During my years as the president of Kendall College, it was a source of considerable pride that I continued my unblemished QSR record of eating bad food as cheaply and quickly as possible and knowing next to nothing about cooking.  I could barely boil water.  Never mind that I was trying to resuscitate and ultimately save this 75-year-old culinary school, which offered a great faculty and training along with a consistent inability to figure out how to pay its bills. Before I got there, it's fair to say that they threw almost everything at the problem, including the kitchen sink.

The only way I made peace with my chef-instructors was to assure them that I would concentrate on fixing the finances, growing enrollment and strategic partnerships, and building them the best technical platform for culinary education, but I would never try to tell them how to season the soup. There's a lot to be said for staying in your own lane.

In any case, it was a viable, if uneasy peace. They knew the place was a mess and that they needed help - they just didn't know if the "new guy" knew anything. They were tired and dizzy from the constant and frantic flipping from one new idea to the next without really giving anything much of a chance to actually take root and grow.

Reinventing Kendall was a long, painful process, but eventually even the most skeptical instructors came around because it's hard to argue with success. In this case, successful innovation was about iteration and successive approximation -- getting a little bit better all the time. In fairness, they ultimately changed their attitudes and their willingness to be part of the process rather than the problem. More importantly, they started to see that our carefully-focused and meticulously-measured actions were beginning to produce quick wins and concrete changes. A clearly articulated vision, a straightforward and realistic path (no "miracles happen here" jumps), and a patient and progressive plan to get there began to make it look like a more secure future was possible.

And, in the process, I also learned a few valuable tricks of the trade that helped to make things better, as well as a great deal about what didn't work and needed to be avoided. And, to be clear, the things that didn't work for Kendall wouldn't work whether your business had been around for decades or you were building a brand-new business.

One of the earliest lessons is that the "al dente" method of throwing stuff against the wall and seeing what sticks doesn't actually work for spaghetti or for startups. Kendall's leaders had tried a little bit of everything and had managed to do almost nothing because they were spread a mile wide and an inch deep. They were trying to do things quickly or cheaply that they shouldn't have tried to do at all. Good strategy is always the same-- it's deciding what not to do.

Marketers used to call this shotgun method of chasing a bunch of rabbits at the same time the "spray and pray" approach. I still see it, especially in the area of new CPG introductions, where the idea seems to be that sheer volume and variety -- tossing dozens of different products into the market as rapidly as possible -- is deemed the key to successful innovation rather than adopting a more targeted and limited approach based on consumer research and testing, followed by launching a few strong contenders and then constantly iterating from there. There's a lot less frenzy and maybe even a little less fun, but it's much more likely to lead to near-perfect pasta. More isn't necessarily better -- only better is better -- and less is often more.

It comes down to a simple question and then developing an approach that leads to the right answer. The question is this: if you consistently hit three home runs (winning new products) out of 10 fairly costly attempts, would you rather achieve the same three wins with only five better-crafted and thought-out product launches? And if the answer is obviously "Yes", then how do you create an innovation process and strategy to assure those far more cost-effective results?

The first stage of the process is to reach a broad consensus on what constitutes success and a commitment to develop and hold fast to certain metrics so that everyone is objectively clear on what a "win" looks like. And so that: (a) the goal posts don't move and (b) you're prepared to kill off the projects that aren't hitting the marks without debate and without delay.

The biggest problem in business is permitting lousy projects to persist to avoid making hard choices and hurting people's feelings. It's a little easier in a startup because this is often an existential issue-- if you're not making it, you won't be around long enough to worry about it. This is why I always say that there are no skid marks when a startup shuts down.

You can pick and choose your own criteria for success, but in general terms I always look for half the story to be internal requirements and the other half to be customer/market driven. So, for example, if I were picking six, I'd split the pie something like this for a typical consumer product:

1. Ultimately meets the company's financial requirements

2. Achieves sufficient and agreed-upon market penetration (% All Commodity Volume)

3. Generates profits across the distribution chain and channels

1. Pre-demand for product exists based on customers' awareness

2. Meets customers' expectations/desires

3. Fulfills customers' requirements/needs in a distinct way

The second stage is to develop and flesh out the funnel and any necessary filters. Here again, rigorous adherence to the rules and requirements of each gate and benchmark in the funnel is critical to keep weak offerings from slipping through. You need to avoid diluting the entire process by having too many offerings moving from the earliest evaluation phase on to the investment, development and action phases.

1. Prune the Pile
You need to spend the necessary time and resources right from the start to sort, evaluate, scrap and settle on a few key ideas. Test a bunch at the beginning because early testing is a cheap way to insure that you don't over-invest down the line. Take your time. Then start cutting, but don't narrow the field too soon or too quickly - we make better decisions when we're presented with multiple alternatives. Imagine being at the bottom of a diamond-shaped path. You want to go wide for a while so you're thinking broadly and outside of the box and then tighten the selection set down to the final most viable ideas. Focus on the problem to be solved rather than the particular solution being advanced. And don't let anyone (even the boss) push their own agendas or favorites too far.

2. Expand the Evaluation Group
The wider the pool of people and opinions that you can bring to bear on the problems, possibilities and products, and whom you can engage in the process, the more likely that you will reach better and smarter conclusions. Getting out of the echo chamber starts with the first phase of consumer research, but the same idea extends to the need to engage a broad cross section of the people in your own organization to make sure that the initial ideas make sense. You're going to want to pull lab and tech people into the conversation to address issues of feasibility and practicality - can we do it? You'll need finance people to look at whether it's worth doing across several dimensions - can we afford it? And finally, now that the project is getting some legs and a better definition, you want to go back to the well and ask the market and your prospective customers whether it still makes sense to them - will the dogs eat the dog food? And this is also the time to start thinking about time - schedules, milestones, timetables and realistic launch dates. Plenty of products fail because no one bothered to look at the calendar and the appropriate buying cycles.

3. Make a Serious Commitment to the Rollout
Once you decide to go, there's really no room for half measures. Thinking small is a self-fulfilling prophecy. This is why it's so important to have a few deep dives rather than trying a dozen different deals without sufficient energy, marketing or other essential resources behind each one. And you've got to have all the necessary follow-ups and follow-throughs as well because getting the ball rolling isn't enough - you've got to get it over the goal line.

I've seen dozens of failed launches over the years and excuses a plenty. Worse yet, too many people try to declare victory too soon or celebrate based on metrics that just don't tell the whole story. It's not enough to ship; the product has to sell through and this is why metrics like % ACV (All Commodity Volume) are crucial. Having your product sitting smartly on the shelf instead of in someone's shopping cart isn't what this is all about. While having people talk about your product is nice, it's more essential that they buy it.

Bottom line: the best innovators focus on quality not quantity and on doing a few important things really well.