One problem with being a tech journalist is that it can really skew your expectations.
You talk to analysts about "best practices," you read about Global 1000 company innovations, you find out about the very latest offerings from vendors, and slowly but surely, you start to expect better of the businesses you encounter in day-to-day life.
You know better is possible, but you forget that "best" isn't always practical or economical for every company. And, frankly, it's really annoying.
Last Wednesday, Ken-Hardin blogged about his frustration with a leading national discount retailer. He called it, "A Personal Tale of Data Integration Woe" and related how the company, despite using UPC scanners, couldn't tell him whether it had an item in stock without having an employee walk back to check the stock room.
You know how hard it is to get anyone to walk these days, much less provide adequate customer service. Hardin spent more than two hours and 40 miles of driving trying to buy the Espresso, four-piece high-top dining set allegedly offered on sale by said-discount chain.
What he couldn't figure out is why technology couldn't solve this problem. After all, he could see employees scanning tags with those UPC scanners, which appeared to be wireless. Why couldn't they access the warehouse data in real time?
"...I can only imagine that any consumer would be left scratching his head about why a business that already tags and collects data about stock flow won't - or can't - use that data to serve consumers a little better. I'm going to jump out on a limb and guess that the loading dock and customer checkout systems did not come as a matched set, so the irritating issue of integration affected me, the consumer, directly. And irksomely."
Hardin suspected the technology existed to allow the store to do better. And so did I. That's the promise of real-time data, right?
But is it really possible? Or, more to the point, practical?
Being the busybody that I am, I decided to ask Phillip Russom, the senior manager at TDWI Research, what he thought had gone wrong in Hardin's encounter. TDWI regularly researches such topics. In fact, I'd recently interviewed Russom about the quest for a 360-degree view of the customer and a new trend, enterprise asset programs.
In my e-mail to Russom, I theorized that perhaps the store did not have real-time integration because of the cost, and so the system required an overnight batch process to update.
But I did wonder: Why didn't they move to real-time? Was it cost? Technology? Lack of skills?
To my surprise, Russom explained that sometimes, overnight is as real-time as it gets:
"'Real time' is a highly elastic concept, and a wide range of people have a wide range of definitions for it. And the definitions regularly range from nanoseconds to 24 hours. Data movement, in reality, is far closer to the latter than the former, as you might've guessed, based on personal experience. How can over night be considered real? Because relatively few business or customer processes can react in nanoseconds or even hours."
Overnight data processing, which includes replication as well as batch processing, has several concrete advantages over true real-time technology, Russom explained:
- It's cheap.
- It's reliable.
- It's risk-free.
When you move to real-time, you gain speed, but you also add risk. Russom pointed out real-time technologies "are more expensive, fragile, serve a much smaller percentage of the business, and risky in how they are forced ignore data integrity, data quality, and other best practices for the sake of speed."
So, you can see why a discount retailer, even a national one, might opt to stick with the old ways.
Russom also noted that many UPC hand-held devices may appear wireless, but actually require you to put them in a cradle and sync them with the enterprise database. Once you do that, the information will be up-to-date-but it's rare that employees have the time or are even allowed to sync during the business day, he said.
Likewise, he added, most cash registers, point-of-sale, and inventory systems are on a 24-hour data sync schedule.
So what technology could have altered the outcome of Hardin's encounter, resulting in a much happier customer? What solution could the IT department give the sales person so this doesn't happen again?
Heads up, CEOs and CIOs. Here's what Russom suggested:
"The sales person got a little more information by reaching for a hand-held UPC device. But, in this case, he could've gotten fresher information by reaching for a different hand-held device: the good ole telephone."
I would add there's one more thing you'll need to put in place: A commitment to customer service. Unfortunately, too many companies still haven't applied the patch to that particular system flaw.