Peter Traynor recently sat with Dave Menninger, who is the VP Global Marketing & Product Management at InforSense and discussed corporate strategy and product development at InforSense.
Peter Traynor: Dave, what’s new at InforSense?
Dave Menninger: We recently announced a number of new products, specifically VisualSense, Document Explorer, Workflow Library and Virtual Machine, which are all technology enhancements; and Translational Medicine Solution, which is a solution targeted to our core market of life sciences.
PT: Life sciences is one of your big markets, right?
DM: Yes. We do about 60-70% of our business in the life science segment. Just to give you a quick recap of the year – it was a really good year for us. We grew from 75 employees to 125 employees and, over the course of the last two years, we’ve more than doubled our revenue. So things are going pretty well.
A couple of elements of recognition I can mention are that we were just selected as a Finalist in the Red Herring 100 Europe 2008 and also named as one of the Top 25 companies in The Sunday Times Microsoft Tech Track 100.
PT: That’s excellent, congratulations.
DM: We’re really well represented in the pharmaceutical industry with 12 of the top 15 pharmaceuticals.
InforSense is focused on delivering analytics as part of the process and embedding it into an organization. Bill Gassman recently talked about the convergence of business intelligence with business process management and we fit quite well into that space.
We do, as I mentioned, real-time analysis and one of the technologies I’ll talk about in this new product set deals specifically with some scoring services and a new product for interactive visualization. Obviously there’s a lot going on in the market – a lot of consolidation. We think that really creates more opportunity for us.
One of the comments at the Keynotes at the Gartner BI Summit in Chicago was that the megavendors would be tied up with integrating their product stacks. The other comment was whether the game is over. No, the game is not over. Innovation and new vendors are coming into the market and will fill the void that the big vendors are leaving.
PT: That will be an interesting space to watch. Tell me more about your products.
DM: There are 4 new product announcements relative to the technology stack and one related to the applications. The first product is VisualSense. VisualSense is an interactive visualization product. Companies like Spotfire, Visor, Tableau and so forth are in the same space. We’ve created our own thin client visualization technology to supplement what we’ve been doing. So we’ve partnered with Spotfire for some time now but we want to be able to offer an end-to-end solution. We continue to partner with Spotfire and they’re well entrenched in the life sciences community and there will be “co-opetition”. We’ve had this conversation with Spotfire’s Chris Ahlberg that we’re not a visualization company. We’re a BI platform company.
As part of our platform, we offer data integration capabilities. It doesn’t mean that we’re a direct competitor with Informatica but, by the same token, we want to be able to offer an end-to-end solution.
Another product is Workflow Library which enables organizations to collaborate, manage and share the various analyses they create using InforSense. The analytical workflows can be posted to a central repository along with comments, documentation and versioning information. Document Explorer is a thin client navigation and visualization tool for text mining. The point here is that while the “Search” function is very well understood and very common it is not very effective. Everybody’s done the Google search where you get back a million hits – that doesn’t really help you. Document Explorer provides the capability to combine domain-specific categorization of documents with free text searches resulting in a much more precise set of search results.
PT: So news feeds, blogs, you name it – any kind of unstructured data out there?
DM: Comment fields in customer service, database, RSS, blogs and forums, HTML pages, you know, searching the web and so forth.
PT: Does it take a long time to set that up?
DM: There are two elements to it. One is identifying the feeds – what sources you want to analyze, and the other is training. So if you’re deriving the ontology, you want to do some training to eliminate noise. Some of it is parts of speech but some of it is just terms that are not really important but occur a lot.
PT: So you have to continue to teach it?
DM: Yes. You set it up and extract terms and phrases and review them over a period of time designating some as “unimportant” so that they do not rise to the top and show up as high-level characterizations of documents.
PT: Can you give some examples of where this would be applied?
DM: Well, the life sciences community does a lot of searching of patents and publications for research and knowledge management obviously makes sense with internal document repositories.
We’ve seen it applied in asset management where news releases are analyzed. It is also used in customer service applications for analyzing positive and negative sentiments.
Another application is compliance analysis where you’re trying to determine how best to deal with different claims. For example, the city of New York gets an inordinate number of liability claims filed against them every year. For a million-dollar claim, for example, the city may predict based on the documents they have associated with the claim that it’s more than likely that there will be litigation. And while they may not be able to predict with certainty that they will win, there is still enough of a chance and the value of fighting it is worth it.
PT: Can you put a number on that - like a percentage?
DM: Yes. It’s all based on numbers and percentages. So there’s a weight for the type of document and there’s a weight for the certain levels of evidence within the document.
