Thursday, April 30, 2015

How Globe Testing helps startups make the leap to cloud- and mobile-first development

This latest BriefingsDirect mobile development innovation discussion examines how Globe Testing, based in Madrid, helps startups make the leap to cloud-first and mobile-first software development.

We'll explore how Globe Testing pushes the envelope on Agile development and applications development management using HP tools and platforms.

 Listen to the podcast. Find it on iTunes. Get the mobile app for iOS or Android. Read a full transcript or download a copy.

To learn more about modern software testing as a service we're joined by Jose Aracil, CEO of Globe Testing, based in the company's Berlin office. The discussion is moderated by me, Dana Gardner, Principal Analyst at Interarbor Solutions.

Here are some excerpts:

Gardner: Tell us about Globe Testing. Are you strictly a testing organization? Do you do anything else? And how long have you been in existence?

Aracil
Aracil: We're a testing organization, and our services are around the Application Development Management (ADM) portfolio for HP Software. We work with tools such as HP LoadRunner, HP Quality Center, HP Diagnostics, and so on. We've been around for four years now, although most of our employees actually come from either HP Software or, back in the day, from Mercury Interactive. So, you could say that we're the real experts in this arena.

Gardner: Jose, what are the big issues facing software developers today? Obviously, speed has always been an issue and working quality into the process from start to finish has always been important, but is there anything particularly new or pressing about today's market when it comes to software development?

Scalability is key

Aracil: Scalability is a big issue. These days, most of the cloud providers would say that they can easily scale your instances, but for startups there are some hidden costs. If you're not coding properly, if your code is not properly optimized, the app might be able to scale -- but that’s going to have a huge impact on your books.

Therefore, the return on investment (ROI) when you're looking at HP Software is very clear. You work with the toolset. You have proper services, such as Globe Testing. You optimize your applications. And that’s going to make them cheaper to run in the long term.
Reduce post-production issues by 80 percent
Download the HP white paper
Build applications that meet business requirements
There are also things such as response time. Customers are very impatient. The old rule was that websites shouldn't take more than three seconds to load, but these days it's one second. If it's not instant, you just go and look for a different website. So response time is also something that is very worrying for our customers.

Gardner: So it sounds like cloud-first. We're talking about high scale, availability, and performance, but not being able to anticipate what that high scale might be in any given time. Therefore, creating a test environment, where you can make the assumption that cloud performance is going to be required and test against it, becomes all more important.

Aracil: Definitely. You need to look at performance in two ways. The first one is before the app goes into production in your environment. You need to be able to optimize the code there and make sure that your code is working properly and that the performance is up to your standard. Then, you need to run a number of simulations to see how the application is going to scale.

You might not reach the final numbers, and obviously it's very expensive to have those staging environments. You might not want to test with large numbers of users, but at least you need to know how the app behaves whenever you increase the load by 20 percent, 50 percent, and so on.
The second aspect that you need to be looking at is when the app is in production. You can't just go into production and forget about the app.

The second aspect that you need to be looking at is when the app is in production. You can't just go into production and forget about the app. You need to carry on monitoring that app, make sure that you anticipate problems, and know about those problems before your end users call to tell you that your app is not up and running.

For both situations HP Software has different tools. You can count on HP Performance Center and HP Diagnostics when you're in preproduction in your staging environment. Once you go live, you have different toolsets such as AppPulse, for example, which can monitor your application constantly. It's available as software as a service (SaaS). So it's very well-suited for new startups that are coming out every day with very interesting pricing models.

Gardner: You're based in Berlin, and that's a hotbed of startup activity in Europe. Tell us what else is important to startups. I have to imagine that mobile and being ready to produce an application that can run in a variety of mobile environments is important, too.

Mobile is hot

Aracil: Definitely. Mobile is very hot right now in Berlin. Most of the startups we talk to are facing the same issue, which is compatibility. They all want to support every single platform available. We're not only talking about mobile and tablet devices, but we're also talking about the smart TVs and the wide array of systems that now should support the different applications that they're developing.

