Smarter Integration in Seven Steps

Share it on Twitter  
Share it on Facebook  
Share it on Linked in  

We talk about the hype cycle a lot in the tech press. Everybody knows it gets out of control sometimes, making it hard to identify the truth of a new technology or trend from the marketing hyperbole.


But what we don't discuss-and what is equally true-is that there's a crowd of IT managers out there who use the hype cycle as a reason to avoid change. These people prefer the status quo, refusing to invest in new tools or technologies, even if it means paying more in the long term.


Take, for instance, hand-coded integration. Any data integration vendor will tell you their number-one competitor is hand-coded integration. I've talked with IT leaders who say it's just simpler to write a few lines of code-they don't need a "data integration platform." I have to wonder how much those few lines of code-and the problems they create later, when there's no documentation on the integration-cost them in the long run.


Sometimes, change can be worthwhile. The key is knowing what's worth pursuing and what's not. One way I tell the difference is to see what emerges time and again as a best practice recommendation, from various sources. With that in mind, I've identified seven integration steps that experts say are worth the time and money.


  1. Put it on your calendar. This isn't a throw-away suggestion. I write about integration every weekday, and I can tell you the companies that resolve on-going integration problems and find innovative and strategic ways to approach integration are those that deliberately decide, "What we've done in the past isn't working. Let's fix it." Instead of handling integration on a project-by-project basis, they focus on the actual integration process.
  2. Break down IT silos. Note I said "IT" silos, not technology silos. IT silos can hurt in two ways. First, they can contribute to too many tools and too many different approaches to integration. Second, data integration is slipping out of the data warehousing domain. Even application developers now have to manage data-and the potential issue is that they'll be more focused on getting the application to work than getting the data right. All of which means the organization is paying for the same mistakes and redundant work-even redundant products. It's also not uncommon for operational data projects to pilfer staff from analytical data projects. And it's likely to get worse before it gets better: Rick Sherman of Athena IT Solutions recently predicted the demand for data integration will continue to exceed IT's ability to deliver it. All of which makes the case for ensuring everyone's on the same page when it comes to integration.
  3. Invite the business to participate, too. This is always a good idea for IT, but it's easy to forget when you're dealing with something as technical as integration. But business analysts are increasingly handling integration. And business users can provide valuable feedback on what's working, what's not working, as well as any unmet information requirements. While you're at it, ask them about their plans, hopes and dreams for using information in the next year and beyond. Then you can...
  4. Create an integration portfolio. Rick Sherman suggested organizations do this in his series on People, Process and Politics last year. "Creating the integration portfolio involves assessing where you are, determining where you want to be and how you are going to close the gap," wrote Sherman. It also allows you to make smarter decisions about your priorities, how you spend your integration money, and-again-ensure everyone's signed off on the same integration play book.
  5. Form an Integration Competency Center. Getting IT and business to agree on an integration plan will be hard, long work. An ICC is the way you put the plan into action by setting standards, sharing best practices and continuing the discussion. If you think ICCs are only for global conglomerates, think again: The benefits of an ICC are worthwhile, even at small organizations. It doesn't have to be formal and stodgy. At the very least, consider starting a wiki to document and share integration practices.
  6. Invest in standards. We all know standards aren't a panacea, but a lack of data standards can cost you significantly-although you may not realize it, as SOA architect and integration specialist Akiva Marks explains in a recent blog post. If you don't have standards, you're creating what Marks calls "embedded expense in the integration process." Standards take some time initially, but by adopting standards, you're actually saving yourself a lot of cost and heartache by benefiting from someone else's experience. "They appear rather complex (they are rather complex) as they're designed to cover all aspects of an industry object with necessary flexibility," Marks writes. "However, literally years of thought have gone into covering their respective industries with enough flexibility to cover individual company variation yet complete enough to have the full range of business operations."
  7. Invest in a data integration platform. Time and time again, the experts say it's worth the investment. Philip Russom, senior manager of research and services at The Data Warehousing Institute, says a tool pays for itself quickly because it substantially reduces the time your six-figure programmer takes to code data integration from scratch. But these tools don't just simplify integration-they can help you with data governance and data quality as well-a fact that companies in India have figured out, according to a recent Express Computer article. Furthermore, many platforms have built-in support for data standards-a plus if you followed step six and adopted a data standard. The best news: They're no longer just for enterprises. Data platforms are proliferating as the data integration market grows, writes Julie Hunt, a market intelligence analyst.