Data modeling and reporting with Quantrix

Posted on: 18 Jan 2010

Categories: Data modeling, Financial modeling, and Interviews

Recently I upgraded a really great piece of software that I use to do data modeling, analysis, and some reporting. The software is called Quantrix Modeler and I have owned it since 2003. I use Quantrix when I want to get a clearer picture of what data means. For instance, when I was getting ready to launch my business a few years ago, I used Quantrix to develop a model against which I could pose what-if scenarios to better understand cashflow issues and to do tax planning. Another time, I used it to help a family member make a decision about which car to purchase. In both cases, and many others, Quantrix has been a joy to use and helpful in facilitating good decision making.

My motivation for using Quantrix is not limited to my internal operations though. One of the lines of business that I engage in is ERP implementation and software integration. My clients need reports to help them run their businesses and Quantrix Modeler is a powerful tool for many reporting and discovery scenarios.

Being a software designer/developer myself, I am always curious to get more information about projects that deliver. I started thinking that it would be really great to talk with the people at Quantrix to learn more about them and their product, especially since I recall reading an interesting story about where the Quantrix technology came from. I want to know more. I contacted the nice folks at Quantrix and they agreed to do an interview. As a treat for my readers, I present the interview with David Phillips from Quantrix.

So, without further introduction, let’s get to the interview:

Lang: David, when I was first considering buying Quantrix back in 2003, I did a lot of research on multidimensional modeling technology, tried demos, etc., and Quantrix ended up being the clear winner for me. During that due diligence period I remember reading an interesting story about Quantrix’s origin. Can you share with us briefly about where the technology came from and how Quantrix the company came to be?

David: Hi Lang. First, congratulations on your new website and blog. Quantrix was founded in 2002 by Peter Murray and Chris Houle. Pete formerly led the development team at Lighthouse Design, which had developed a “Quantrix-like” product for NextStep and OpenStep. In fact, that product was also named Quantrix. Lighthouse was acquired by Sun Microsystems, who decided to discontinue the Quantrix project. Pete opted out of moving to Sun and, along with Chris, former CEO of ERP company Scala, started a company based on the belief that there was a need in the marketplace for a better alternative to spreadsheets when it came to more advanced financial modeling and analytics.

Lang: It seems like a strong suit of Quantrix the company is education about how to use the product in the form of webinars, screencasts, demo models, and customer support. I know I have had a very favorable experience with email support with respect to the very few questions I have had. The product is a sophisticated piece of software, but because it is usable in so many domains and because Quantrix conducts conferences, training, and modeling that cater to a wide range of client needs, you must have a wide range of expertise or at least deep modeling skills represented on the Quantrix team. I think my readers would be interested to know what areas of speciality are represented in the Quantrix team?

David: We have invested a lot of time and effort into developing self-learning materials for our customers. Your readers might be surprised to learn that over half of our customers have learned to use Quantrix on their own. Many of our larger corporate clients opt for hands-on training which they later supplement with our online materials.

In addition to learning materials, we feel that we differentiate from other software vendors by the quality and responsiveness of customer support. We encourage our customers to send in their models to get expert advice on how to optimize structure and formulas. Try sending your spreadsheet to “you know who” and see what happens. We also make use of technology such as online chat and web demo software. When people have issues or need some advice, they want answers quickly. Quantrix culture drives us to provide a level of support that is second to none.

As you alluded to, Quantrix is a very flexible tool that can be applied to all sorts of domains and business challenges. Our professional services and support teams have seen some very complex models in financial services, retail, higher education, manufacturing, retail and just about any other vertical you can think of. This breadth of experience enables them to apply “lessons learned” across domains. So, to date we really haven’t been stumped. However, there are times when some really specialized knowledge is the most efficient path forward. In that case, we often will bring in our partners who have expertise in a specific area.

