Since the time Amazon announced its Database-as-a-Service, most IT vendors, big and small, have either spun off services utilizing Amazon or built services that are complementary to Amazon. And many others have rolled out competing products. Each of them is targeting SMBs with the promise of agility, reliability, scalability and integration capability. The market has become crowded with options that are very difficult to sift through for SMBs. Nevertheless, the interest in database-as-a-service has never been higher. As per Techaisle’s 2013 SMB Cloud Computing study, between 6% and 64% (varies by employee sizes) of SMBs are either using or planning to use database-as-a service. However with an average number of formal IT staff at 4.2 and percentage of SMBs having formal IT staff varying from a low of 3% (for 1-4 employee size) to high of 97% (for 500-999 employee size) businesses the adoption is sporadic and hindered because of inherent possibility of “lights-out” situation due to outages, network latency and connectivity issues.
GenieDB, a company founded in 2011, has brought into the market a MySQL-database-as-a-service that that removes outages from the equation. Phyken Media, a video game studio for mobile platforms, developer of “Wizard Ops Tactics” is GenieDB’s marquee customer. Kunal Patel, President of Phyken Media, was faced with two huge challenges;
1/ during development of the game he had to have access to robust technology that could scale to production rapidly,
2/ be able to deploy into multiple geo locations for multiple-cross-region-player challenges without having to install and manage multiple servers.
In such a rapid-fire environment, where all the data is “hot”, Kunal and his team of developer-artists turned to GenieDB to utilize its MySQL database-as-a-service offering which he said is hugely disruptive enabling globally distributed databases where all nodes remain synchronized.
To get to know more about GenieDB we had a very detailed Q&A with Sumeet Sheokand, CTO, GenieDB. Given below is an excerpt from the interview.
Techaisle: What is the motivation behind GenieDB and in particular MySQL-as-a-Service?
Sumeet Sheokand, CTO: GenieDB is built with the aim of dramatically simplifying database management in the cloud for business critical applications. Dealing with cloud outages, network latency, redundancy, replication, tuning, etc. can be very painful and time consuming, distracting precious IT resources away from other core business areas. We want to evolve this vision to the point where you no longer need a team of DBAs to manage complex, distributed database platforms. We will either automate processes or provide a very simple interface to manage hundreds of highly available, low latency MySQL databases ready for any cloud infrastructure, anywhere in the world. It is well known that the only way of overcoming the challenges of cloud outages and network latency is to distribute copies of the database across wide geographical areas, a historically difficult problem for relational database architectures. GenieDB makes this critical functionality available to all businesses and all budget sizes, in a very easy-to-use package. In keeping with the theme of making it easy, we offer our core technology as a Database as a Service (DBaaS). This allows us to offer our users a point and click ability to deploy geographically distributed database servers with automated management and monitoring already in place.
Techaisle: What problem are you trying to solve? Especially, how can small and mid-market businesses benefit from GenieDB?
Sumeet Sheokand, CTO: We have created a geographically distributed database fabric that removes the database as a single point of failure and makes cloud-enabled MySQL database provisioning, management and monitoring an exceedingly simple affair. With our service, the small and mid-market businesses can focus on running their businesses or building their applications rather than worrying about architecting complex distribution, replication and failover systems, not to mention installing, patching and maintaining backups. All the rote tasks of using databases will be taken care of by our service.
Techaisle: Is the service a better mousetrap than others that are available in the market?
Sumeet Sheokand, CTO: The comparison between what is available in the marketplace today and GenieDB can best be described as the difference between a database-in-the-cloud vs. a Cloud Database. The existing DBaaS solutions are simply databases which have been put into the cloud utilizing a variety of scripts (i.e. "A Few Scripts-as-a-Service"). These offerings are partial solutions as they do not remove the complexity of building distributed systems in the cloud. Some only offer single locations; some only offer it in certain locations, while others don’t offer an easy path to grow the database with business traffic. Through our proprietary storage engine, GenieDB has fundamentally transformed the way a relational database functions in the cloud, removing all of these common pain points within the database layer. Businesses can choose their cloud providers, their locations and size, and know that their database will be available, responsive and grow as their business grows, with minimal effort.
Techaisle: How do you think it is different from Translattice, NuoDB, Galera?
Sumeet Sheokand, CTO: GenieDB stays true to MySQL, the most commonly used database in the world, as compared to Translattice (Postgres) or NuoDB (Not MySQL, custom). GenieDB also offers an ‘Eventually Consistent’ model that allows us to place the nodes as far apart around the world as the business needs and deliver local database performance, instead of being limited by network distance (Galera).
Techaisle: Is GenieDB’s approach similar to Google’s Spanner?
