When evaluating the performance of a new initiative, we typically ask ourselves whether we should keep doing what we're doing or make a change--whether to pivot or persevere. To answer this question, we might create quantitative success metrics. For example, "If 20% of people sign up for our service this month, we will continue to offer it." Success metrics aren't always black and white, and when things don't work out, it's not always clear how or what to change. Should we give up? Is there a simple improvement we can make?

For questions like these, I follow a 3-step method to determine, "What next?". As a light and fluffy example, here's how I applied my method to coconut flour pancakes, which I've spent the last couple weekends designing, iterating, and improving.

Weeks ago, I found a recipe for healthy pancakes which used coconut flour. A lover of both healthy eating and pancakes, I immediately ordered coconut flour on Amazon and starting counting down the days until Saturday morning.

Carefully following the recipe, I was already planning my Instagram caption.

That Instagram was never posted, because the pancakes were a disaster--burnt and crumbly, and nothing like the photo in the recipe.

It was time to apply my qualitative method of determining how to proceed:

1. What went right? What went wrong?

Staring at an almost-full packet of leftover coconut flour, I wondered whether I should abort this mission. My pancakes were not light and fluffy as promised by the recipe. I wondered, "Should I completely give up and eat something else?" In search of my answer, I enumerated what had and hadn't worked. What went right? I used all the ingredients suggested by the recipe, and the batter had turned out the way it was supposed to. What went wrong? Everything after the batter hit the stove. The pancakes crumbled when I tried to flip them, and when I took it slower, they got burned. Something was wrong about the environment and tools I was using.

When reviewing initial performance of an initiative you are exploring, focus on what went right and what went wrong. This will help you identify what (if anything) to maintain and what to change in making improvements.

2. Is there one thing I can change without changing the product?

Still with leftover batter, I asked whether there was an easy way I could improve my pancakes without wasting it or spending more money. My answer was yes: I could change the environment in which I was heating the batter to see if I could prevent burning and use a different method of cooking them so that I could avoid crumbling them when I flipped them. I put the rest of the batter it in a baking dish and baked it in the oven. This time around, it came out beautifully light and fluffy, but still a bit crumbly.

You might be tempted to totally trash your failed creation, but instead you should identify the simplest change that might make a difference. Often, this will be the environment or conditions, instead of the product itself. This might include aspects like instructions for using your product, the city in which you're testing, the distribution channel, or even the types of people with whom you're testing it out.

3. What changes can I make to the build process?

Still unsatisfied with the outcome of my pancakes, I knew that the next step was to refine my process. I researched coconut flour and how to make it less crumbly. I found several tricks to make coconut flour stick, like sifting the dry flour before adding in wet ingredients, pureeing the fruit that you add into the batter, and letting the batter sit fifteen minutes before baking. Notably, none required new ingredients. These were all procedural changes. The following Saturday, I incorporated these tricks into my batter making process. The outcome? Light and fluffy healthy pancakes!

If simple changes to the environment or conditions still don't do the trick, find ways in which you can improve your process. With a product or service, this may be something like adding or removing a step in the user experience. With internal initiatives, this could be something like adding a checkpoint or removing a meeting.

Things get messy when they don't go according to plan. If things don't turn out as expected the first time around, you don't necessarily need to take your idea off the table. Instead, explore ways to make changes to environmental factors and your process before you let your idea completely crumble.

Published on: Oct 10, 2016
The opinions expressed here by Inc.com columnists are their own, not those of Inc.com.