In this episode of the ITProPortal podcast we meet a company with a unique name and a unique take on big data and machine data. James Murray, Vice President and General Manager for EMEA gives us a company spotlight on Splunk.
For more related podcasts click here.
To subscribe to receive new podcast episodes for free click here.
Firstly, Splunk is an awesome name where did it come from?
Splunk comes from Speleology, which is caving rooting around in caves with a head torch on looking for stalactites and stalagmites and generally finding things underground. When our founders set up Splunk they were rooting around in the logs of computers trying to understand why a website had crashed and getting data from different sources. They likened that to ferreting around in a cave so the name came from speleology in America it is called spelunking and we shortened that to Splunk.
Splunk is known as a provider of services around big data so, let’s begin by defining what you do in that area?
Well big data obviously does lots of it and by lots of it we can talk in terms of terabits so if you wanted to visualise what a terabit is, it’s about 600 hours of high quality audio recording. If we had to analyse that amount of data to find issues and challenges within it as part of our daily business operation then we would have a pretty tough job. So ‘Big Data’ is all about whether you can handle that quantity of data coming at you at fast arrival rates in order to protect the stock price of the company and optimise your revenue.
Tell us a little bit more about the type of clients you have and how your solutions have been designed to benefit them?
Well if you look at the pattern of our customer base there is a lot of financial services, retail and Telecommunications companies and I think that is largely down to the fact that they have huge customer bases and mission critical websites where they are transacting B to C. They have to know on a minute by minute, second by second basis what is going on within their infrastructure in order to detect security threats, to see that there is a performance problem with the website or to discover that there is a new customer segment emerging. Those three areas financial, Telecommunication and retail compete on the basis of service to their customers and they have to be spot on everyday within that real time web environment.
When you go into an organisation to help them out what do you start by doing?
The main thrust of what we do is analysing machine data. Machine data simply put, is the logs that every computer program spits out. It’s the trail that it leaves behind like exhaust fumes and by analysing those exhaust fumes you can see everything that is going on within a business. A program that runs an application that runs a website is recording every customer/ entry of data, user names, products, purchase and also time that is taken to move from one page step to another. So by analysing all of that stuff you can discover what is going on in your infrastructure in real time. It is real time because you are taking it direct from the machine as opposed to waiting for it to be aggregated and inserted into a data base and then analysed a couple of weeks later. So main difference between what we are doing and what we call operational intelligence is that we are looking at what is happening right now rather than the historical sort of business intelligence where you are more sort of looking at rear view mirror at what happened yesterday and last week.
Have you seen the importance of big data initiatives growing?
Every major company has got some kind of big data initiative going on but not every single company is clear on what it is going to get from it. What we are doing is focusing with our customers on specific applications. An obvious big data application is security which is not a discreet area of its own, it’s in the website, it’s in your ERP application and it is across every part of your infrastructure. So big data is critically important and when you look at the volume of information coming at you in terabits and the need to keep on top of that on a daily basis it is increasingly important.
What do you think has been the biggest change or development over the last 12 months for you? I guess mobile has had a significant impact?
Yes. Mobile has had a massive impact as has the Internet of things. There is this new type of machine data, which is all about transaction information, that you can suddenly harness and then correlate with your historical information data bases and also with unstructured human information. So to bring that to life, it is now possible for us to correlate what is happening on a website in terms of the percentage decrease in visits with the call centre stats that show an increase in calls coming through and then correlate that with the performance of the middle ware layer to see if there is a problem in supporting the website that is causing customers to go from the website to the call centre. Then in doing that it means that you can set up alerts so that rather than wait for a negative customer feedback a few days later you can instantly trigger a check to be made by IT to look at the middle ware layer and ensure that the website is back on track. It is really a new way of looking at the world as it’s a much timelier source of information than has ever been available to us.
Well your main client base includes some very large companies but what is applicable to smaller companies when talking about big data. Are there small companies that have big data to consider?