So being able to test on multiple operating systems and platforms and being able to automate as much as possible is very important for them. They need the tools that are very flexible and that can handle any given protocol. Again, HP Software, with things such as Unified Functional Testing (UFT), can help them.

Mobile Center, which was just released from HP Software, is also very interesting for startups and large enterprise as well, because we're seeing the same need there. Banking, for example, an industry which is usually very stable and very slow paced is also adopting mobile very quickly. Everyone wants to check their bank accounts online using their iPad, iPhone, or Android tablets and phones, and it needs to work on all of those.
Most of the startups we talk to are facing the same issue, which is compatibility. They all want to support every single platform available.

Gardner: Now going to those enterprise customers, they're concerned about mobile of course, but they're also now more-and-more concerned about DevOps and being able to tighten the relationship between their operating environment and their test and development organizations. How do some of these tools and approaches, particularly using testing as a service, come to bear on helping organizations become better at DevOps?

Aracil: DevOps is a very hot word these days. HP has come a long way. They're producing lots of innovation, especially with the latest releases. They not only need to take care of the testers like in the old days with manual testing, automation, and test management. Now, you need to make sure that whatever assets you're developing on pre-production can then be reused when you go in production.
Reduce post-production issues by 80 percent
Download the HP white paper
Build applications that meet business requirements
Just to give you an example, with HP LoadRunner, the same scripts can be run in production to make sure that the system is still up and running. That also tightens the relationship between your Dev team and your Operations team. They work together much more than they used to.

Gardner: Okay, looking increasingly at performance and testing and development in general as a service, how are these organizations, both the startups and the enterprises, adapting to that? A lot of times cloud was attractive early to developers, they could fire up environments, virtualize environments, use them, shut them down, and be flexible. But what about the testing for your organization? Do you rely on the cloud entirely and how do you see that progressing?

Aracil: To give you an example, customers want their applications tested in the same way as real users would access them, which means they are accessing them from the Internet. So it's not valid to test their applications from inside the data center. You need to use the cloud. You need to access them from multiple locations. The old testing strategy isn't valid any more.

For us, Globe Testing as a Service is very important. Right now, we're providing customers with teams that are geographically distributed. They can do things such as test automation remotely, and that can then be sent to the customers so they are tested locally, and things such as performance testing, which is run directly from the cloud in the same way as users will do.

And you can choose multiple locations, even simulating the kind of connections that these users are using. So you can simulate a 3G connection, a Wi-Fi connection, and the like.

Other trends

Gardner: I suppose other trends we're seeing are rapid iterations and microservices. The use of  application programming interfaces (APIs) is increasing. All of these, I think, are conducive to to a cloud testing environment, so that you could be rapid and bring in services. How is that working? How do you see your customers, and maybe you can provide some examples to illustrate this, working toward cloud-first, mobile-first and these more rapid innovations; even microservices?

Aracil: In the old days, most of the testing was done from an end-to-end perspective. You would run a test case that was heavily focused on the front end, and that would run the end-to-end case. These days, for these kinds of customers that you mentioned we're focusing on these services. We need to be able to develop some of the scripts before the end services are up and running, in which case things such as Service Virtualization from HP Software are very useful as well.

For example, one of our customers is Ticketmaster, a large online retailer. They sell tickets for concerts. Whenever there's a big gig happening in town, whenever one of these large bands is showing up, tickets run out extremely quickly.

Their platform goes from an average of hundreds of users a day to all of a sudden thousands of users in a very short period of time. They need to be able to scale very quickly to cope with that load. For that, we need to test from the cloud and we need to test constantly on each one of those little microservices to make sure that everything is going to scale properly. For that, HP LoadRunner is the tool that we chose.
We need to be able to develop some of the scripts before the end services are up and running.

Gardner: Do you have any examples of companies that are doing Application Development Management (ADM), that is to say more of an inclusive complete application lifecycle approach? Are they thinking about this holistically, making it a core competency for them? How does that help them? Is there an economic benefit, in addition to some of these technical benefits, when you adopt a full lifecycle approach to development, test, and deployment?