PT: That’s really interesting. The other place I thought this might work is investor relations in large companies – finding out what the sentiment out there about the company is.
DM: When we talk about customer sentiment I use the phrase “asset management” and that’s exactly where we apply that. So fund managers, be it investor relations or the people actually managing the portfolios are using that information to get some sense of what might be happening with the investment.
PT: Earlier you had mentioned an InforSense solution targeted to the life sciences segment; tell me more about it.
DM: Translational research is a specific solution targeted at our core market segment. One of the key issues in the life science community is the dichotomy between the research side of things and the application to individual patients and these two worlds are largely separate. The providers and the clinics are the ones who have the medical research institutes so places like Dana-Farber for cancer research, Windber research which works with Walter Reed Army Medical Center, and Erasmus Medical Center in the UK are examples.
The teaching hospitals are affiliated with actual care providers. There is this trend toward taking the patient information, taking the lab information and bringing it together – and that’s referred to as translational research – so that you can provide individualized patient care. That’s the goal ultimately.
PT: So who does the translational research?
DM: Well today it’s mostly done at teaching hospitals and medical research institutes who are closer to the patient side of things. But the pharmaceuticals are also interested in translational research because what it allows them to do is develop drugs, treatments and procedures targeted at specific individuals.
PT: I can see why the teaching hospital would be in this – because they’re synthesizing what they’re getting from the research side and other side and they’re making it into a direct application for the patient. But why the pharmaceuticals?
DM: Oh, they are very interested in it, but you’re right, they’re a little further behind than the medical research institutes in terms of applying it on a practical basis.
One of the big issues has to do with HIPAA, the Health Insurance and Portability Accountability Act. You’ve got a wall, if you will, between the pharmaceuticals and the actual patient records. The information the pharmaceuticals get is cleansed, consolidated and without patient identification. So what the pharmaceuticals are doing is using that information to then provide information to the caregivers/providers on what characteristics of an individual to look for in terms of prescribing a particular treatment.
PT: I see, and InforSense has created a solution for that.
DM: It’s bringing all this different information together – genetic information about the individual, molecular information about the drug and the compounds, external public information about adverse effects or other research that’s been done and so forth.
But the interesting thing for InforSense is that we’ve created a packaged solution that puts all this together. Everybody else is doing this on a custom, consulting one-off type of basis. So we went into Dana-Farber for instance and within a couple of months had them up and running. And similarly at Erasmus, they were up and running very quickly. Whereas companies like IBM are doing this but they’re bringing in an army of IBM Global Services folks and building something and trying to do a lot of other things as well.
PT: Are you looking to partner with other data visualization companies?
DM: I think it would probably become more difficult now because when we partnered with Spotfire. We didn’t have our own visualization capabilities and they were well entrenched in the market. If you’re somebody well entrenched in the market segment, we’re happy to partner with anybody. We’ve partnered with SAS, MatLab, R, for example.
PT: Did you say R? That’s a growing segment.
DM: Do you know what really surprised us? R was one of the key integrations we did in the life sciences community and the market segment we’ve broadened into more recently is financial services – we had no idea that R is deeply embedded there as well.
PT: It is. And you get more and more students graduating who used R throughout university and they’re going to take it into their own industries and extrapolate it out there.
Now corporately, what is InforSense planning or doing organizationally or in terms of M&A?
DM: From a corporate perspective the focus has been on building out our sales and marketing organization. I think we’re somewhere around 8 to 10 sales people now and we’re looking to add another five or so during the course of the year.
PT: In North America or the UK?
DM: Both. And the sales force expansion is related to the market segment expansion. We’ve had our focus on the life science market since inception. We’re now hiring and pursuing directly the financial services market segment as well, in addition to the healthcare market segment. And we’re pursuing across industries, the sales and marketing customer-oriented analyses.
PT: Are you seeing a lot of continued investment in this space?
DM: Business intelligence? Yes. It’s interesting. As I said at the beginning, the consolidation at the top of the market I think has created more opportunity, not less, for the smaller vendors. It’s similar to how a forest fire sweeps through and creates an opportunity for lots of new growth underneath.
PT: Are you doing anything new and different with Spotfire or are you still doing more of the same?
DM: No, still the same. It was interesting because there was a little bit of concern when we came out with VisualSense – but our CEOs had a heart-to-heart conversation about it and recognized that yes, there is going to be some overlap but we’re not fundamentally in their space in the same way that we’re not in the database space. But we have the ability to save data in an InforSense file format just for purposes of making everything self contained. But it’s not our core focus.
PT: There are lots of cross pollination opportunities even if there is overlap. Dave, I thank you for your time; it was a real pleasure.