We have a lot of companies that are smaller in size actually. For instance, if you look at the betting market, those companies are not huge in terms of the number of employees or their geographic spread or even their revenue but they do have a big data problem. If you look at your average soccer match, it might have 100-200 different markets on it from first throw in, to first corner, first free kick, first goal and so on. When you multiply that out by the number of customers then you very quickly get to big data issue.
Has the global recession do you think opened up any opportunities for you at Splunk or has it been a hindrance?
I think what it has brought into focus more and more is the importance of the web as way of doing business and a way of grabbing customers from your competition. The one area of growth even through recessionary times seems to be the e-commerce area where more and more funding is going in to grab the competitions customers and to grow that one part of the business that does seem to be more recession proof. More people go to the web when they have got less money to spend rather than the High Street. We very much capitalise on that because if you are competing on the effectiveness of your website then you need to know that your infrastructure is behaving in the right way, you have got security threats covered and also that you are analysing the customer order patterns and how they are experiencing the website. That throws up a big opportunity for us and typically in a recession the other thing you see is the spending on cost control and security those two areas are very strong for Splunk. So, not only are we able to help companies generate more from their websites but also keep their infrastructure secure and take costs out of the operational inefficiencies in the business.
You will see from our results in the last year we have grown by 64% year over year and we are now just a shade over 200 million in revenue and we IPO ‘d last year and our stock price had the first day pop of 109% which was only matched in recent times by LinkedIn and unlike other high profile stocks we have not only maintained the level of our stock price but have actually increased it in the last year. Regardless of the recession we have done extremely well. I think it is a testament to the value that Splunk brings to businesses that needs a response real time.
So what are your focuses for the year ahead at Splunk in terms of growing the business and improving your services?
I think it is to capitalise on our position as first mover in machine data. We have also been recognised by the market as being the first true mover in big data. In the last year we have moved from a position of knowing that we held a lot of value to the security professional to actually being voted as number one provider worldwide. The opportunities are vast and I think what we need to do is get the message out particularly in EMEA we have 1300 customers out of the 5000 plus customers worldwide.
We have to focus on continuing to give good value to those customers by helping them to achieve the operation and intelligence vision. This involves creating a platform upon which any aspect of the business, sales, finance, marketing or logistics can very quickly (and I mean within 15 – 29 minutes) build a dashboard that starts to show insights that they never had before. That is a huge opportunity for us just within our 1300 customer base and then going into markets such as security where typically our presence has been more modest we see an opportunity to clean up and become the dominant player. We have a different view of the world in security to all of the other players and what we are doing is bringing all the information into one place. So, for a security professional that is very powerful because all they have to do is ask the right questions and the answers appear. That is very different from historic seam solutions that have to narrow down the volume of data because they are not built to withstand big data. With our map reduced technologies and our scalable solution we should clean up on the security market but then look to become a platform across our customers for all of their real time operations and intelligence needs.
To finish us off then by looking into your crystal ball, how do you see the security industry changing over the coming years?
Well back in the 1950s Alan Turing posed the question; “Can machines think” and if you think of the Terminator etc then it would definitely be useful to know what the machines thoughts were and those thoughts would be stored in logs. I think more and more as awareness of the value of machine data gets out there we will see that companies are looking to organisations and especially Splunk to harness the value of that data. Just think about the potential of a sensor on a car park at a major retailer – if you are taking data off that sensor about how many cars are moving through, you could correlate that with last year’s data and the revenue that was gained on that day and forecast how much you are going to sell on that particular trading day in advance. So that at 8.30am you already know what your sales are going to be at 5.30pm. This is all about taking data from machines. When you look around you at the machines that we use it’s your car, your mobile phone. So the more machines out there with sensors on them generating machine data the more there will be a need for Splunk. I think looking into my crystal ball it would be that light bulb going on across major organisations around the world that will create our biggest opportunity.