Aracil: To give you an example of economic benefit, we did a project for a very large startup, where all their systems were cloud-based. We basically used HP LoadRunner and HP Diagnostics to look at the code and try to optimize it in conjunction with their development team. By optimizing that code, they reduced the amount of cloud instances required by one-third, which means a 33 percent savings on their monthly bill. That’s straight savings, very important.

Another example is large telecommunication company in Switzerland. Sometimes we focus not only on the benefits for IT, but also the people that they are actually using those services. For example those guys that go to their retail shops to get a new iPhone or to activate a new contract.
Reduce post-production issues by 80 percent
Download the HP white paper
Build applications that meet business requirements
If the systems are not fast enough, sometimes you will see queues of people, which turns into lower sales. If you optimize those systems, that means that the agents are going to be able to process contracts much quicker. This specific example will reduce to one-fifth of the time by using Performance Center. That means that the following Christmas, queues literally disappear from all those retail shops. That turns into higher sales for the customer.

Gardner: Jose, what about the future? What is of interest to you as a HP partner? You mentioned the mobile test products and services. Is there anything else particularly of interest, or anything on the big data side that you can bring to bear on development or help developers make better use of analytics?

Big data

Aracil: There are a number of innovations that are coming out this year that  are extremely interesting to us. These are things such as HP AppPulse Mobile, StormRunner, both are new tools and they are very innovative.

When it comes to big data, I'm very excited to see the next releases in the ALM suite from HP, because I think they will make a very big use of big data, and obviously they will try to get all the information, all the data that testers are entering into the application from requirements. The predictive test and the traceability will be much better handled by this kind of big data system. I think we will need to wait a few more months, but there are some new innovations coming out in that area as well.

 Listen to the podcast. Find it on iTunes. Get the mobile app for iOS or Android. Read a full transcript or download a copy. Sponsor: HP.

You may also be interested in:

at 3:11 PM 0 comments

Tuesday, April 28, 2015

Big data detects and eradicates slavery risks from across B2B supply chains

Business networks and big data are joining to give companies new means to identify and eliminate slavery and other labor risks from across their complex global supply chains.

Data-savvy B2B participants in these networks can now newly uncover unsavory and illegal labor practices that may be hidden deep inside multi-level supplier ecosystems.

The next BriefingsDirect best business practices panel discussion focuses on how Made in a Free World, a nonprofit group in San Francisco, is partnering with Ariba, an SAP company, to shine more light across the supply chain networks to not only stem these labor practices, but also reduce the risks that companies may unwittingly incur from within their own pool of buying.

Listen to the podcast. Find it on iTunes. Get the mobile app for iOS or Android. Read a full transcript or download a copy.

To explain practical and effective approaches to forced-labor risk determination and mitigation, we're joined by Tim Minahan, Senior Vice President of Ariba, and Justin Dillon, CEO and Founder at Made In A Free World in San Francisco. The panel is moderated by me, Dana Gardner, Principal Analyst at Interarbor Solutions.

Here are some excerpts:

Dillon: I learned about this slavery issue about 10 years ago, just reading an article in The New York Times, and really wanted to figure out a way to get more people involved. That led to Made In A Free World, a nonprofit organization that was started just few years ago.

Dillon
We began with consumers, just everyday individuals getting involved and being able to leverage their own network, their personal networks. That's now graduated into figuring out how to get businesses to leverage their own networks.

Gardner: What is the problem we are talking about? Is this just slavery? Is this breaking the law in terms of other labor regulations?

Dillon: We define slavery as anyone who is being forced to work without pay under threat of violence, being economically exploited and unable to walk away. There are over 30 million people who fall under that definition today. In some cases, these people find themselves in the sex industry, but in most cases, they're in informal sectors, agricultural or service industries, much of which is finding its way into supply chains.

Part of what we believe is that we have to find a way to be able to connect the dots and figure out how we can use the systems that currently exist to protect the world's most vulnerable resource.

Gardner: Tim, why is this becoming such an important issue now for nearly any business?

Minahan: Over the past decade, as companies begin to outsource more processes and manufacturing and assembly to low-cost regions, they've really looked to drive costs down. Unfortunately, what they haven't done is really take a close look at their sub-tier supply chain.

