Magic squirrels and magic tree roots helps create a great process

Summary:

• Start with why (tip of hat to Simon Sinek) for any process improvement or automation -- Why do we need to do something, why now (leave out the why a certain technology or vendor until you have a fully baked process for now)
• Treat technology like magic squirrels / magic tree roots -- it's magic and we can figure out a way to make it work
• Explore and then fully understand the process flow at a high level and then in detail
• Walk through the process with real world examples and data
• What does good or great look like? What metrics help us measure and ultimately know we are doing a good/great job?
• Get the process to work manually if you need to and understand each step before you automate -- note this is likely messy at first!
• Make it production ready -- think about as many ways as possible this process can break and then build in safeguards to stop common issues from occurring upfront (or at least have a fast workaround until you can get a permanent fix)
• Good enough solutions are many times good enough! Value you and your team's time and work on the highest/fastest return on your investment which may be a completely different process - if in doubt, explain what you did to executive leadership and have them help decide where to go next

Article (about 5 minute read)

In the early days of COVID-19, my friend Robert and I had a meeting at the local park – sitting 6 feet apart

Robert came to me with what he thought was a technical problem.

Robert’s team had him spinning in circles around technology all week. Robert started talking about programming languages, scripts and templates and I quickly said whoa whoa whoa -- hold on.? I said for our discussion right now, the technology we are going to use was magic tree roots and magic squirrels (we were in a park after all).?

Robert was baffled, but I said for him to continue.

Let’s start with the first question -- “start with why” (tip of the hat to author Simon Sinek) – why do we need this integration, what would it do for our customers?? What is the primary benefit?? Why is it important?

The why was obvious – this process was the primary workflow for his company to take in credit cards to run transactions through 3rd party systems like Toast and then tie these transactions to loyalty cards for restaurants. ?The transactions were core to the top line revenue and the loyalty card information was critical to understand key information such as who bought what, when did they buy it, day of week and time of day all of which was the going to be used to market to the most loyal audience members and drive revenue.

Question 1:? Is the process documented?

Robert gave me a long look after I asked that simple question and then slowly shook his head…not really.

I pulled out some blank pieces of paper.? As Robert described the process, I drew out the various components of the process on the paper at a high level.

I then showed Robert the piece of paper and it described exactly what we wanted the process to do.? Robert went further to say that no one in his company or any of the partner companies in the integration had shown him a simple, easy to follow process diagram like this.? Internally, all the discussions were around technical details and I pointed at the ground -- magic squirrels and magic tree roots.

Question 2 – What are the triggers and what data (required and optional) is involved with each step – how does the data flow through the system?

We went through the previous diagram and started digging in deeper to the process. I added notes on when and how data moved through the system -- triggers, required/optional data, any other outputs like email out to the customer or notifications on mobile devices, etc. Robert started struggling to answer these basic data related questions in his own process and I gave Robert a little smile.? I patiently and gently gathered as much information as he had…but I knew he didn’t have all the answers.

Question 3 – Waving a magic squirrel tail, how might this process work well?

I walked Robert through the business process and the way the data most likely needed to flow through the process.? I had experience with credit card processing via multiple engagements with credit card companies and financial institutions. I kept asking Robert if this made sense and he chuckled and said no one had walked through the process end to end before with him using real world data and examples.

Robert then shook his head and said quietly to himself we don’t know much do we.? I asked him to go on and elaborate. Robert said the technology isn’t the key problem – we don’t understand the process well, we don’t have the data elements, triggers and what/how this should flow end to end.? I said this is common – many people start (and get stuck on) the technology. The technical people love to talk about the “how” which many times is the last piece of the puzzle…not the first piece.

Question 4 – Metrics and Reporting

I wondered out loud – how do we know we are doing a good or great job with this process? What is key set of measurements / metrics to make sure this process runs smoothly.? Error handling?? Error rate?? Expected end to end time to complete the process – especially important for any credit card processing?

Robert chuckled and said that this was the type of things that he brought up multiple times, but the team kept pivoting back to the technology over and over again.

Immediate Result – Successful Proof of Concept works within a day

Robert took a picture of the diagram with his phone and went back to the team internally.? Every time anyone brought up technology, he grinned and said that it would be solved by magic squirrels and magic tree roots. Robert refocused the team back on the underlying problem…we need to know the process flow, the data and metrics in order to be successful.

The project had been stuck for at least 2 weeks and within just 1 day, they were able to figure out the correct process flow and data elements and lo and behold, the proof of concept integration worked perfectly with the technology that they were already using for other projects…. once we had all the right elements click into place.? The team went further and put into place error handling of what we all collectively thought would be the most common errors (credit card information must be in a certain format for example) and started collecting metrics for data analysis.?

Moving from Proof of Concept towards Production

I also advised them that once the proof of concept worked, the next step was to figure out not just how to find the errors but more importantly stop errors early – assist users and systems to enter correct data, automatically flag and try to fix , throw “smart” errors and automatically fix common errors when they occur.

Robert was able to focus on the team effort, communicate what was truly important and do all of this without a single PowerPoint slide – just discussions and drawings on pieces of paper....and the help of some magic squirrels and magic tree roots.