Getting value from data: Creating the future SaaS

Where is the future of data headed for SaaS companies? 

Salesloft’s Chief Product & Engineering Officer Ellie Fields recently appeared on the podcast “SaaS Scaled” with Arman Eshraghi, the CEO and Founder of Qrvey, to hash it out.

During the show they discussed the changing trends in how companies use data, like drawing insights from unstructured data, putting workflow at the center of what you do, and more.

We’re covering the highlights in this article, but you can watch or listen to the full podcast episode.

Hallmark of modern SaaS: It doesn’t matter where the data is

Arman: Are the SaaS apps being built today more frequently optimized for data entry, workflows, or analytics?

Ellie: I think that’s one of the major areas of disruption in the industry, which is apps used to be one of the three, and in fact the more legacy apps were optimized for data entry. 

When I think about one of the most powerful, largest categories of SaaS apps, in fact where SaaS started was with CRM. And CRM was disruptive at the time. It was really a window into the database more than anything else. It was not optimized for workflow, and it required sellers to do a lot of data entry. 

The new generation of apps, and Salesloft is one of them, is where you’re really putting workflow at the center. You’re putting the user at the center. The core of what you’re trying to do is help a person work better. When you take that perspective, it doesn’t matter where the data is.

For example, many modern apps will integrate a lot of different kinds of data. We work with emails and calendars in Google and Outlook, meetings platforms, CRM data from Salesforce Dynamics, and so on. I think that’s really a hallmark of modern SaaS because it’s not about this app and the silo of this data. It’s about what you’re trying to get done and the data you need to do it. Because data ecosystems are open now, you can get that data at the right point. You don’t need to own it.

Removing barriers and creating structure from unstructured data

Arman: With data being more elastic and flexible and users requiring more self-service, will we see barriers being reduced?

Ellie: Yes, absolutely. You’ve touched on unstructured data. One of the most powerful advances in the industry has been the ability to transcribe something that’s been said and get insights from it. I think unstructured and structured data have a lot of intersections. 

For example, you could take a lot of conversations between buyers and sellers, this is something we do on our platform, and look at all the transcripts and say, “How many times was this certain product name mentioned?” You’ve actually created structure out of unstructured data, which again is just massively powerful. But it’s also powerful to go back to the unstructured data itself in that original transcript.

To get value from data, traverse up and down

Ellie: In our app, data is anything from the number of times you reached out to someone by email, phone, or SMS, in addition to what you said when you reached out. So, that’s structured and unstructured. That’s just one example. But to really get the power from the data at the time you need it, you actually need to traverse up and down. 

Coaching employees is really important in sales. A lot of data shows that coaching results in higher close rates and shorter sales cycles. It’s powerful, but it’s hard. To coach effectively, managers need a few things. They need to see all the data. They need to see everything their sellers are doing, not just the meetings they’re having, but the emails they’re sending. They need to see all the interactions between the buyer and the seller in aggregate, because there are probably thousands a week and you can’t look at those one by one. 

But then you need to be able to drill all the way down. If a seller is sending a lot of emails but not closing many deals, you want to look at emails and the conversations with the buyer. There’s something somewhere in there in either the structured data or the unstructured data that’s the clue. 

Data is really only useful when you have it in that moment when you’re doing a thing, and if you can go all the way from the high level all the way down the bottom and back up.

Forecast meetings: From painful to productive

Ellie: Another place we see this is forecasting, which I think is a classic example of how SaaS can change a market. Forecasting is something every company does, and they all hate it.

We watched people have a bunch of these meetings and noticed a few things:

1. It was all done in Excel, which is a classic opportunity to take a manual, highly error-prone process and move it to a best practice flow. 
2. We also saw the entire team, from the Chief Revenue Officer down, doing a lot of basic fact-finding, such as the last outreach date. The business-focused conversations that can move deals along weren’t happening because people were stuck trying to gather the data. 

So, in forecasting what you classically want to do is to traverse all the way from a time trend or your number this quarter, and then go down by region or product, and go down again to rep, and down to deal, and even go all the way down to the last meeting, and then go all the way back up.

That’s the kind of power that a data-enriched workflow app can give you. All of a sudden, forecast conversations change and become ways to move business ahead and prioritize. They become interesting, rather than a painful, grinding exercise and trying to figure out what happened last week.

SaaS apps should deliver best practices – not just technology 

Arman: What are the impacts of SaaS providers putting workflow at the center?

Ellie: In a SaaS application, you want to move fast, and you want to deliver a single flow. I think it’s been an anti-pattern in a lot of software to get really custom, deliver multiple versions to multiple customers. You can’t really do that in a SaaS app and be successful. So that brings up the question of, how customizable is your product? The point of view I always take is that we want to have an opinion on best practices. 

When you deliver technology, you’re not just delivering bits and bytes in an app somebody works in. You are studying the field. Hopefully you’re watching customers and learning about best practice. I believe you should be encoding that into your app as the default. So, if somebody just bought your software and did the minimal setup possible, which hopefully is a very fast setup, they are running a best practice workflow.

We certainly try to do that at Salesloft — to understand best practice sales flow. But it’s never going to work for everybody, especially as you go to bigger organizations. So, then you ask the question, “What customizability?” You want that best practice flow in the default and then the right set of knobs. To me that’s one of the hardest questions to navigate, and one of the most important questions: What’s the core flow you’re supporting? How do you allow modifications of that flow? When are you going too far?

Using data to manage development

Ellie: Data culture gets misunderstood sometimes. I think about data culture as just having the right information to make decisions. I think there’s a certain level of information you just need in any organization. For example, in engineering we’d be irresponsible if we didn’t monitor our systems. 

I tend to think of a data-informed culture rather than data-driven. By that, I mean looking at velocity is important. Looking at performance of the different things that you ship, whether that be capabilities or performance improvements. It would be crazy to spend three months of a team’s time, which is quite a bit of money, on something and then never check if it actually worked.

Creating the future

Ellie: But I hesitate to goal on those metrics, especially in an engineering organization. Fundamentally, a lot of what we’re doing is creating the future and we’re working with unknowns.

I think that we can be accountable for things like providing features that very important customers want and making sure that we can bring in some revenue. Making sure that we’re thoughtful about the way we build things. But then a lot of the data becomes a way to understand, for example, velocity. I would never goal teams on velocity. But what you can do is observe if typical velocity falls off and evaluate the causes.

But we have a responsibility to look at the information that we can to try to understand what’s going on. Great organizations — and especially great product and engineering organizations — continuously improve. 

It’s one of the fun things about being in our industry. You can’t just come out day one and 30 years later be doing things the exact same way. You have to keep pushing. There’s always a better way and data helps us figure that out.

Check out the full conversation with Arman and Ellie.