Minahan
So they might have outsourced a process, but they didn't outsource accountability for the fact that there may be forced labor in their suppliers' suppliers. That's a real issue, getting that transparency into that problem, understanding whether there's a potential risk for the threat of forced labor in a sub-tier supply chain.

Gardner: How big of a problem is this in terms of scope, depth and prevalence? Is this something that happens from time to time in rare instances or is this perhaps something that's quite a bit more common than most people think?

Minahan: Dana, this is far more pervasive than most people think. Slavery really has no boundaries. There are incidences of forced labor in all industries, from conflict minerals in the Congo to fishing in Malaysia to, unfortunately, migrant workers right here in the United States.

Palm oil problem

Dillon: It's really not a problem that “exists over there." Just one statistic alone: By 2020, half of all the products you can purchase in a grocery store will have a commodity, palm oil, which has a huge incidence of forced labor, particularly around the Malaysia region and Indonesia.

This small commodity, mostly because it's so easy to pick and harvest, is now finding its way into a myriad of products -- and that's just in agriculture.

Gardner: Why should companies be concerned? Isn’t that someone else's issue, if it’s below the level of their immediate reach?

Minahan: You can certainly outsource process or manufacturing, but you can’t really outsource accountability. Secondly, there is a big movement afoot from regulators, both here in the United States, Federal laws, California state laws, as well as overseas in the UK to hold companies accountable, not just for their first-tier supply chain, but for their sub-tier supply chain.

Gardner: And, Justin, of course visibility being so prominent now -- a camera around every corner, social media -- people can easily react and irreparable damage to a company's reputation. Do you have any examples of where this has come to bear, where not knowing what's going on within your supply chain can be such an issue economically and otherwise?
You can certainly outsource process or manufacturing, but you can’t really outsource accountability.

Dillon: Well, people love to hate the ones they love. So everyone is complaining about Apple products, while they're using their Apple iPhone. But in a recent article in The New York Times, one of the supply-chain folks for Apple said that the days of sloppy globalization is over and they're taking this quite seriously. I would argue that some Apple's work is some of the most innovative human rights work that we've seen.

They've dug deeper down into the sub-tier suppliers and, if their reputation as innovators has anything to do with their products, it certainly has to do with their supply chain as well. They are a great example of a company who sees these challenges, sees the connection to not only their brand, but also to their products and they're taking some remediation work against it.

Gardner: Okay, we understand the depth of the issue and why it’s important. Now, now how do we do get the tools to combat this effectively? What have you done? I understand you have a database, and it’s ongoing. Tell me a little bit about the tools that you’re bringing to bear to solve this problem and to help companies take better accountability.

Dillon: The most important thing to do first is to realize that we aren't operating in the 1990s anymore, in the sweatshop era. That was an era that where we found problems in supply chains and we started to build solutions around that, auditing, monitoring, all the rest of that. That was 25 years ago.

Era of big data

We're now in the 21st Century, in the era of big data, and we need to be able to use those tools to be able to combat 21st Century problems. For us as an organization, we've realized this was a space where we could be helpful to business. Our mission is to use the free market to free people. That’s what we do as an organization.

We've realized that there is a lot of data missing, and there is a lot of synthesis of data missing. So we, as an organization, decided to pull together the bible of databases, when it comes to supply chains, the UNSPSC. Then, we built off of that taxonomy a risk analysis on every single thing good, service, or commodity that can be bought or sold. That becomes the lingua franca, so to speak, when it comes to supply chain, risk management.

Gardner: How does it work? When a company comes in and gives you information about who their suppliers are, how do you come up with inference, insight and some sort of a predictive capability that identifies the risk or the probability that they could be in trouble?

Dillon: We are using all the best databases that currently exist on the issue. Everything from forced-labor databases to child-labor databases to rule-of-law, governance, migration, trade flows. All of that is synthesized into an algorithm that can be applied to any individual’s spend data. You're able to get a dashboard on all that spend, which gives you some optics into your sub-tier suppliers, which is where we need the optics. It’s not a crystal ball, but it’s the next best thing.

