Thursday, November 29, 2007

Trillium Software Customer Conference

The field marketing team told me today that the dates for the Trillium Software Customer Conference are confirmed now on May 20-23, 2008. You heard it here first. The official announcement comes out from our PR team in a week or so.

I’m excited about it this year, since the location will be here in Boston at the Marriott Boston Cambridge. Being so close to home, it will give everyone working in our Billerica office an opportunity to interact with customers.

The event is specifically aimed at customers. If there’s a feature in the software that you’ve been dying to see, or if you think Trillium Software needs to lead the way in a new direction, you can tell it to the VP of development and even the head of the Trillium division of Harte-Hanks. There are no guarantees your suggestions will make it, but the interaction is valuable to both customer and company.

We’ll have learning sessions, demonstrations and industry experts all contributing to the event. In looking over the mini-site, I can see that the final agenda hasn’t been announced yet, but they have been known to do a great job getting guest speakers, customer presentations and a few key Trillium Software employee presentations.

I had the privilege of attending and presenting at last year’s event in Las Vegas. I heard very positive feedback. Field marketing usually throws in a couple of nice surprises to make the event fun. Last year, one of the highlights was a “motivation speaker” who was anything BUT motivational. There were some Vegas shows that the whole group enjoyed, including Blue Man Group. I’m looking forward to a great time in Boston, too.

Wednesday, November 28, 2007

Wanted: Information Engineer

CNN Money has release a list of the hottest up and coming jobs. Number 3 is the "information engineer", a new title with a $70K to $120K salary potential. It's not surprising. I would estimate that salary range to be low, as corporations compete for resources to build data governance teams.
If you think about it, information engineering as a career move is relatively outsource-safe. Corporations will have a hard time outsourcing data governance, for the following reasons:


  • Data needs to be secure - with many laws on the books about data security, corporations know that it's hard to maintain security if the data is sent off shore.

  • Accessing data is half the problem - data tends to be locked up in silos, often controlled by someone who doesn't want to give it up, or someone who has left the company. Corporations know they need inside resources to work through the politics and opposition to access to the data.

  • No one knows the ins and outs of the company data like a company employee - an outsourced data quality won't understand the industry-specific and corporate-specific data challenges and how to solve them. It'll be tough enough for someone who has complete access to the business users of the data.


Let’s face it, as companies grow and compete, we’re not going to run out of data that needs to be cleansed, standardized, and matched. Building your career around data governance may just be the best move you can make.

Monday, November 26, 2007

Winners and Losers of Data Quality: Nominations

When companies have poor information quality practices, it’s hard to miss. The issues mostly manifest themselves in customer service (CRM) type interactions. As a customer, you may get unwanted or unnecessary contact, too little contact, or you’ll be struck with the feeling that the company doesn’t know you at all during a transaction. It’s more difficult to notice companies with good data quality practices. It’s more a good feeling you get when you do business with them. Good customer service, powered by proper data governance is becoming, more and more, an expected modus operandi.

Starting with this week’s blog entry, I’m going to nominate companies who either exemplify to me the epitome of good data governance with thumbs up, or show signs of data governance inefficiencies with thumbs down.


Thumbs Up: MGM Mirage Resorts and Casinos

I traveled often to Las Vegas in 2007 for industry trade shows, staying at hotels in the MGM system like Mandalay Bay and Monte Carlo, and those outside the MGM system like Caesars Palace. Depending upon who made the reservations for the trip, there were slight variations in my name and/or e-mail address that naturally occur.

Still, I am quite satisfied with the apparent knowledge that MGM has about me. They seem to understand that, as I am not exactly a high roller, discounted rooms and free buffets are appropriate offers for me. I don’t get duplicate e-mails and direct mail pieces from them, even though there was ample opportunity for those types of things to happen this year. The marketing materials that I get from them seem like a service, not at all a bother.

Nicely done, MGM. I’ll be back to bet the daily double at Monmouth some time soon.

Thumbs Down: Hewlett-Packard

Conversely, from my perspective, HP seems to have a data governance problem. Every quarter or so, HP sends me a catalog, listing home office store products. But every catalog comes in duplicate. Sometimes two, sometimes more. This time, one of the catalogs says that I'm "Steve Sarsfield" and the other calls me "Steven Sarsfield". For some reason, the one addressed to Steven also has a line that identifies the street that intersects my address. Not sure why, since that doesn't have anything to do with my official postal address.

So, I'm going to throw some numbers at you HP. Let's say conservatively, you send out 500,000 catalogs each quarter. Let's estimate that 20% of your database contains duplicates - a fairly conservative number. Finally, I'm going to estimate the catalog postage and printing costs at $1 ea.

That would be 20% of 500,000 = 100,000 catalogs that you send unnecessarily. At $1 each, that's $100,000 per mailing wasted. You mail it every quarter? 4 x $100,000 = $400,000 per year.