Lang: One of the things that makes Quantrix a winner for me is that, like a lot of the tools that I use, it doesn’t get in my way too often. A lot of the things that Quantrix can do, a traditional spreadsheet can do, but not with the same ease and certainly not with the same elegance and robustness. Being able to write functions and formulas that are dimension specific is a huge win for conceptual integrity of the model, but also for ease of maintenance. Can you talk about those advantages and maybe give an example of a customer situation where they gained from adopting Quantrix Modeler in place of say their old spreadsheet approach?

David: Sure. A lot of our customers come to us out of frustration with spreadsheets. While many people focus on the “time-to-answer”, which is often significantly quicker with Quantrix than spreadsheets, much of the real return on investment comes with the major reduction in maintenance of the model over time. In fact, we’ve explored that issue with a whitepaper called The True Cost of Spreadsheets. We also have an ROI calculator on our website that takes several factors into consideration. With Quantrix, changes or updates to the model can occur very quickly. New data can be brought in via DataLink with a click of the mouse button. The model expands or contracts automatically without the user having to worry about new rows or columns. Because Quantrix employs plain language formulas, it is much more transparent, thus making it much easier to ensure the integrity and accuracy of the model. Our manager of professional services, Steve Bailey, likes to tell the story of the time he was at a meeting with a Fortune 100 food manufacturer. He and the client had developed a manufacturing model that included a “bill of materials” for all the possible ingredient paths that could go into production batches. The model would calculate cost of materials, labor requirements, batch outputs, etc. While they were walking through the model with the extended group, one of the team members who was not involved in creating the model was able to point out a number that didn’t seem to fit. Upon taking a closer look, it was determined that a formula needed to be modified. Steve made the change “on the fly” and the model recalculated to display the correct value. You simply won’t find a story like that with spreadsheets, in most cases, just the opposite.

Lang: As I said earlier, I have been using Quantrix Modeler Pro since 2003 on the PC and later on the Mac. I recently upgraded to Quantrix Modeler Pro 3.6 with DataNAV and DataPUSH on Snow Leopard, one of the first things I notice is how improved the user interface is. Can you explain to our readers some of the features that are available now that increase usability of the software?

David: As an early adopter of our software, you’ve seen the evolution of our technology. Version 3 contains several enhancements to the end user experience. Mind you, most of the improvements are based on customer feedback, which is vitally important to us. First of all we created a tabbed user interface which allows the user to arrange matrices, chart views and presentation canvasses on their screen in a way that is intuitive and productive. They can save these “perspectives” and call them up whenever they like. We’ve also developed a set of user roles and permissions which allow model authors to securely share their models with colleagues and clients. As your readers probably know, when you share a spreadsheet you do so at your peril. User roles and permission control the interaction of users with model elements. They also enable the author to hide elements of the model. Obviously, its easier to navigate and interact when there are fewer options on the screen. We also created a very innovative feature called the Presentation Canvas. We chose the term canvas because Quantrix allows the user to be creative when designing dashboards and reports. It’s a drag and drop feature that enables the user to combine various elements of a model on a single or set of canvasses. It also comes with a full set of widgets that enable interaction with, and visualization of, the data.

Lang: I have to say, that as a creator of software solutions, I am excited by the fact that Quantrix has an API that I can leverage to build tools for clients and for my own business that use your core engine. Can you talk about improvements in the API and about what kind of access the API provides?

David: We currently have two flavors of the API: the Standard API (QAPI) is included with Professional Edition and the Quantrix Development Kit which offers significantly greater capabilities. QAPI enables users with java programming skills to build plug-ins and extensions. Some examples that come to mind are a stock market ticker that gets real-time price quotes, a vocabulary manager for working with models in multiple languages, and a spreadsheet pivot table converter. The Quantrix Development Kit is designed for more intensive enterprise use. For example, IDBS, a UK-based life sciences software company, has built an electronic laboratory notebook for discovery research with Quantrix at the core. This application is used by thousands of researchers in several of the leading pharmaceutical companies. QDK enables integration with plug-ins and servers. It also allows Quantrix to be embedded into other applications. Other features include bulk loading of data, improved calculation performance, full access to the permission system, and the exposing of extension points.

