By Noel Ward, Executive Editor February 2, 2004 -- Talk to any printer who has wrestled with a recalcitrant database and you'll likely find someone with less than kind words to say about their customer's database and IT department. Even seasoned data mavens in direct mail and transactional service bureaus are quick to point out that databases are the major barrier for printers seeking to offer customers variable data printing. While service bureaus can usually triumph over the gnarliest databases, most digital and commercial print providers are justifiably intimidated by them. And solutions are, at best, elusive. At the GATF Variable Data Printing Conference last November, a company that wasn't there was at the core of two of the solutions showcased by printers who used the NexPress 2100. The Rochester Group is an information technologies firm that's finding a rapidly expanding niche in helping digital print providers take on the vagaries of customer databases. We were wondering what they did that was different and what they bring to the variable data printing party. It turns out they have a very practical approach that develops a win for all concerned. Bob Miller, president of Rochester Group took the time to answer a few questions to give ODJ a high-level view of how his company approaches the database challenges facing printers. You should consider this the opening act of a continuing dialog. Bob has agreed to join our team of columnists here at ODJ and will be talking each month about the many issues printers face in working with customers, their databases, and how to solve them to make the data work. ODJ: Who is the Rochester Group and how long have they been in business? BM: Well, to answer your question in reverse, we've been in business since 1982. My two partners and I have been owners since 1993. In that time the world has gone from mainframe computers to minicomputers to personal computers, and data communications has gone from 300 bit per second telecom links to multi-megabit per second network connections. All of these advances have only added to the complexity of the systems and data that are out there. Still, most of the systems we and others worked on twenty years ago are still in operation, so the key to leveraging the information stored in them is to integrate them with all the newer information stores. Otherwise it's just chaos. So what we do is help our customers make sense of the chaos. ODJ: I've noticed that over the past few years most IT companies have had trouble sustaining profitability and have had to reduce their staff and cut costs. The Rochester Group seems to have weathered the storm and not had to reduce it's staff while maintaining profitability. How were you able to survive in these hard economic times? BM: First of all, I don't think most of our customers think of us as an IT company. Too often, IT companies want you to throw out everything you've got and replace it with the latest three-letter acronym. We integrate and extend what's there. But I think the real reason we weathered the storm is that, as a result of how we work, we have a really deep knowledge of our customers' businesses, and we develop and support systems which are central to their ability to understand and reach out to their customers. That becomes, if anything, more important during a recession. ODJ: I recently attended the GATF Variable Data Printing Conference where E.J. Flammer of BatesJackson, commented The Rochester Groups ability to collect and manipulate data was the key to its success. Can you gives us some insight on the complexity of this task and how your company's skill intellectual properties saved the day for BatesJackson. BM: The project E.J. was talking about is a good example of my favorite kind of partnership. What BatesJackson does to create such high quality printed material seems like rocket science to us. And what we did for them was actually pretty easy. We did basic name/address cleanup. When I say basic though, it goes beyond the Postal Service checking that most people think of. The BatesJackson data came from multiple individual contact managers, so there were duplicates and misspellings and the same person with different names -- Bob and Robert, for instance. Difference users had come up with creative ways to use the software, so we had company names in the address fields and county names in the city fields and so forth. Fixing these problems is a craft really, as opposed to a science. Luckily we have some people who are very skilled and have good tools, so in a day or two we were able to clean up some 5000 records. When we do work like this we retain everything we do in scripts, so the next batch of 20,000 just took a couple of hours. ODJ: I understand that the Rochester Group's key to success is its ability to understand the true business needs behind your customers technology solutions. Do you have any success stories that help explain how you've achieved success? Bob Miller , President Bob Miller's model for success is one of participation and cooperation. President of The Rochester Group since 1993, he has collaborated with the senior management team to grow the company from eight to over seventy employees and increase annual revenues from $750,000 to an estimated $6.6 million in 2002. With a reputation for setting aggressive goals, Bob's plans include expanding The Rochester Group into a nationwide network of offices slated to open over the next eight years. Bob is known for his unconventional creativity, leadership, and depth of expertise on strategic issues such as forming partnerships and charting growth opportunities. A firm believer in encouraging the right culture and working environment in a business, Bob works closely with the senior management team to ensure that the culture at TRG is "family-like, fun and caring... Everyone has individual talent and creativity, and our job is to create an atmosphere that brings that out." In providing value to clients, Bob's philosophy is to start with TRG's employees. "Our company is only as good as our people—they're the most important part of what we have to offer." Like the rest of TRG's senior leadership, Bob brings a unique combination of business intelligence and technical expertise to the company's consulting services. After earning his bachelor's of science degree with honors from Penn State, Bob moved to Rochester in 1980 to join I.P. Sharp & Associates as a business applications consultant, product development