Tired of Products Schedules Slips? Use These Two Practices
Deliver your products on time, every time, with high quality!. In the B2B world, this is the foundation for your success. Not only your own revenue stream is on the line, but your customer's revenue stream is dependent on it. My interpretation of this reality is that aggressive products execution schedules are nice to have but a predictable schedule is a must have. To that end, I have learned from Jeff VerHeul a great method.
I have met Jeff when he was the Senior Vice President of corporate engineering at SanDisk, one of the brightest and nicest people I have ever known (you can tell immediately that he is from the midwest and not from the valley...). While overseeing hundreds of engineers across three continents and tens of products programs, he has the remarkable ability to drill down in minutes to a specific bug and understand deeply the technical details, gaps in processes & methodology and infrastructural improvements that need to be implemented. And as nice person as he was he had no problem to hold a daily 7 am meeting 7 days a week for a program that went off tracks…and last anecdote about Jeff, he would hold you up in slide 22 because the data is inconsistent with the second bullet on slide 4. Amazing guy, a role model!
Quite early in his tenure, Jeff introduced a concept of two governing rules within his engineering organization. I would just call them “The VerHeul Rule”:
1. If you got to a point where you must push out a milestone, do it only once! Never-ever push out the same milestone twice!
2. The push-out is bounded to your current distance from the milestone. For example, if you are two weeks before the milestone, you can only move it by up two weeks.
The first one seems obvious but it’s not. Why? Because the people in the front lines (teams that interact directly with customers on the BU /R&D /Quality organizations) tend to adopt optimistic assessments of the situation or “hope strategies”, and make commitments to customers based on those. This is a slippery slope that leads to multiple customer disappointments. For example, you communicate to a customer a delay in delivery of one week, then another five days, then another two days… Letting down a customer is ugly. Be brave enough to do it only once.
The second one is a tricky one but think of it: if you are two weeks before a milestone and need to push out by a month, you’re either being dishonest with your peers or you just don't know what you are doing. In today’s development world, you have pretty good models of bug discovery rate, bugs closing rate, mountain charts tools and so on. Good engineering manager knows where she stands with the program.
As Jeff used to say, “If you are in a control posture of your project, you should be able to follow these rules.”
Establishing those rules between engineering and marketing turns out to be very useful as it brought stability and predictability to our products roadmap execution. The engineering team would do everything they could to deliver product milestones on schedule. But if things didn’t go as planned the business unit would not be caught in a surprise and be put in an extremely tough position to deal with disappointed customers.
Another positive outcome of those rules – it built a lot of trust between marketing (BU) and engineering. In development, things can go wrong and will go wrong at times. But once you put a governing policy that handles those situations, ground rules for all parties to follow, everyone will have the same level of expectations and much of unnecessary stress and friction can be prevented.
A good friend of mine once told me that tier 1 OEMs are like elephants: they are big, they move slowly and have a very long-term memory. If you didn't deliver they will remember it for a long time....I have found these two rules helpful to deal with the elephants.
Key takeaway: A milestone can be pushed out only once and not further than your current distance to it.