Sumeet Sheokand, CTO: GenieDB has some conceptual similarities to Google Spanner, such as a consistent Clock around the cluster. GenieDB uses software Lamport Timestamp based clock compared to Google’s hardware based clock. Even though there are similarities, GenieDB was developed independent of the Spanner effort. It was nonetheless a great validation of our vision and approach. Google Spanner is also not available as a product for businesses at large to use in their own applications. GenieDB is available today to reap the same benefits.
Techaisle: Can it be deployed only across Amazon or other cloud providers as well? Do end-users have a choice of cloud provider they feel comfortable with?
Sumeet Sheokand, CTO: GenieDB is currently available across Amazon, Rackspace and Google’s cloud services. We are actively working on integrating a handful of other popular providers, including HP, into the service. We will be happy to work with our customers to accelerate their cloud of choice for integration.
Techaisle: The future is Hybrid as we all know it, so how does it work if in a customer’s environment there are both on-premise and cloud based MySQL databases?
Sumeet Sheokand, CTO: GenieDB core software is agnostic to machine location. It just needs a Linux machine with network access. It is because of this capability, that we can build a database cluster across multiple cloud providers. Hence, we can support any Hybrid configuration, if we have access to the machines. There are business and security concerns on providing full access that would need to be figured out.
Techaisle: Why did you choose MySQL for database-as-a-service?
Sumeet Sheokand, CTO: MySQL is the ubiquitous database for modern applications. It is by far the most widely installed and used database in the cloud today and hence is the basis of our service.
Techaisle: Most of the SMBs do not have IT staff let alone database administrators. Can it still be deployed with limited technology knowledge?
Sumeet Sheokand, CTO: Deploying GenieDB does not need any technical knowledge, staff or application changes besides pointing the application to the GenieDB provided database location. It is this [SMB] specific target audience for which we designed GenieDB.
Techaisle: How do you help SMBs in understanding the technology and how the service solves their business pain points?
Sumeet Sheokand, CTO: GenieDB has extensive amount of information available at its site including FAQs, White Papers and Demonstrations. We also offer a one week, free trial of the service and support so that potential customers can experience the service before buying. We are of course happy to spend as much time as necessary to get any new client comfortable with our service.
Techaisle: For type of SMB customer is this solution most suitable?
Sumeet Sheokand, CTO: GenieDB is a Storage Engine for MySQL and as such is a general solution. Hence, GenieDB works with any application that works with MySQL today. From a business perspective, any business that is running a critical application on a single database could benefit from GenieDB, so that if one database server goes down, others are still available and the application will continue to work without any downtime. Typically we see customers that are starting a new project or that have an existing project that is anticipating growing traffic or is becoming more business critical.
Techaisle: What is a typical deployment timeframe?
Sumeet Sheokand, CTO: Nodes are spun up and the cluster made available within minutes. Most applications can be up and running against GenieDB under an hour, including data transfer. The actual duration does depend on the amount of data to transfer and the upload capacity available from the customer site.
Techaisle Take
At last count, MySQL is still the most widely used database for cloud deployments and is easily the one that SMBs use extensively. However, MySQL’s continued dominance is being questioned since it was ingested by Oracle through its Sun acquisition. Although Postgres is usually considered to be more powerful and meant for big data sets, there is yet no visible mass migration from MySQL to Postgres. The success of GenieDB and the decision to use its DBaaS is not dependent upon the debate of MySQL vs. Postgres but GenieDB’s relevance for the most-used open source database. GenieDB has created a very useful solution for SMBs but will be challenged to capture mind-share of IT consultants, service providers and the developers who are advisors to SMBs for development of applications based on multi-nodal, geographically dispersed, tightly synchronized MySQL databases. Apart from focusing on growing its customer base GenieDB will have to continue to add feature sets, capabilities and integrate with different regional cloud providers. For now, GenieDB’s MySQL-as-a-Service receives check marks for its simplified usage, ability to integrate with all MySQL applications, rapid deployment, cloud portability (not being locked-in with a single provider) and above all making the database immune to outages.
Techaisle Analyst Insights
CRM has become a core application for businesses and we have already seen that Sales Force Automation and Marketing Automation functions have been quickly incorporated along with Business Intelligence. All of these can use the same or linked tables to provide a 360 degree view of the sales and marketing process. However, today, we have finally come to a place where it should be easy enough for SMBs to plan and execute business strategy using a structured performance management system, like the Balanced Scorecard. Key Performance Indicators (KPIs) should be a standard part of the application architecture as should a meta-directory of KPIs that all applications can access. To measure the effectiveness of Sales, Marketing, Operations, and industry-specific activities, each area should have standard metrics and access to benchmark data that lets the SMB know how they are doing compared to peers, but rather than only using historical data it should be based on forward-looking objectives (leading indicators) that are tied directly or indirectly to activities designed to ultimately improve financial results. SMBs are seriously interested in measuring elusive objectives like Return on Marketing Investment, Optimal Pricing, Cost of Acquisition, Lifetime Customer Value. They want integrated applications that can not only measure these objectives but also be able to optimize effectively. This is what we call the Enterprise Performance Management (EPM).