designer, and programmer. In 1984, he went to Computer Consoles Inc. as a software engineer. His projects there included developing ported system software for a CCI minicomputer and a database system for an MCI WorldCom Inc. switching system. He was also employed as a business applications consultant at LPA Software Inc. Bob is on the Board of Directors for both the Rochester Business Alliance and Publick Musick, as well as on the Board of Managers of the Memorial Art Gallery. He has also served as board member and president of the Society for Information Management. Bob and his wife Alana have two children and live in Brighton, NY. BM: I have lots of stories. One of my favorite is the time a customer called at 11:00 on a Thursday morning in December. A mainframe database that was used to create new customer accounts had been reorganized for the new year by mistake, and because of this the company couldn't take any orders from new customers for the rest of the year! I can't tell you who the customer was but I can tell you that this was a very expensive problem. We weren't the first people called, but everyone else had tried to find a way to restore the mainframe database and concluded that it couldn't be done. We recognized that what was needed was just a simple search capability on a DUNS database. We put that up on one of our servers and had the customer back up by 8:00 the following Monday morning. Although it was a 12 million record database, the application was so straightforward that it worked quite well. Although it might go against the grain to build a new system rather than fix the old one, in this case the new system was very simple and cheap. Considering the lost orders, the payback period for the system had to be less than a minute. ODJ: Accurate business information is key to achieving high return. What should companies do to better understand and manage their information? BM: I am going to disagree with ‘accurate' a bit. I think an awful lot of the failed data warehouses out there did so because the people implementing them thought only perfect data could be included. In fact, the more you dig the more find that there's no such thing as perfect customer data. People move, change jobs and even lie about themselves on a regular basis. ODJ: Having said that, we find the basic technique is to take whatever they've got and gain the ability to do flexible queries on it -- use the data to help them do their jobs. Then, as they start to use it they will find problems and you have to fix them one by one. As they move ahead, they'll realize there's other information they need, so you add it. It's really that simple. What makes this work is that the traditional approach is to specify everything required ahead of time, take one massive whack at making the data perfect, do release one, fix a few things, do release two, and then move on to the next project. This doesn't work because nobody ever knows what they need up front, and things change by release one. And because there's no such thing as perfect data, only good enough data, and the only way you know what is good enough is by using it. BM: I think this sounds very inefficient to most IT people, but since you never need the massive D-Day assault to get to release one, the overall level of effort is much smaller. It's just sustained. But that's the nature of a database versus a piece of software. A database is meant to reflect reality, and reality changes constantly. ODJ: How do you help customers stay abreast of the issues presented with data warehousing, online analytical processing (OLAP), and knowledge management? BM: We publish a quarterly newsletter which covers some of these topics, although we really focus on the business issues and not the technology. Still, our customers like us to worry about the technology so they can do their jobs. We also convene some customer groups for lunch meetings. I've found these invaluable in helping me understand our customers better, and they often tell me the same things. Interestingly, I think the biggest value of these meetings is what our customers learn from each other. We just get them in the room and buy lunch. I'm glad to do it. ODJ: How can print providers capitalize on your resources and talents? BM: One of my customers called me once to say that a third party had asked her for some reports from a database we maintained. She had told him to call us direct. That took a lot of convincing, because he thought we would put him through a drawn-out process of telling us exactly what he needed and why and when and how often, and of course we would try to tell him all the other things he would need and he would have to fight us off. The usual experience. Finally he said, ‘Okay. So you're telling me I just call them up, tell them what I want, give them some money and they'll do it?' In a way, I think we're hard for a lot of people to do business with, because they expect to be sold a product. We have some unique abilities and we look for opportunities to partner with others to create value. I would say that if someone reads this interview and thinks that we could work with them to make some money together then they should contact me. I won't ever do anything that doesn't work for all involved. I've never seen that model succeed, and I don't want to live that way. ODJ: How do you see The Rochester Group Integrating with digital printers? BM: We complete the value proposition for digital printing. The printer has the ability to create materials that are customized on an individual basis using data provided by his customer. However, what the customer has is data that's hard to access and inaccurate. That's where we come in In terms of the customer relationship, I think the printer has to stay in the lead. It's still the printer's value proposition. We're part of the solution. ODJ: If you could look into a crystal ball and see into the future, how do you see The Rochester Group helping the digital print industry? BM: Our mission, really, is to clean up this whole customer data mess. We're already doing that in partnership with printers. We're also developing a suite of sales and marketing management software packages which will improve the situation. Everybody I talk to in the digital print industry says what's holding them back is the inability of their customers to provide the data. So I think the more successful we are at accomplishing our mission the more successful the digital print industry will be.