasks that nobody had realized needed to be done. The last Post-it Note said "convert Post-It system to computerized system" with a series of sub-tasks that would accomplish that.
Once we had a plan, we went back and assigned tasks to team members. I did not try to mandate the time needed for a task; that was for the person doing the work to do. I also put an end to the "Death March" work schedule; working seven days a week isn't the solution. I made it clear that these were promises, not estimates, and that the only punishment for being late is that every morning you had to put a big fluorescent note above your task saying "X DAYS LATE." Next I had everyone answer the following question for every task: Define "done" - no "coding 90% done" fake milestones. Then we decided who would verify that it was actually done and sign off that they personally verified it.
Finally, I forbid any investors asking any of the workers any questions outside of a new weekly status meeting I set up. Every week we would gather in the war room and show them a bunch of tasks signed of "done on time", "done X days early", "done X days late" and they could see what was left. Then I would send the workers back to their tasks and go through what we had spent that week and what expenses I anticipated coming up. By the second month we were hitting nearly every goal on time and under budget, and were able to make up any unanticipated delays.
The project ended when a foreign investor bought out our group of investors and the patent-holder and moved the project out of the country. I left on good terms, having accomplished everything I set out to accomplish.