How the cloud makes data integrations easier

Is the cloud everything it’s cracked up to be? Any new technology goes through a hype curve, but generally speaking, I believe cloud-based solutions have done a pretty good job of matching some very lofty expectations. The growth numbers continue to prove this out, as companies like Amazon, Microsoft, Salesforce, and others experience extraordinary growth on what are already multibillion-dollar cloud businesses. By solving very real problems for their customers, they’re being amply rewarded.

That’s not to say there aren’t significant areas for improvement, however. One of the biggest challenges facing cloud adopters is integrating applications and data, something that has also been challenging for traditional IT departments ever since they moved beyond the mainframe. Not sure what integrations are? Trust me, you’ve been impacted by them – at least when they haven’t been implemented well. Do any of these scenarios sound familiar?

• You’re shopping at the grocery store, and find an empty shelf instead of the product you wanted.
• You order an in-stock item from a website, only to find that it’s actually not in-stock after you’ve paid for it.
• The contact information in your company’s customer relationship management (CRM) system doesn’t match the contact information in your billing system.
• You called a customer support center to upgrade or downgrade a service (like cable TV, or your wireless phone), but when you login to your online account, the change doesn’t appear.

All of these scenarios are likely the result of applications and data that aren’t effectively integrated:

• Point-of-sale systems that don’t automatically update an ordering system.

• E-commerce front ends that aren’t directly tied to warehouse inventory systems.

• CRMs that don’t automatically push names and phone numbers and email addresses to financial systems.

• And customer service center systems that aren’t synchronized with customer portals.

Fixing these types of issues, or even better, avoiding them altogether by being very thoughtful about system selection and integration implications, is time well spent. Fortunately, the cloud can make it considerably easier than might be the case for legacy, on-premise IT systems. Here are a few tips:

• Take a serious look at what I call cloud-based “ERP-lite” solutions, especially if you’re a smaller shop with relatively simple needs and little IT support. There are a number that support selling, customer service, billing, HR, and other key functions under a single login, no integrations needed, and that are relatively easy to stand-up.

• If you need something with more features, a number of very credible, full-blown cloud ERPs are available that account for virtually all business functions, but know that you may need significant IT talent to implement and operate. Even so, the technical challenges and implementation risks are far lower than traditional on-premise solutions.

• If you’re determined to go best-of-breed, or require the functionality present in more robust enterprise SaaS solutions, keep in-mind that point-to-point integrations don’t scale well, so look to one of the increasing number of cloud based integration tools, many of which already come with connectors to the most popular cloud applications.

• I’ve always thought Salesforce.com has an interesting model that is a sort of hybrid ERP & best-of-breed. While Salesforce.com doesn’t provide all the native functionality that’s need to run an enterprise, they’ve opened their platform to other vendors, making data integration either unnecessary, or much easier.

Before you venture down any of these paths, I strongly suggest you ensure ownership for entering and maintaining business data is clear. Integrating “dirty” data between systems is a pointless exercise, and recovering from it can be tedious. Establishing accountability for this can be harder than it looks, but solve it first, as it’s foundational.