Interview: Matt Calkins speaks about Appian & low-code data integration
When we last spoke to Appian's Matt Calkins in November 2020, it was an exciting but demanding time for the company. Employees were volunteering in the community, and the company worked hard with its customers to ease the 'shock' of disruption through better use of technology like low-code.
Appian's efforts have not gone unnoticed. Gartner named Appian a Leader in the 2020 Magic Quadrant for Enterprise Low-Code Application Platforms. The most exciting achievement, however, was being named a Gartner Peer Insights 2021 'Customer's Choice' for Enterprise Low-Code Application Platforms.
"It's terrific to be ranked in that study because it's based on customer feedback and shows our commitment to the low-code market," he says.
Customers have plenty to be happy about. Appian's Low-code Automation Platform, which was updated this month, includes low-code capabilities like intelligent document processing (IDP), robotic process automation (RPA), and a new feature called low-code data.
Before we get to low-code data, we want to explore the concept of what Calkins calls 'slow software', which is a rigid method of development that is resource-intensive and rather cumbersome.
"It's the opposite of agile - it usually takes a large investment to create software, it takes a long time to develop, and it's very difficult to change. You can't tweak it if a business model changes, and you can't consume it in expected ways."
"You can't develop a piece of slow software and say, 'hey, this week, we're all working from home so I expect to be able to log on with a mobile device' - it doesn't work."
Unfortunately, though, it's how most software was written. Not so long ago, organisations assumed that their IT needs wouldn't change too much, so they only needed a small investment.
"That left them with an infrastructure that's not ready to pivot when it's necessary. It's a trade-off they made then, but it's not a good trade-off anymore."
Luckily, IT departments are looking for a new model that makes software agile. A model that can adapt to business changes, security changes, platform changes, cloud changes, and a model that can adapt with as little work as possible.
"If things change in the future, organisations may not have time to make those changes. There wasn't time last year when everybody had to work from home. Nobody had time to go back to the applications and say, 'we're going to build in a mobile access module'. Too late. Your application does what it does.
"So you either invest in the ability to pivot, or it's too late to pivot and you lose the benefit you would have gotten from that agility. That's why it makes sense to spend now versus later because later there won't be time."
He says organisations should invest in flexibility so they can meet all their circumstances with confidence. Slow software cannot do that. This is where low-code comes in. By nature, it is the polar opposite of slow software because it simplifies application development and supports flexibility.
"Last year, worldwide mobile application usage tripled. Appian mobile application usage multiplied by 20 times in the last four quarters. That shows what happens when you make it completely easy to consume an application in any way," says Calkins.
One of the things Appian is also focusing on this year is the concept of 'low-code data'. It's now part of Appian's Low-code Automation platform and it takes the focus away from the database skills traditionally needed for data integration.
Data, Calkins says, has largely been absent from the low-code process.
"Appian has been working on low-code for a long time. Dragging and dropping is an intuitive way to communicate a logical sequence of events. But data and the integration of data has not been done in a low-code manner."
He adds that the hardest thing about building applications is data. Builders have to know where it is, if they have to join it, how to get it to perform well, what permissions they need, and much more.
These kinds of things need people to be data experts. But those who want to build applications may not be experts, so they may leave a lot of useful data behind.
Calkins adds, "Data integration behaviour involves a lot of coding, typing and scripting. It's not a way that empowers people the most."
"It's not useful to own data, what's useful is having the data in front of you when you make an important decision that pertains to that data. Should this customer get a discount? Should this customer get an upgrade? Should you know within the moment you make a decision? If you bring the data to bear that decision, you made your data valuable."
By creating low-code data, Appian seeks to empower those making the applications and decisions. And it's always easier when an app can seamlessly connect to data in a way that is both intuitive and simple.
Calkins digs a bit deeper: "If you're into data, you know how hard it is to do a join across many databases, the data can be dirty, there aren't any index relationships, there's no foreign key or primary key. There's no good way to bring that data together. We're solving it by building a third-party index, like kind of a join key of our own.
"It works behind the scenes. The person who builds this application doesn't need to know what a key is, because it works. The person asks for the data they want and we make sure they have it when they ask a question or when the decision faces them. It has a big impact because data is wasted most of the time."
With low-code and data integration now fundamental to Appian's capabilities, it naturally raises the question about what's next for the company. Could analytics and decision-making capabilities be the next natural step? It's a question that remains unanswered for now, but there are countless opportunities ripe for the picking. We can't wait to see what Appian does next.
