Taking a Capital Efficient Company Public and Beyond: Medidata CEO Tarek Sherif (Part 4)

Sramana Mitra: The first customers that you were getting traction with, how much were they paying you?

Tarek Sherif: We tried a couple of different charging models. At first, we thought about a per-patient model. Back then, people were charging per data point. We tried a couple of those and they were frustrating for us.

So when we tried the per-patient model, one of the things that you learn very quickly in clinical trials is, you almost never enroll patients at the pace that you think you will. It’s always slower. So it takes forever to enroll patients.

Depending on what drug you have, it could be six months or whatever. If you had a per-patient model and you’re waiting for a patient to be enrolled, you would go out of business. That didn’t work so well. We looked at the data point model.

The problem with that was that our customers had a hard time estimating how many data points they were going to collect. They didn’t really like that even though other people were using it. So we finally said, “We’re just going to charge you a set amount of money to use our software over the course of the trial. Then we’ll charge you ratably for doing that.”

So our earliest trials were actually pretty inexpensive. Maybe the earliest ones were $50,000 to $100,000 for the duration of the trial. That started to creep up pretty quickly to an ASP per trial of around $250,000. Of which, software was about two thirds. Part of it was services as well. Then it just went from there.

Sramana Mitra: That’s the beauty of these enterprise software models that get large chunks of money per customer. That’s what helps you bootstrap and make very capital efficient company-building possible.

Tarek Sherif: It actually got better than that. What we did subsequent to that is, if you’re spending a couple of hundred thousand dollars per trial, we started to bundle. Think of a debit card concept. It took us a few years to build to this level.

So we were running bundles of studies – maybe two or three studies for a customer. We kept building up our credibility. We service a lot of our customers. We didn’t always get the software right. When we conceived Medidata, we always said the mission was to build technology that was going to impact patients’ lives. Our customers like that alignment between the mission of our business and what they were trying to accomplish. That helped us build the culture, helped us attract and retain employees. It also helped with our customers.

However, building the software was complex. Scaling it was massively complicated. There was a year where we went from about 50 employees to over 150 with massive turnover. We were having all the issues you could imagine, but we survived it. In part, our customer retention has always been amazing because we had very good service.

Our customers knew that even if the technology had an issue, we were going to go fix it. We owned the problems. We never pushed them off and said, “It’s your fault.” We just owned the problem and we came back with a fix. That made a huge difference.

So getting back to the pricing model, what happened was, we would bundle a few studies together. Then we came up with the idea of instead of just having the piecemeal based on individual decisions of when they wanted to give us a trial, we’d look at their entire portfolio and said, “Fine. If you run 50 studies a year, why don’t you sign a contract with us for 150 studies that you can start over a three-year period? We’ll give you a discount but pay us either upfront or yearly upfront for the value of running those studies.”

We’d take the revenue ratably, but we got larger cash infusions either on a yearly basis or in multiple years. It was very rare that we got multiple years upfront, but usually it was a year in advance. They would consume the studies out of that agreement. It was use it or lose it.

If they got to the end of the agreement and they hadn’t used them all, they were still on the hook for paying us. But if they went over, they had to pay us more. That model became wildly successful because of the way they made a lot of decisions based on budgeting.

So when they budgeted for a clinical trial, the certainty of knowing what the software was going to cost for that individual trial became very important because this wasn’t a purchase that was done at a corporate level by the CIO. It was a decision that was made lower in the organizations within data management. So we made their budgeting process easier. They liked this model quite a bit and it worked extremely well for us.

Today, we’re close to three quarters of a billion in revenue. We have over 100 agreements in place that are worth over a million dollars each. So we write agreements that are in the tens of millions of dollars with individual customers, which is quite unusual for a company our size. It’s something we’ve done for the past decade.

Sramana Mitra: How long did it take you from launch to reach a million dollars in revenue?

Tarek Sherif: We launched in late 1999. It took us till 2002 to get to a million in revenue.

Sramana Mitra: How much financing went into that part of the company building?

Tarek Sherif: About $2.5 million.

Sramana Mitra: When is the next round of financing? When does the next round of financing begin?

Tarek Sherif: In late 2003, we decided to just go talk to some venture capitalists just because we wanted to get some feedback. We thought we’re building a great business, but we didn’t want to be sitting there with rose-colored glasses thinking we had a great business. So we weren’t really contemplating a next round of financing, but we went out.

The reception we got was so positive that we ended up doing our next round of financing. That was right before Memorial Day of 2004. Insight Venture Partners put $10 million into the company. We did $7 million of revenue in 2004. By 2009 when we went public, our total revenue was $140 million, and a large part of that was subscription. I don’t remember the exact ratio of subscription to services. Then obviously, over the last decade, we’ve gone from $140 million in revenue to almost to $740 million this year.

This segment is part 4 in the series : Taking a Capital Efficient Company Public and Beyond: Medidata CEO Tarek Sherif
1 2 3 4 5 6 7