Business Experiment or Business Play? Let’s Do it Right.

performance improvement quality improvement management plan

This isn’t what we want!

Should we “experiment” or “play” with our businesses?  And how do we do business innovation the right way?

Businesses need to change in order to grow.  Sometimes it’s hard to figure out how our business might change for the better, and there might be more than one appealing approach.  So… how do we change when there are choices?

I’ve been reading advice telling me I should have fun with my business.  It’s well intentioned; the pundits are trying to break nervous, clutching business owners out of their self-imposed inaction.  The pundits talk about exploring, trying, “failing fast”, and even playing with the business.  I appreciate the sentiment.

But I don’t, for one minute, take their advice.  Here’s why.  The definition of play is “activity engaged in for enjoyment and recreation.”  I’ve written this before and will again, many times; the first rule of business is to survive and second is to make a profit.  Let’s have a tight, serious, focused approach to business innovation.

Let’s try the word “experiment”, meaning “a test, trial or tentative procedure for the purpose of discovering something unknown.”  In definition there’s a procedure, and that’s what we’re going to focus on today.  That procedure, tailored for business, is what we want to get right.

The procedure might be used for any business situation were we have more than one approach to improving our business, and no obvious deciding factor.  We might be looking at a new sales tool, a new marketing approach, a new production method, a new financial reporting format, etc.

The Business Experiment Procedure

Let’s step through the procedure.  It looks like the scientific method we learned in middle school science class, but this procedure has been carefully rethought for business experiments.  I’ll explain how and why, step by step.

Compose the assertion… and include the downside risk.  In the scientific method we’re supposed to form a hypothesis, a way we think the world should work.  In this business experimental method we describe, in one sentence, an action, the expected outcome, and the worst-case alternative.

In business, getting it wrong usually hurts.  We lose market share and sales, lower profits, pile up unsold inventory, sink morale… we don’t want any of that.  Refusing to take a foolish chance is just good business judgement… especially when there are so many ways to improve a business.  This is a good place to stop and reconsider whether we should go ahead.

Design the experiment… including the warning indicator.  In the scientific method we plan the actions, the needed observations, and the instruments needed.  In this business experimental method we do the same, but we include one more important step; we describe observations that might give us early warning that we’re off track.   (Usually the warning indictor tells us that our worst-case alternative is announcing its business-killing presence.)

Occasionally, medical trials are stopped because an unintended outcome is hurting the participants.  The same can be true in business experiments.  That’s why we pay attention to our warning indicator and let it tell us to stop.

Collect the data.  Much of the advice I read about collecting data tends to focus on ethical/legal issues (remember the fuss that Facebook provoked?), or methodological issues like ensuring a statistically significant sample sets.  Let’s let ethicists, attorneys, and statisticians advise us if we really need that kind of help.  More important, I believe, are avoiding two mistakes I’ve seen time after time.  First, we must collect the same way every time.  If we’re using sales staff with checklists, we won’t change the instructions half way through the experiment.  If we’re using a plugin to capture questionnaire responses on our WordPress blog, we won’t change the survey questions.  Second, we won’t stop collecting discouraging or disappointing data.  (See the comment above about warning indicators; either stop the experiment or continue to the end.)

Here’s a very good reason to collect the same way every time, even for disappointing data.  That data may be our doorway to the unexpected.

Analyze the data, looking for new knowledge.  Digital businesses supposedly have an easier time using this experimental process than  brick-and-mortar businesses… or do they?  A/B comparison tools and analytics make it easy for digital businesses to set up alternatives (design the experiment) and collect weekly, daily or even hourly results (collect data)  But this step, analyzing the data, is where more traditional companies have an edge; it’s possible to take directly with the those involved, whether they’re customers, employees, or even investors.  Just a few observations, followed up with face-to-face conversation, can yield much more insight than a hundred more detached data items.  The answers to a three-fold question tells us how successfully we’ve analyzed our data:

  • What do I know that I didn’t know before?
  • Do I mistrust this new  knowledge and can I say why?
  • Can I use this new knowledge to do something?

Also, remember what we said about disappointing data being a doorway to the unexpected?  The unexpected is a rich source of innovation possibilities and disappointing data deserves our close inspection.  This is a key distinction between business experiments and the scientific method, where disappointing data is often treated as a form of experimental contamination.

Act on the knowledge.  For either online or brick-and-mortar business operations, this last step is the most critical.  For scientists, the knowledge is an end in itself, but for business people it’s just facts filling storage until we do something with the knowledge.  We change the structure and behavior of our businesses, we focus on a different aspect of our customer’s needs, we change how we describe our products and how we get the word out.  We spend time and money.  WE ACT.


Let’s experiment with our businesses and not play with them, using the experimental business procedure:

  1. Compose the assertion… and include the downside risk
  2. Design the experiment… including the warning indicator
  3. Collect the data
  4. Analyze the data, looking for new knowledge
  5. Act on the knowledge

We always need to act.  We need to wholeheartedly commit to the change, and be patiently persistent.

(What other qualities should entrepreneurs have?  Here’s an introduction to business tradeskill.)




  1. Chris Chadbourne says:

    Yes, this is suitable for just about any kind of business.

  2. Experimenting with business can be great or challenging– It’s great if you find changes that are useful, and challenging if you make a move you shouldn’t have. Sometimes change can be great, but overall it’s up to the customers.

    • Chris Chadbourne says:

      Tony, that’s so true. I especially like your phrase “it’s up to the customers.” When we stay close to our customers and change incrementally, we raise the odds we’ll get it right. In the last few days I’ve been listening to folks telling their business stories, and I’ve noticed that fortunate serendipity can play a big role… if we’re prepared to take advantage.

  3. Chris, you are welcomed .

  4. I’ve been looking for best business experiment blog post,found this one.I wouldn’t mind sharing it with other business enthusiast. Well crafted details.

  5. Fantastic read. This is especially good for new business owners who aren’t particularly fluent with basic preparations when it comes to starting up and running a business. As a business owner, you have to take full advantage of data and statistics to help predict your potential outcome.

  6. You choose the right kind of images. You know what you are doing big man! Great!

Leave a Reply

Your email address will not be published. Required fields are marked *