One key aspect of how of any new tool gets adopted is : how well does it integrate within the existing environment.

If the value delivered is good enough, you could get away with more friction to usage and adoption.

If not (and even if it does), it can pay huge dividend to think about that part.

There are mainly 3 components to consider : the trigger, the inputs and the outputs of each process.

The process itself being a "black box" that would just do its thing.

The trigger

The trigger for a process is what gets it started. It is usually a single event that can be of multiple kinds :

• Manual : a response to some action by a user, like a click on a page for example.
• Time-based : for planned actions, things that happen when the clock reaches a specific time or frequency.
• Value-based : reacting to a change of value, or when a set of values matches some conditions. In many ways similar to manual triggers, the difference is the source.

Think about what makes the most sense, one or possibly a combination of those? What's possible is dependant on the control you want to have as well as what the process requires to run.

The inputs and outputs

For both inputs and outputs, the question you should ask yourself are :

• Are there any necessary ones ?
• If yes, what's the best possible way to ask or present them (depending on your constraints)

If there aren't any input or output, you can likely automate everything (make sure to have a way to control that things work as expected though).

For inputs, ask yourself, can they be retrieved and automated in some ways ? Do they have to be manually entered. And if so, what's the most straightforward way to do that.

The tool I have built that I use the most is Podyt. My goal with it was to be able to listen to Youtube videos (not all are worth watching) within my podcast player feed. One key aspect of it is : how do I know which video should be included ? My solution : monitoring a dedicated youtube playlist. When I add a video to it, it gets picked up, converted to mp3 and added to a podcast feed. All I need is 2 clicks on Youtube interface to make it work. Simple and it works everywhere, on the web, any mobile app, ...

For outputs, the question is : what's the best way for you to consume the information? Do you need some realtime notification ? Is a weekly email enough ? ... Consider your needs, how you expect to work with your new tool, process, automation in place, ... and leave space to imagination to uncover unexpected opportunities.

For whatshoes.run, a tools that allows managing and tracking running shoes easily, my goal was to be able to remember and update the shoes used for a training session within the same timeslot I am looking at the activity itself. The original process, using the Strava app was a bit cumbersome, lot of clicks, when I remembered to do it. Now, I receive a notification on my phone (or other device) as soon as the activity gets recorded and have a direct link to confirm or update the shoe. It simplifies the process and the way it is set up ensures that I follow through.

You can be creative with those 3 aspects of external interface for your processes to not only automate more and more, but also gain a competitive edge. Start with the core of what you need accomplished and then expand from there.