That database and analysis are now available to anyone, any size, any sector, to leverage their influence to the extent that they can. We've recognized that you can't be everywhere at all times but you’re somewhere at some times and that's the place we feel like any company can make the greatest influence.

Gardner: Now, Tim, the labor issues that we’re talking about are certainly a risk issue, but at the same time, we’re looking to find more ways to automate and make visible other types of risk, any kind of risk, in the supply chain in procurement, buy-and-sell environment.
That database and analysis are now available to anyone, any size, any sector, to leverage their influence to the extent that they can.

Tell us a little bit about how this particular risk fits into a larger risk category, and then how the databases for each of them, or many of them, come together for a whole greater than the sum of the parts.

Minahan: As Justin said, the information here is shining a light on a particular issue, being able to mass, for the very first time, data points from hundreds of different data sources around the world to predict and project the threat of forced labor.

Extending that further, that's one risk indicator that companies need to manage. Companies are managing a whole host of sustainability issues around social and eco-responsibility, but also financial risk, and threat of disruption risk.

Being able to pull all those together into a common risk factor is what we're attempting to do through the power of business networks. By connecting the world’s businesses and connecting their supply chains and automating their processes, that was phase one.

Phase two is harnessing all the information from all of those interactions to provide a better level of visibility that raises that transparency for everyone. Then, you can predict future risk. Being able to tap into what Made In A Free World has done in this particular area, pull that into the network, and expose that as another risk factor that companies can evaluate is a very powerful opportunity for us together.

Gardner: As we've seen in other aspects of the networked economy, the better the data, the more influence, the more action as a result of that data, that encourages more people to see value and add more data back into the pool and so on and so forth. Tell us about the news you made in April at the Ariba LIVE conference and how that works into this notion of a virtuous adoption cycle.

New partnership

Minahan: Ariba and Made In A Free World announced a new partnership to bring the power of the Made In A Free World database together for predictive analytics of forced labor with the power of the Ariba toolset and the Ariba Network to help the Global 2000 and beyond be able to have the transparency they need to identify potential risk, in this case, of forced labor and their supply chain, and be able to take action and rectify it.

Gardner: Justin, how do you see this announcement benefiting your cause?

Dillon: Well, it moves us past the point of saying there is nothing that you can do as a company. That is no longer an option on the table. What you're going to do is the next question. We have removed the word "if" in this conversation -- and we couldn’t have done that without Ariba.

And we can see how the Ariba Network is going to be helpful, because companies are coming to us and coming at the federal government saying, "What are we supposed to do? How do we move forward? This seems like a huge problem." They're right, and the solution is right in front of them.

Gardner: Tim, for organizations that are now more aware of this problem, and of the general risk issue across supply chains, how do they start? Where do they go to begin the process for getting better control that allows them to have better accountability?

Minahan: That’s exactly what Ariba and Made In A Free World are trying to do --  provide the tools necessary for companies to get started. We mentioned together the technical solution that we 're bringing to bear to allow folks within the Ariba community to be able to access the Made In A Free World input and be able to project potential issues of forced labor in their supply chain.

Together, through that effort, we've developed a playbook that provides suggested guidelines for folks to get started. You can’t fix what you don’t know about. You can’t improve without having some understanding of where you fit in the mix. We believe that data and doing an analysis on the freedom platform is the best way for a company to get started.

We think it’s the best synthesis of practices, data, influence, and the network effect that anyone can begin to take. We suggest that they start to look at their own risk based on what they are buying and based on their own exposure. Every company that comes to us, we're able to do this risk analysis with them, and they're are getting new insights.

In addition, there is a whole ecosystem of companies that are really looking at this issue hard. The thing that’s most exciting about it is that everyone is very transparent and they're willing to share. So you have companies from Patagonia to Apple and the like that are sharing their practices freely, because this is an issue that we want to address. The business world has the opportunity to address probably the most serious issue facing us today, and that is modern slavery.