For EPM applications to be really effective, they should be able to collect data from all applications and break into several areas; for people, productivity should be monitored through activity and results (as it already is in the new generation of SaaS applications), and effectiveness of software and equipment should be measured through algorithms that follow click paths, analyze application usage, optimize the process flow and usability of the systems. In some cases, like network optimization, filtering potential employees and ecommerce, systems should optimize themselves and human intervention should only be required when something is way outside the parameters defined by the administrator – who may increasingly be the LOB management.
With the EPM (Enterprise Performance Management) system SMBs will have a new attitude and culture that values and uses data visualization as the quickest way to gauge overall performance and specific areas of interest at a glance.
Most SMBs that have used CRM and ERP systems within the past few years are familiar with the dashboards that are available with many of these applications, either embedded or purchased separately. We believe that Dashboards will continue to evolve and be dynamic in several ways; the way they use data from subsystems like ecommerce and other real time feed sources, the way users can personalize the layout of their dashboards. Similarly, within the EPM, the actual KPIs should be dynamic and have the ability to build KPIs “on-the-fly” by calculating variables on the screen and saving the result in a meta-repository for all to use. It will have to become the norm.
While several SaaS vendors allow this kind of metric building and start the user at a dashboard, we have yet to see anything targeted to the mid-market or SMBs that connects the performance across front office, production, fulfillment and customer service. NetSuite does it to some extent almost out of the box. The market has to catch up. While this level of functionality is an excellent target, small businesses can probably get by with a good understanding of leads, opportunities, customers, invoicing, billing and customer service (or the appropriate subset) by integrating together several applications from different IT vendors. But the need for EPM is genuine and the industry has to quickly design solutions to empower SMBs with enterprise-level EPM technology at an affordable price.
In the thirty years from the time Shunu Sen posed the marketing-mix problems, I have been busy with marketing research. I tried modeling most of the studies and discovered that market research data alone is not amenable to statistical predictive modeling. Take for example, imagery. Is there a correlation between Image parameters and Purchase Intention scores? There should be. But rarely does one get more than a 0.35 correlation coefficient. Try and link awareness, imagery, intention to buy, product knowledge, brand equity, etc. to the performance of the brand in the market place and one discovers land mines, unanswered questions and inactionability.
This is where ANN steps in.
Technically ANN (Artificial Neural Networks) offers a number of advantages that statistical models do not. I will list a few of them.
- Non-linear models are a tremendous advantage to a modeler. The real world is non-linear and any linear model is a huge approximation.
- In a statistical model, the model gives the error and one can do precious little to decrease the error. In ANN one can specify the error tolerance. For example we can fit a model for 85, 90, 95 or 99% error. It requires some expertise to figure out whether there is an over fit and what is the optimum error one can accept.
- Statistical models make assumptions on distributions that are not real in the real world. ANNs make no distribution assumptions.
- Most ANN software available today do not identify the functions that are fitted. We, on the other hand, have been able to identify the functions that are fitted and how to extract the weights and build them into an algorithm.
How do we bring the differentiation?
Our biggest strength is in data integration that combines market research and economic data with transaction data into a single file. This is tricky and requires some ingenuity. We use Monte Carlo techniques to build these files and then use ANN for building the Simulation models. Optimization then becomes clear and straight forward since we do not use statistical models. Optimization using statistical modeling, which most modelers use, is a nightmare. Most of the large IT vendors and even analytics companies continue to use statistical modeling for Optimization. And therein lays the problem. Neither are these companies aware of the possibilities that ANN can provide. Most modeling is done using aggregate data, whereas we handle the data at the respondent level. The conventional modeling is macro data oriented whereas we are micro data oriented. Hence the possibilities that we can generate with micro data for modeling is huge, compared to macro data.
We have crossed the stage of theories. There are many projects that we have executed successfully that have gone on to become a must-have analytical marketing input mechanism.
Doc
Techaisle
We have covered Marketing Automation as a major topic, especially through the Techaisle SMB Marketing Automation Study conducted in US, UK, Germany and several blog posts, commenting on the rapid growth and consolidation in the market. In one fell swoop Oracle is now addressing a US$3.5 billion opportunity in the US by 2015 and US$6.0 billion opportunity globally.
Oracle’s announcement that it would acquire Eloqua for $871M, a leading Marketing Automation vendor, seems a little odd at first, mainly because majority of Eloqua’s customers are said to be using Salesforce.com as their CRM platform, and because Oracle competes with Siebel-on-Demand and recently released its’ own Fusion CRM product.
