MuleSoft CEO Brent Hayward on Salesforce integrations, APIs

While most Salesforce customers intuitively grasp the concept of the Sales, Marketing, Commerce and Service clouds, it's not so obvious where Salesforce's MuleSoft toolkits sit in the ecosystem.

We sat down with MuleSoft CEO Brent Hayward -- who remains at Salesforce amid a number of recent CEO resignations -- to discuss the direction MuleSoft has taken since Salesforce acquired it for $6.5 billion in 2018. Also, what the future holds for MuleSoft, which powers integrations between Salesforce clouds as well as with Salesforce and other platforms such as ERP and HR.

Some companies Salesforce acquires live on as independent businesses. Others get subsumed into the platform. It looks like MuleSoft is still one of those independent units. Is that the reality?

Brent Hayward: It's been an evolution. We were integrating with Salesforce 75% of the time before we were acquired, often connecting e-commerce, ERP and service. Salesforce customers were a large subset of our enterprise customers.

Certainly, we had a class of customers that didn't use any CRM products yet. But that also posed an interesting opportunity -- sometimes they couldn't take advantage of Salesforce because the cost of integrating their legacy infrastructure to it was more expensive than the value they were going to get out of it.

Brent Hayward

After the last two years, we've looked at opportunities to align functionally to the way Salesforce works. One of the big moves was to change our distribution organization to better align to Salesforce's go-to-market approach -- now we have the same teams on the same accounts that will go out to large customers, such as AT&T and Rocket Mortgage, and deliver that full relationship and that full opportunity.

The second thing we saw was in the engineering and product organizations. We wanted to make sure that the future was using MuleSoft to power a lot of these core services, such as Salesforce Genie.

Microsoft, Oracle, SAP, ServiceNow and Adobe are introducing vertical clouds, as is Salesforce. Is this the ebb and flow of the tech industry -- and you're going back to horizontal in three or four years -- or is this here to stay?

Hayward: I think it's here to stay. I think customers want to know that we're not just delivering horizontal IT capability that they then have to make work for their industries. They want to know that we're thinking, how do we serve a patient, or how do we serve a retail customer?

I don't know that you're going to see us chunk out the MuleSoft service and have a different platform for each [type of] customer, because the power is in the horizontal platform. How do we take it to market with services, accelerators, industry specific data models -- the application of that to, for example, go solve a patient problem in healthcare -- is where I think the rubber hits the road with how we serve the customer.

MuleSoft, at its core, is data plumbing. It's hard to make it sound as sexy as the latest AI innovation. How do you, as MuleSoft's CEO, do it?

Hayward: What isn't interesting is sitting around and talking about plumbing. No one wants to talk pipes. Just like if you go into a beautiful modern bathroom, you're going to get yelled at if the water doesn't turn on -- but no one's appreciating the elegance of the plumbing. People are just saying, 'The damn water is not turning on!'

Really, what people are looking for are these incredible outcomes. If you're a new customer of RBC Wealth Management, it used to take weeks to onboard you as a customer. To collect all the information to get your accounts now takes 24 minutes. You don't have to appreciate the pipes in the walls that make that happen. But that's damn sexy in terms of a customer experience.

Who are MuleSoft customers that are not using Salesforce?

Hayward: MuleSoft customers are ultimately trying to create composable businesses with enough flexibility that they can turn on a more digital business model to sell goods, or service the customer, as many did during the pandemic. They don't want to get stuck in a business model that ends up being the end of them.

You can't have a flexible and composable business model if you don't have a flexible and composable IT landscape. So, a lot of customers are looking to us to take what used to be custom, hand-coded, point-to-point, rigid, brittle things and put that into an architecture that has APIs, lets you extract the data from the processes -- from the engagement layer. This allows things to almost become Lego pieces, reusable building blocks. It's a different mindset to do this type of integration work, which ultimately ends up with far less sprawl, where things are reusable. That's what increases development speed.

What is Genie MuleSoft going to look like?

Hayward: The first [major component] is: How do we unlock the data that Genie needs to be successful? If the data is already in Snowflake, if it's in Amazon, great -- we can access it natively. But a lot of enterprise data is not already in that data architecture. It may be in a legacy system or a modern system. Let's make sure Genie has billions and billions of data points to be able to perform AI. It's not enough to unify that data into a single view of the customer -- you have to be able to act on it.

The second involves real-time triggers. What we want to do with that is enable business users, nontechnical users, to go in and create automation workflows. For example, a user might have a trigger for an outage or an issue. We can go design a workflow that actually delivers a different output. With MuleSoft, we can extend that beyond the four walls of CRM.

Let's say there's a service case where there is a service outage. We're going to manage all the ticket flow. But the trigger also could invoke a third-party diagnostic system, or maybe we do geocoding to see if the power's out in the location where a service outage has been reported, and we can prevent rolling a truck. So, think about it as data, a high level of analytics and real-time processing. But then what do you do with all these triggers? MuleSoft takes a Salesforce user's triggers well beyond the four walls of commerce, service, sales and marketing.

At the time of Salesforce's MuleSoft acquisition, the rumor was Salesforce had to do it because it couldn't make its clouds -- some of which also were obtained through acquisitions -- talk to each other. Is that true?

Hayward: I don't think that's true. It is true that when you make big acquisitions such as ExactTarget or something that comes with its own infrastructure, the ultimate goal is to put them together. But it would be a fallacy to say that MuleSoft is running around stitching together a Frankenstein. Genie is the true vision. It is the next-generation platform where we've been doing all the hard work to exist on a single object model.

MuleSoft is really what's connecting the [world outside Salesforce to Salesforce and vice versa]. Serving the customer is much more than a sales request or an account record. It's just gotten really broad. The primary reason for the acquisition, in my opinion, is because customers saw the possibility of what they could do in terms of these customer experiences with Salesforce. A lot of what we do every day is harmonizing these different data systems so that we can serve the customer right.

Editor's note: This Q&A was edited for clarity and brevity.

Don Fluckinger covers enterprise content management, CRM, marketing automation, e-commerce, customer service and enabling technologies for TechTarget Editorial.