Learning more

We have every expectation that our partnership with Ariba is going to improve this tool, not just for us, but for the planet.
Gardner: It strikes me that this is a game of numbers. If enough companies examine their supply chain sufficiently and then eradicate the areas where there is trouble, the almighty Dollar, Peso, or what currency it may be, comes to bear, and things like corruption don’t work and bad labor practices don’t get supported. Is this a rolling-thunder sort of thing. If so, what’s the timeframe? Is this something that can be solved in a fairly short amount of time if people actually come together and work on it?

Dillon: To me, it’s the greatest story ever told. This is the thing that we all appreciate, and frankly we all benefit from it. We're all benefiting from a free market and the way that we look at change in our organization is that we say that change is more about judo than karate. How can we use the force against itself to actually change it, and the marketplace is the way to do that.

So yes, this can move quite quickly. In my opinion, it's much quicker than governments can move. We think that they're going to be followers in this case and we highly expect them to be followers.

Gardner: Tim, if companies do this right, if they leverage big data, if they examine risk properly, what do they get, other than of course the direct reduction of that particular risk? It strikes me that we are not just tackling risk, but we're putting into place systematic ability to manage risk over time, which is way more powerful. Is that what we're going to get?

Looking at risk

Minahan: Absolutely. Companies are looking at risk holistically, and this is one key vital input into that. As they continue to address that and, as Justin said, we're using free markets as a powerful lever to free people. They can, through their buying patterns and their buying policies, begin to adjust the market, shine a light not just on their own supply chain but change the sub-tier supply chain practices to make sure that there is fair labor. It becomes unappealing to have forced labor and it becomes a detriment to their ability to win new business by doing that.

That's really what the power of data and the power of business networks can deliver in this scenario.

Dillon: Businesses are the heroes in the story. It's not the non-profit, and it’s not government. Anytime you associate business with human rights, businesses are often qualified as Darth Vader who actually think that they are Luke Skywalker in the story. And for us an Ariba, I don’t need to put that on them, but we are just Yoda.

We're giving them the tools that they need to fight the evil empire and we are going to do it together and celebrate it together. But this is one thing that we'll have to do in concert. We all have individual roles to play, and business has a very clear, distinct role to play.

Listen to the podcast. Find it on iTunes. Get the mobile app for iOS or Android. Read a full transcript or download a copy. Sponsor: Ariba, an SAP company.

You may also be interested in:

at 1:09 PM 0 comments

Wednesday, April 22, 2015

ECommerce portal Avito uses big data to master rapid fraud detection

This BriefingsDirect big data innovation discussion examines how Avito, a Russian eCommerce site and portal, uses big data technology to improve fraud detection, as well as better understand how their users adapt to new advertising approaches.

Listen to the podcast. Find it on iTunes. Get the mobile app for iOS or Android. Read a full transcript or download a copy.

To learn more about how big data offers new insights to the eCommerce portal user experience, BriefingsDirect sat down with Nikolay Golov, Chief Data Warehousing Architect at Avito in Moscow. The discussion is moderated by me, Dana Gardner, Principal Analyst at Interarbor Solutions.

Here are some excerpts:

Gardner: It sounds like Avito is the Craigslist of Russia. Tell us a little bit about your site and your business.

Golov: Yes, Avito is a Russian Craigslist. It's a big site and also the biggest search engine for some goods. We at Avito have more searches, for example, from iPhones than Google or Yandex. Yandex is a Russian Google.

Gardner: Does Avito cover all types of goods, services, business-to-business commerce?
Become a member of MyVertica
Register now
And gain access to the Free HP Vertica Community Edition.
Golov: You can sell almost anything that can be bought in the market on Avito. You can sell cars, you can sell houses, or rent them, for example. You can even find boats or business jets. We right now have about three business jets listed.

Size advantage

The main advantages of Avito is, firstly, its size. Everybody in Russia knows that if you want to buy or sell something, the best place for it is Avito. It’s first.

Golov
Second is speed. It is very easy to use it. We have a very easy interface. So we must keep these two advantages.

But there are also some people who want to use Avito to sell weapons, drugs, and prohibited medicines. It's absolutely critical for Avito to keep it all clean, to prevent such items from appearing in the queries of our visitors.

We're growing very fast, and if we use moderators we'll have to increase our expense on moderation in a linear progressions as we grow. So, the only solution to avoid a linear increase in expenses is to use automation.