In the typical Oracle fashion of acquiring existing market leaders, i.e. Siebel Systems, PeopleSoft and JDE, Oracle snatched up another jewel for the Ellison crown, apparently valuing B2B and Enterprise-level functionality over SMB and Social Media Marketing automation that have been the focus of arch-rival Salesforce.com. It also sends a message to Redmond, whose recent acquisition of MarketingPilot seems to offer a substantial list of features and functions, but does not carry the weight of Eloqua’s brand. In the short term, it kind of looks like a mixed bag; the repercussions of the purchase seem to revolve along these areas:
SMB Customers:
1) Oracle says they will continue to support third party applications, but they have a huge vested interest in on-premise CRM in Siebel and other solutions that will compete for resources,
2) Enterprise customers who still have a staff to manage applications may be open to another level of integration with a combined Fusion/Eloqua/Siebel offer, but for those SMBs already on SFDC, it is very unlikely that there would be a compelling reason to move from SaaS to an on-premise model; getting away from capital purchases, IT headcount, maintenance fees and software upgrades was the main reason for going with Salesforce.com from the beginning. Same for those who are already managing their marketing campaigns and customer communication programs using Eloqua; it is hard to take that away without some revenue risk and employee dissatisfaction.
3) In a recent Techaisle survey, 77% of SMBs interviewed stated they were looking for vendors to reduce complexity. The type and level of integration of Eloqua into the larger Siebel suite will either make things more or less complicated depending on the approach taken by product managers to create a seamless experience.
Competitors:
4) Oracle denies access to Eloqua’s technology to Salesforce.com, which would have been a very good fit both in terms of customer acquisition/migration and start up culture. This may well be the most important (short-term) advantage gained by Oracle. This moves Marketo, SliverPop and others up the ladder as the large independents in the space, with Hubspot and Marketo obvious next-in-line M&A targets, in a market that has seen scores of start-ups, mergers and acquisitions over the past few years.
5) Despite a 30% premium paid by Oracle for Eloqua ($23/share over the market $17), there is already a class action suit alleging that the board should have shopped a buyer more aggressively, suggesting a $27 price as more reasonable. No comment.
6) Over the longer term the implications are larger in a market that is moving fast, which will be influenced increasingly by Big Data, Automation and Optimization - Enterprise capabilities to compete with the big players like IBM, Salesforce.com and SAP, who will be bringing competition and automation to a new level in the next few years. A recent Wikibon Post is a good example of how hardware and software are evolving to meet these emerging real time challenges. The post describes how fast the bar is rising in optimization in general and online advertising in particular:
“Many commercial Web publishers make space available on their Web pages for banner and display advertisements. Typically, when a user opens such a Web page, the browser reaches out to an online ad exchange network and requests an ad unit to serve to that user. The ad exchange broadcasts this information, often enriched with behavior data specific to the user in question, to multiple advertisers. Each advertiser compares the information against its internal ad inventory and existing ad campaigns to determine what that ad impression is "worth" to them. It then decides whether to place a bid and at what amount. Bids are returned to the ad exchange, which determines who the highest bidder is and delivers the winning advertisement.”
- Wikibon
All without noticeable lag to the user. These are the kind of industrial strength capabilities that are on the way as the market compounds at almost 130%, dominated by Mobile spending as devices grow by the hundreds of millions, as shown in this Kleiner-Perkins forecast.
Who’s Next?
So in our opinion this is a significant acquisition for Oracle, mainly because Eloqua has a strong base of satisfied users and a strong brand, and strong technology that can be applied to Oracle’s stack in the long run. The $871M price tag was not enough to prevent a lawsuit for some shareholders, but represents a sizable investment for Oracle as they strive to define the ultimate Customer-Centric, Multi-Channel Relationship Management platform in their race against the large horizontal vendors in the space (IBM, SAP, SAS, Google, Teradata). To this end Oracle has acquired eight companies in the last two years (ATG, FatWire. Endeca, RightNow, Inquira, Vitrue, Collective Intellect, and finally with this announcement, Eloqua). The scope of what used to be called the “360ᴼ customer perspective” has evolved to include pre-sales, sales, post-sales, customer service and lifetime customer value application components, with a relentless push to automate and integrate each piece of the puzzle.
In the wake of this acquisition, an obvious question is who's next? As mentioned earlier, Marketo is an obvious choice, as are Silverpop Hubspot, Responsys, and others. Will SFDC respond in kind or continue to focus on the lower end of the market and Social Media acquisitions? Regardless, we think 2013 will continue to see a rich market for Marketing Automation M&A activity, following two years that have seen scores of transactions in the space.
