I want to write about internal triggers, the topic of the TechCrunch post I wrote with Nir Eyal. 

I think that internal triggers are absolutely fascinating, but incredibly complex. They’re random/unpredictable by nature, which makes them all the more seductive. 

===

Before we start to talk about internal triggers, it’s important for us to understand triggers in general.

Triggers are calls-to-action, reminders. An email telling you to go to the store and pick up groceries is a trigger. So are your mother’s persistent comments to “sit down and study”. Pop-up ads telling you to “purchase cheap viagra” are triggers - so are billboards encouraging you to “stop by the store to pick up some Coca Cola”.

Truly, anything can be a trigger. Any stimulus that causes an organism to perform a behavior is a trigger.

At this point, you might be rolling your eyes. You might be thinking that this concept is too vague to be meaningful. However, on that point I would disagree.

While some stimuli will cause a diverse array of behaviors, others will cause predictable responses. This is especially true when we’re dealing with people in highly-constrained environments - such as the internet. While an individual can perform thousands of different behaviors in the average physical/3D environment, most web environments only allow users to perform a constrained set of behaviors. Conversion focused web sites will often only allow users to perform 4 or 5 behaviors (this includes clicking to different informational pages), while retention/time-sink sites will allow users to perform up to 200 or 300 different behaviors (Facebook).

However large the set of possible behaviors in an online service, it’s still minimal compared to the field of all possible behaviors one encounters in the real world.  

This is a great thing, however. It allows us to design behaviors more effectively than we can in the physical/real world. Since we’re dealing with a relatively small number of variables, it means that we can painstakingly map out stimulus-response relationships between information and calls-to-action on our sites, and the tracked behavior of our users. This makes the internet one of the greatest behavioral laboratories on the planet - we’re dealing with individuals in constrained environments in which we can record/track everything. 

Of course this means that it’s going to be extremely hard for us to generalize our findings to “real” world - but since our lives are moving into the digital realm, it’s not necessarily a bad thing to be an expert in analyzing and designing online behaviors. 

Anyways, the point is that certain stimuli will produce fairly predictable responses in groups of people. This is especially true when the stimuli we’re talking about are explicit - such as billboards which tell you to “purchase Domino’s Pizza for your football parties”. 

As we said before, any stimulus that causes a behavior is a trigger - but for the purposes of the rest of this article, we’ll say that triggers are stimuli that produce predictable behavioral responses in the particular population in which we’re interested. That is, triggers are stimuli with a fairly constrained/uniform range of responses. 

Great. Let’s call this part 1 -  I don’t have time to finish this train of thought right now. I’ll continue it tomorrow.