Gardner: In order to rapidly decide which should or should not be appearing on your site, you’ve decided to use a data warehouse that provides a streaming real-time data automation effect. What your requirements are for that technology?

Golov: We need to be able to perform fast fraud detection. The warehouse has to have very little delay.
Our data warehouse has to be big. It has to store months, possibly years, of data.

Second, we have to have data for long periods of time to learn our data mining algorithms -- to create reports, and to analyze trends. So our data warehouse has to be very big. It has to store months, possibly years, of data. It has to be fast, or only slightly delayed, and it has to be big.

Third, we're developing very quickly. We're adding some new services, and we're integrating with partners. Not long ago, for example, we added information from Google AdWords to optimize banners. So the warehouse must be very flexible. It must be able to grow in all three ways.

Gardner: How long have you been using HP Vertica and how did you come to choose that particular platform?

Golov: Well over a year now. We chose Vertica for two two main advantages. First, speed of load and data. The I/O speed provided by Vertica is awesome.
Become a member of MyVertica
Register now
And gain access to the Free HP Vertica Community Edition.
Second is its ability to upgrade, thanks to the commodity hardware. So if you have some new requirements that require you to increase performance, you can just buy new hardware -- commodity hardware -- and its power just increases.

It’s great and it can be done really fast. Vertica was the winner.

Measuring the impact

Gardner: Do you have a sense of what the performance and characteristics of Vertica and your data warehouse have gotten for you? Do you have a sense of reduced fraud by X percent or better analytics that have given you a business advantage of some sort? Are there ways to measure the true impact?

Golov: Avito grew really fast over the past year. We have a moderation team of about 250 people at the beginning of this process. Now, we have the same moderation team, but the number of items has increased two-fold. I suppose that's one of the best measures that can be used.

Gardner: Fair enough. Now, looking to the future, when you're working in a business where your margins, your business, your revenue comes from the ability to provide advertisement placements, improving the performance and the value on the actual distribution of ads and the costs associated is critical.

In addition to rapid fraud detection and protection, is there a value from your analytics that refines the business algorithms and therefore the retail value to your customers?
We're starting few more products. The main aim of them is to create our own tool for optimizing the directions of advertising.

Golov: We're creating more products. The main aim of them is to create our own tool for optimizing the directions of advertising. We have banners, marketing campaigns, and SMS. So we've achieved some results in our reporting and in fraud prevention. We'll continue to work in that direction, and we are planning to add some new types of functionality to our data warehouse.

Gardner: It certainly seems that a data warehouse delivers a tactical benefit but then over time moves to a strategic benefit. The more data, inference, and understanding you have of your processes, the more powerful you can become as a total business.

Golov: Yes. One of my teachers in data warehouses explained the role of data warehouses in an enterprise. It’s like a diesel engine inside a ship. It just works, works, and works, and it’s hot around it. You can create various tools to increase it, to make it better.

But there must always be something deep inside that continuously provides all of the associated tools with power and strong data services from all sides of the business.
Become a member of MyVertica
Register now
And gain access to the Free HP Vertica Community Edition.
Gardner: I wonder for others who are listening to you and saying, "We really need to have that core platform in order to build out these other values over time." Do you have any lessons that you have learned that you might share. That is to say, if you're starting out to develop your own data warehouse and your own business intelligence and analytics capabilities, do you have any advice?

Be flexible

Golov: First, you have to be flexible. If you ask a business about changing, they'll tell you that they can’t. It will be absolutely this, every time. And in two months, it will still change. If you're not ready to change using your data warehouse to get needed data and analytics, it would be a disaster. That's first.

Second, there always will be errors in data, there will be gaps, and it's absolutely critical to start building a data warehouse together with an automated data quality system that will automatically control and monitor the quality of all the data. This will help you to see the problems when they occur.

Listen to the podcast. Find it on iTunes. Get the mobile app for iOS or Android. Read a full transcript or download a copy. Sponsor: HP.

You may also be interested in:

at 3:18 PM 0 comments
Subscribe to: Posts (Atom)