I know margins on computer hardware are very slim these days. How many computers do you have to sell to add $400k directly to the bottom line? In this one division in this very large company, there seems to be inefficiencies.

Besides that, it’s mildly irksome to receive duplicate catalogs – just more paper that I have to make sure goes into the recycle bin and not into a landfill.
Now, don’t get me wrong. I’ve been very, very happy with my HP printer and PC. Heed this, HP. Despite making some of the best products in the world, I believe that poor data governance is stealing away your goodwill.

Tuesday, November 20, 2007

Creating Structure from Unstructured Data


A lot of focus in the data quality industry has turned to cleansing and standardizing unstructured data. An example of this is shown above.

At Trillium Software, we continue to teach the engine more and more about supply chain and ERP data as well. We can take what would otherwise be very difficult to use description data, and putting it into buckets. The Trillium Software System understands the distinction between an item name, and size, and packaging and is able to standardize that information into proper fields.

Of course, the benefit of this is that if you want to understand how much polypropylene you have in inventory, you can't easily do it with the data at the top of the diagram. However, you can get a complete understanding after it has been put into its proper buckets (data on the lower part of the diagram). It comes in handy for that meeting with the polypropylene sales rep, since now you can fully understand the volume of your purchases.

One of the first customers for whom Trillium accomplished such a task was back in 1996 at a major food manufacturing company. The company had descriptions of ingredients such as “Frozen Carrots”, “Carrots, Frozen”, and “Frz Car” in their supply chain systems, and Trillium was able to sort it out.

More recently, there was Bombardier, which is available as a case study. It took only three months for a small team of engineers to design, develop, and implement a new process for standardizing 2.9 million inventory items using the Trillium Software System. Now, reports that once took months to generate are created weekly, providing high-quality information for streamlining procurement, reducing inventory, increasing on-time delivery, and boosting sales.

This is my last web log entry before the Thanksgiving break. Have a happy and safe holiday!


Sunday, November 18, 2007

Postal Validation for the Australia Post



One of the very basic functions you can offer as a data quality vendor is to validate data against the local postal services. With this validation, the postal service is saying that it has tested your software and it agrees that your product can effectively cleanse local data. The customer of said products then become eligible for postal discounts and save money when they mail to their customers. The US, Canada, and Australia have their own way of testing software to ensure results.
I took a look at the Australia Post web site to who was ON the latest AMAS (Address Matching Approval System) list and who was missing. It's interesting to note that, as of this posting, only two of the major enterprise software vendors (those in the Gartner Magic Quadrant 'leaders' section) now support AMAS.
According to the AMAS list, only Trillium Software and Business Objects (FirstLogic) support the Australian postal system with software certified by Australia Post.
Sure, a good data quality solution should have connectivity - it should integrate well with your systems. It should be fast, and it should support the business user as well as the technologist. It should have many other features that meet the needs of a global company. However, postal validation for global name and address data is basic. It helps the marketing department hit their targets, it helps the billing department's invoices reach the customer, and it keeps revenue flowing into an organization.

Saturday, November 17, 2007

Weekend Edition: Unique Christmas Gifts

On the weekend edition, I'm apt to take a departure from normal data governance and data quality topics and talk about... well, anything. This weekend, I found two very unique holiday gifts that I wanted to share with you. I'm not making money off this, just passing an idea along.
The first is shown here. My friend Joe Unni is offering a line of very unique, handcrafted clocks. In his custom furniture business, Joe is always on the lookout for interesting and exotic pieces of wood. He then meticulously sands and finishes them, and produces some very cool clocks. I had a beer with Joe a few nights ago, and he told me that he was able to get his hands on some 50,000 year old Kauri wood that was previously buried in New Zealand. The wood isn't petrified, but has some very unique qualities that you can't find anywhere else. He also showed me some other exotic woods that naturally come in oranges and purples. I think it's just a great gift idea. If you want one, it's best to give him a call on the number listed on his Web site.

Winter "Gardening"
I'll tell you my second unique gift idea, as long as you don't tell my sister. My family are avid gardeners, and it doesn't make sense to give gardening items in the winter time. If you've ever tried to shop for a gardener, you know that gardening items are hard to find in November and December. However, I think my sister will really enjoy the home mushroom kit offered by Mushroom Adventures. The kits have everything you need to grow a couple of crops of mushrooms. The mushroom soil comes inoculated with the mushroom mycelium (the fungus). You just add water and wait for the portabellas. Yes, it is a little "out there" in terms of gifts, but perfect for my family. It beats a boring pair of gloves.

