Integration-ready applications

Pat Garrehy

Pat Garrehy is the founder, president, and CEO of Rootstock Software and has an extensive background as a software visionary, architect, and engineer.

Pat Garrehy of Rootstock Software forecasts how users will create their own interfaces to fit their work.

Interview conducted by Vinod Baya and Bo Parker

PwC: Pat, can you tell us about your company and yourself?

PG: Sure. Rootstock was founded in 2008 as part of a venture I was doing called the Danville Group. We got started when NetSuite asked if I could extend its package to handle manufacturing, which I agreed to do. In 2011, Rootstock moved to the salesforce.com AppExchange platform.

We provide cloud-based ERP [enterprise resource planning] manufacturing and supply chain solutions for midmarket manufacturers. We help them cut costs, improve processes, and increase revenue with minimal IT infrastructure investment.

In the early 1970s, when I was at Martin Marietta, I was the primary developer and designer of an MRP [material resource planning] product that was taken commercial, called MASII. It was on the mainframe and sold to large corporations in the Silicon Valley, such as Shaklee, Fairchild, FMC, and Amdahl Corporation.

In the early 1980s, I founded Relevant Business Systems, a client/server ERP software provider that focused on discrete manufacturing companies. Relevant was sold in 2006.

PwC: So you have been in the enterprise applications market for about four decades. What is new today?

In every single technology shift, one thing has stayed true. Customers want all the functionality they’ve enjoyed before, they want more of it, they want it faster, they want it cheaper, and they want it easier to use.

PG: Indeed, I’ve been through many technology shifts—mainframes, minicomputers, microcomputers, client/server and, now, the cloud. In every single technology shift, one thing has stayed true. Customers want all the functionality they’ve enjoyed before, they want more of it, they want it faster, they want it cheaper, and they want it easier to use.

The functionality of the core packages is by and large the same, but the technologies in use are new. For instance, today, things are mobile, which is a strong pull for new features and use cases. There is also talk about social, which means that companies want to be more collaborative and more open with their vendors and their customers. Bottom line: mobile and collaborative features must be in the software because people are getting accustomed to them.

There is another new challenge. In the next 10 years, customers will hook to much more than just their own financial systems. With our solution, for example, they will need to hook into the corporate financial system. They will also need to hook to a number of configurators. Depending on who they are, many may need to hook to a number of quality control systems as well as to a variety of time and attendance systems.

On the AppExchange platform, they ask all the developers to develop in a certain way so solutions are more conducive to integration. You just cannot anticipate who will integrate to whom. Therefore, you must do more in your development than simply create APIs [application programming interfaces]. You must “harden your objects.”

PwC: What does that mean?

PG: It means all the business logic is on the back end and every single file or table you develop is hardened and has the potential to be called by some other application. It means you have exposed all your fields in a manner that is easily accessible but the integrity of that particular table is maintained.

It’s being done and, most importantly, it works. For example, when we hook to a system on the same platform, we can expect that objects in the other system are hardened and they’re easy to access. We also know that we can easily add fields to their screens [interfaces] and they can add fields to our screens. These kinds of things are part of the new development paradigm that is defined around ease of integrations.

PwC: Software packages are large. Do you choose what is hardened in this way and what is not?

PG: That’s exactly what we do. For example, we’ll harden our sales order entry header and sales order entry line. We’ll harden our purchase order, our engineering item, our bill of materials, and our project master file. Right now, we’re in the process of hardening our work order. By the time we are done, we will have hardened more than 75 percent of our objects.

For the most part, you harden those things that are more likely to be called by somebody else. On the other hand, we do not need to harden the control record; no one needs to call a control record.

Importantly, hardening is not a need just for other software vendors. It is invaluable for the end customers. When customers want to build their ancillary apps and do their own customization on top of our system, the same principle holds.

For example, we have our own RMA [return merchandise authorization] module. We do it a certain way and it’s tailored to our warranty system. However, customers might have their own RMA module that has a lot of their cost accounting built into the module and they would not want to lose that. They can and should be able to use their RMA module with their methods of cost accounting, yet they can access the objects in our system to make sure the supply-demand balance stays correct. That’s the kind of thing that is going on and will go on more in the future. You need to build for that.

PwC: How are mobile trends influencing your strategies?

PG: I believe mobile trends are accelerating the push toward cloud [software as a service], as mobile development and usage are better suited for the cloud architecture. There are use cases we don’t even know about that will push people to embrace cloud technology.

Mobile platforms are creating new energy around the user interface. We must really think about how the app is going to be used, which will be different from how it was or will be used on the desktop. This difference in usage is the other reason we are hardening the objects, giving users the freedom to create their own interfaces. Since there is integrity of logic, end customers can slap on another screen [interface] as best fits their work.

PwC: What is the impact of such developments on the design of your solutions?

PG: While the roles and needs may differ, they are all connected to the same ERP system. For example, if you look at a manufacturing organization, the receiving clerk, the person in the stock room, and the worker on the production floor usually have simple but specific needs to process an order, forward a receipt, deal with expiration dates, and so on.

Then, go to the other end of the spectrum, the C-level person who has analytical needs. That person probably has a large screen and is thinking in 15 different dimensions, because that is how he or she runs the business. As you go down from the C-level to the managers, they may have similar needs but fewer items on their desk.

Everybody had the same user interface 20 years ago. That is definitely different now. In the future, there will be Rootstock’s complex screen and there will be Rootstock’s simple screen. Customers can take their favorite third-party app, create new interfaces, and put them on top of ours and other applications. Users will develop their own interfaces in effect.

Five years from now, there will be tools for a power user who understands how work is flowing across the various roles and individuals, and who needs specific information to do his or her job more effectively. I believe those tools will superimpose on packages like ours.