Lang: After becoming a customer of Quantrix I elected to receive email updates from Quantrix. I noticed that Quantrix hosts an event based around the product. I almost went last year to the one in Europe, but my schedule ended up being too full. I hope to possibly attend one of these events. There seems to be quite a few companies investing in both the product and attendance at the event. Can you tell us about the ecosystem surrounding Quantrix Modeler and its API and specifically what kind of opportunities exist for other businesses to be a part of that ecosystem?

David: Yes, we host an event called the Seminar by the Sea. We hold one event in the Fall in Maine and the other in a city in Europe, which we’ve been rotating. We’re hoping to host a Seminar in Asia as well. It’s really more of a customer conference than a seminar as we have many participants who make it a yearly event (some attend both events). We also have many partners attending and presenting the innovative ways their customers are using Quantrix. In addition to building skills, the collegial nature of the event also seeds a lot of new ideas which participants take back to their organizations. We have a variety of sessions for users of different skill levels and include a technical session focused on QDK. It’s not all work though. We always host a fun event in the evening, usually involving water. In Maine, we take a cruise in Casco Bay and have a traditional lobster bake on an island. Last year’s seminar in Amsterdam feature a boat cruise in the canals.

Lang: I upgraded to Quantrix Modeler Professional 3.6 for three reasons. I wanted the improved user interface experience and I wanted access to two new technologies: DataNAV and DataPUSH. I do a lot of work with servers that reside on machines on other networks. I often have need, for my own businesses and for clients, to connect to databases and do things with the data there. Typically this is building reports of one kind or another, but often it is getting insight into what the data means. Can you explain what DataNAV and DataPUSH are and what kind of advantage they afford?

David: Sure, both are add-ons to our flagship product, Quantrix Modeler. DataNAV is a substantial advancement for Data Navigation, Analysis, and Visualization (DataNAV), enabling both business and IT professionals to easily query data sources to develop ad hoc, exploratory, as well as structured analyses. DataNAV allows our customers to improve upon a process that has traditionally been filled by the pairing of BI tools and spreadsheets. This process hasn’t been particularly efficient because it has required IT input and programming skills. DataNAV empowers the business user to create analyses that previously required programming skills.

DataPush, on the other, enables users to publish data from models into databases, data warehouses, and other business applications. Whereas the model and analysis was the “end of the road”, now users can complete the “analytics loop”. A real-world example of this would be a utility customer which uses DataPush to push its forecast and variance analysis to a centralized database, where it can be accessed by other workgroups and applications.

Lang: Whenever I find a tool that is quite different than other tools that are often used to do the same job, I observe that it leads to different insights and fresh perspectives. The force behind the tool is the person operating it, of course, but I would be curious if you could share with us any feedback you have had from clients regarding how it has afforded them similar type gains as I have experienced? What kind of advantages do they report?

David: I think a great example of that is Civitium, a management and technology consulting firm for broadband infrastructure planning. As you know, the federal government has provided stimulus funds to state and municipal governments for infrastructure projects. Because they wanted to inject the money into the economy quickly, many of the projects had very ambitious deadlines for submitting project proposals. Civitium worked with their governmental clients to develop model-based proposals that incorporate engineering data with financials. They enabled their clients to win several of these projects, which will result in projects and jobs in their communities. In fact, Civitium participated in the awards ceremony which was attended by the Vice President of the U.S., Joe Biden.

Lang: I have seen Quantrix Modeler evolve a lot from my first version 1.2 that I purchased quite a while ago. What should we be looking forward to in the next version of Quantrix?

David: Well, there are two answers to that. Our upcoming point release, 3.7, will have an updated menu structure that makes working with external data easier. There are also refinements to some of the new features that came out with version 3.5, such as the chart grid. The development roadmap for version 4 is still in discussion so I can’t really shine too much light on that other than to say it will significantly raise the bar for business modeling and analytics software.