Thursday, November 15, 2007

Siebel UCM at Schneider National


A case study on Schneider National came out today on TechTarget. This appears to be a case study that Oracle/Siebel drove, but Trillium Software is a big part of the success here. It's a classic case of a small company that grew rapidly without regard to information quality processes. Now, smartly, company is looking to clean up the 50+ databases of customer information to get a better handle on its business.
I was intrigued by this:

"Schneider's source data comes from a monolithic mainframe-based system that houses customer information such as billing and customer orders. The UCM also ties into a Lawson ERP system on the back end."
That's why it's so important for data quality vendors to support many platforms and offer many ways to integrate data quality processes. Here's an example of a customer that may possibly need: 1) Mainframe support; 2) Siebel UCM support; and 3) a way to hook up Lawson ERP to the DQ processes. Frankly, the vendors that have folded into the ETL tools or BI tools can't support all that. Only an enterprise DQ platform can do it all, and any future integration that may come up.

Wednesday, November 14, 2007

Rapid Business Growth and Data Quality

My colleagues and I just completed a webinar for Trillium Software that discusses rapid business growth and it's negative effects on data quality. It was called "Improving Data Quality in SAP Netweaver Environments with Trillium Software". In hindsight, it was probably a poor name. The event focussed more on how rapid growth sets up a company for huge data quality problems and how you can use re-usable data quality processes to tackle the problem. We had only about half of the attendees that we usually get.
But, wow! Where there were shortcomings in picking a name, we must have struck a chord with the content.
I had some great follow-up calls directly after the event to follow-up and ask questions. A pharma company called and wanted to chat about doctor's data. You know, doctor data can be quite challenging in that they tend to work in multiple facilities. So, in order to match up duplicate doctors, you have to go off of non-name and address components and compare things like tax id number or provider number.
Another company, who I'll describe as retailer, called to talk about some of the professional services that I mentioned in my webinar. More and more, staffing a data quality initiative is a problem, but we can usually help out with our professional services team. The company wanted to go back and clean up some legacy data for re-use in a marketing campaign. They were asking for advice on how to get started. Since Trillium SW now has strategic planning services, run by some talented folks like my colleague Jim Orr, we were able to help.
That was just two of the eight or so follow-up calls I got. Great projects from some great companies.
As for the webinar, if you get a chance, please check it out and let me know what you think. The idea is that as companies grow, the fact that they are trying to grow rapidly, often with little regard to information quality, can be a major problem. Not only silos of data crop up, but silos of information quality processes appear. Everyone begins to put their own unique spin on what the data should look like, and before you know it, the information quality problem gets worse.

Tuesday, November 13, 2007

Data Quality in Japan

Data quality problems are global! This is a poignant video on data quality in Japan as it relates to the governmental retirement program in Japan. It shows a company called Agrex who has used Trillium Software to cleanse data and find fraud in the Japanese system. Even if you don't speak Japanese, you can still follow the plot, so give it a try:

Monday, November 12, 2007

MIT's Information Quality Program

I just attended MIT's 12th International Conference on Information Quality in Cambridge this weekend. I have to say that I am amazed and delighted how informed the industry is getting about data quality. As I sat through each of the 20 minute talks, it became clear who was more theoretical than practical.... who were the presenters who were throwing it out there and trying to sell it, and those who have done it. Based on the Q&A sessions, the attendees understood this as well. The industry is getting smarter about how to get funding for and implement data quality.
You want an example? I sat in on a case study that BT (British Telecom) delivered by Nigel Turner and Dave Evans. Good guys and funny as they peppered in insults to each other as part of their shtick. Anyway, talk about amazing returns on investment. Back in 2003, BT realized that much of their investment in new enterprise application was being hurt by poor data quality. Nigel and Dave were part of a team that was able to sell the impact of information quality by selling the ROI. The key seems to be to talk about what will happen if you institute good data quality practices, but also include, as they say, the "do nothing" option. What will happen to the data if we continue down a path and do nothing about information quality on any given project? How much will it cost in 1 year, 3 years, or 5 years. That's a good strategy to take when delivering your data quality investment presentation to corporate. Fear, uncertainty and doubt always takes you far.

BT tracked ROI and their data governance initiative and was able to show hundreds of millions of dollars (or British pounds) worth of savings and benefit. Wow! Fantastic stuff.

This is one of the more academic venues for researching data quality, and therefore less commercial. The presentations were interesting in that they often gave you another perspective on the problem of data quality. Some of the information were clearly cutting edge, but I sat in on a couple of sessions that flashed up theoretical formulas for calculating data quality. Yup, that's nice, but when it comes to convincing your boss that you need to invest in DQ, they don't generally want to relive calculus class. Rather, they want to know how it'll effect me and my company. That's how you'll get the funding you'll need for the task at hand. BT had it right.
I talked briefly to Rich Wang, who is the founder and face of the MIT IQ program. He mentioned the possibility of speaking at the July event in Cambridge, and I'm excited by the prospect of it all. I have many ideas... many ideas...

Disclaimer: The opinions expressed here are my own and don't necessarily reflect the opinion of my employer. The material written here is copyright (c) 2010 by Steve Sarsfield. To request permission to reuse, please e-mail me.