Sandy Kemsley’s Vlog - Best practices in business automation application development - design #1
Sandy Kemsley Photo

Business automation best practices

#3 – Application development – Design (part 1)

By Sandy Kemsley

Video Time: 7 Minutes

Hi, I’m Sandy Kemsley of I’m here for the Trisotech blog with the third in a series on best practices in business automation application development.

In the first episode, I talked about the automation imperative: all of our cool new technology lets us automate faster and better and automate things that were never possible before. However, everyone else has access to the same technology and you need to up your game in business automation or your organization won’t be able to compete.

In the second episode, I looked at two best practices in strategic vision for business automation: first make business automation a strategic direction and second pick the right processes to automate, especially at the beginning. Go back and review those two earlier episodes if you haven’t already watched them.

In this video and the next one, I’m going to delve into one of my favorite topics: best practices in the design of business automation. Design best practice that I want to talk about is related to metrics.


Metrics are often something that aren’t thought about until after a process has already been automated, but that’s completely the wrong way around. How do you even know what to design and implement unless you understand what’s important. Metrics are supposed to be measuring what’s important to your organization, and this is not just about departmental efficiency like doing the same things faster better and cheaper. You need to take more of a business architecture view of metrics where you start with the corporate goals at the top level, and then use that to create metrics for the end-to-end processes that might span multiple departments and multiple systems but are aligned back up to those corporate goals.

Improve Customer Satisfaction

Let’s say you have a corporate goal to improve customer satisfaction, and quite frankly who doesn’t? Now, that’s a little bit fuzzy, but if you dig into the reasons for your customers dissatisfaction, you’ll probably find that this translates to operational metrics at the lowest level. Things like quality, end-to-end cycle time, process transparency. In other words, if you could process the customer’s order right the first time, get it done faster, and let them see what’s happening with the order during the process, then your customer satisfaction is going to improve.

Now, what does this translate to from a design standpoint. Well, if you want to improve the quality of your customers orders then automation is definitely your friend. Look for the places in the process where a person is doing repetitive tasks and decisions that are prone to errors. Such as re-entering data between your website and your order fulfillment system or deciding which shipping method to use. You can then design in automated connections between systems and design automated decisions where there are clear business rules. This is going to improve data accuracy and reduce decision variability. Both of which are major contributors to the entire process quality.

Improving the End-to-End Cycle Time

Now, adding this type of automation also has the effect of improving the end-to-end cycle time which is the second metric that would be related to our corporate goal of improving customer satisfaction those activities and decisions which are automated now happen instantaneously at any time of the day or night rather than waiting for a person to process them during regular business hours.

Process Transparency

Now, the third metric commonly tied to customer satisfaction is process transparency. The customer wants to know what’s happening with their order and you need to build in ways for them to get that. It might be notifications sent to their email when a Milestones reached or it could be a portal that they log into and then they can check the order status directly themselves. Once you have an end-to-end process orchestrated, even if it’s a combination of automated and manual steps across all these different systems and departments, you can start to add in these points of visibility for your customer to track their own order. And then, as an extra bonus providing ways for them to serve themselves in terms of monitoring their order, you reduce the number of calls into your call center. This reduces your costs at the same time as you’re improving customer satisfaction. Total win-win!

So that’s just an example of how you can take a corporate goal like customer satisfaction and take that down to the level of operational metrics, which you need to do. You then need to make sure you have those linkages between those operational metrics and the corporate goals so you can make sure that they’re serving the goals in the right way.

Understand what makes the process work

Now, the second design best practice that I want to raise here is that you need to understand what makes the current process work, or not work. This doesn’t mean that you do a micro level as is analysis and then just automate the same steps, rather you need to dig into your current business process to understand the key elements of success.

You might start with some sort of automated analysis like process mining, for introspecting your current process, but the key thing here is to go out and talk to the knowledge workers who currently work in the process. There is a great episode of Michael Lewis’s against the rules podcast that really highlights this Michael Lewis is the guy who wrote the book Moneyball and several other books that dig into why certain things in business work the way that they do. Now this episode is called six levels down, and it looks at a U.S Healthcare billing company and about how they became successful by gaining a deep understanding in how to apply the right billing rule at the right time. And the way that they did this was to find the people six levels down from the top of the organization. In other words, the people who actually do the work and understand how and why things work. These are the people at the front lines of the business processes, even if they’re not customer facing, they understand which parts of the process and rules are necessary and which are not.

And spending time with these frontline people while they’re doing their work has always been part of my design process. I definitely see this as part of this best practice of understanding what a process needs to do in order to be successful and I think that that’s something that you should incorporate into your design practices when you’re looking at business automation.

Now, once you’ve understood what’s needed to make the the process successful you have to look at how to design that into your new automated process. You might have some parts that still remain manual because that’s part of the secret sauce of the process, but you’ll probably find a lot will be able to be automated. You have to start however with that kernel of truth about what’s required for a successful customer transaction within that process.

That’s all we have time for today. Next time I’m going to talk about two more best practices in business automation design and a couple of failure indicators that you can watch out for. I’ll wrap up the series after that with a video on best practices in implementation methodologies.

That’s it for today. You can find more of my writing and videos on the Trisotech blog or on my own blog at See you next time.

Blog Articles

Sandy Kemsley

View all

All Blog Articles

Read our experts’ll blog

View all

Learn how it works

Request Demo

Confirm your budget

Request Pricing

Discuss your project

Request Meeting