Lang: Could you give us a feel for how Quantrix Modeler is being used? Specifically, I am curious to know which industries it is being used in and maybe a few examples of how it is being used?

David: Yes, here are some specific industries and how they are using Quantrix:

• Higher Education: We have several non-profit and for-profit institutions of higher education that are using Quantrix for distributed budgeting and planning. University budgeting and planning is notoriously complex, which makes it an ideal space for Quantrix. In fact, we’ve been endorsed by Bill Massy, former VP of Finance at Stanford University and author of several books on the subject.

• Energy: We have several customers in both the energy generation and utility space. They are doing a range of activities from forecasting to capital budgeting over several years. These modeling activities tend to be complicated by the sheer number of variables which can affect plans.

• Financial Services: We count several of the Top 10 banks as customers. One client is modeling the performance of their real estate investment portfolio. They previously relied on 25 twenty megabyte spreadsheet models with 8 million formulas to a single Quantrix model with 500 formulas. They also went from a one year annual planning model to a three year quarterly planning model.

Lang: Like a lot of internet entrepreneurs, I have systems on various transactional servers and I have an interest in compositing a fuller view of my operations from these disparate sources. I have my own ideas on the use of Quantrix Modeler in such a scenario, but for the sake of my readers can you talk about how Quantrix Modeler with DataNAV and DataPUSH can help an internet entrepreneur with several bespoke database backed systems tie it all together?

David: Yes, the ability to tie various external datasources into a single model is a key benefit of Quantrix. By using DataLink, our inbound data integration tool, users can link a datasource to a matrix. Oftentimes, models are comprised of multiple matrices. By using business logic, the matrices can be linked together so that disparate data is seamlessly integrated in the model. DataNAV, on the other hand, is a great tool for navigating through datasources as you seek out the data that would be meaningful for your analysis and model. It allows you to create libraries in a column-store database and then, with in-memory analytics, quickly pull extremely large amounts of data into the matrix. DataPush allows you to take the key elements of your model and then publish them back into your enterprise IT systems.

Lang: My last question is not so product specific. I like companies and products that are outside the box in a good way. I see Quantrix as being that kind of company. One of the first things that impressed me about Quantrix is that you folks are based in Portland, Maine. There is a lot of talk these days in the internet entrepreneur world about how location affects the culture of a company, especially in the software world. Can you talk about what it is like to operate from Portland, Maine and specifically what advantages and what challenges you see in operating from there?

David: Sure, the paradigm for locating a company has really changed due to the technologies that enable you to conduct business from any place that has a good internet connection. In the past, you had to locate in area with a technology and business cluster that provided all the resources and expertise for taking an idea and transforming into a successful company. Quantrix is located in Maine due to the desire of the founders and employees to work in an area that has a very high quality of life. If you like the ocean, mountains, and roads that are relatively free of congestion, you would probably like to work and live in Maine. While I would say the culture here is relaxed, we still have the strong work ethic and drive that you would find in a company in any major city. In terms of drawbacks, it would mostly relate to working in a smaller labor pool and fewer professional investors. If you are a start-up, it can be more challenging to attract significant capital. That being said, it is easier to access the investment community here than it would be in Silicon Valley.

Lang: I always like to ask my interview guests if they have anything they would like to share in closing. Can you give us some insight or a parting thought to consider or feel free to drop some exciting news here if you like?

David: I think that our distributed budgeting and planning solution is a very exciting development. The solution basically pairs Quantrix software with a centralized database of the client’s choosing. It provides much more flexibility, dynamism and scalability than other solutions on the market, which tend to force organizations to work within rigid templates or the “lowest common denominator”. Not so with Quantrix. It enables geographically dispersed business units, campuses and other entities to develop budgets and plans that are specific to them, while being able to consolidate to a master model. Our solution maps to the business processes and unique requirements of companies. Challenges such as allocations, labor modeling, currencies, etc are all addressed with relative ease. As budgeting and planning are fundamental financial activities that most companies struggle with, it really delivers value in short order.