How this works:

You or someone else gives you an idea.

We should be running Google Ads targeting our competitor's keywords

Boss: Why aren't we running ads on Twitter?

We should be more active in groups and slack channels around our core competencies

That idea has an underlying hypothesis. What is it?

A lot of people are searching for our competitor, we should piggy back off that

Don't be afraid to go deeper

Boss sees competitor ads on Twitter all the time

Brand awareness and some possible leads in those channels

What is the quickest way to test this idea?

Launch a single Google Ad search campaign

Don't make it a huge campaign. Find a way to start
proving (or disproving) the hypothesis.

Launch a single Twitter campaign

Find a group to start engaging in

The trick is finding that simple test. But if you understand
the hypothesis, you have the ammunition you need to get
started.

Most of these tests are meant to be something you can launch within one or two weeks, like a software sprint.

You want to learn something quickly, and then iterate on to another experiment around that hypothesis (or move on to something new).

KNOW YOUR BANDWIDTH

Some of these projects are a big commitment, so I don't normally do more than 2 at a time.

Every Monday morning I schedule time during the week that I will need to get going on the experiments.

I review this each week with our founder so he knows the experiments we're running (accountability).

Depending on how things go, I try to stage things ahead of time if I can. For example, if I need a contractor to help setup a Google search campaign, I start looking before I actually need them.

SECRET: When you have these experiments running, you'll be less distracted (or stressed) by what you COULD be doing. You can just add it to the list and prioritize it with the others.