Building Successful Information Systems: The Right Time
This is the fifth in a series of posts on building successful information systems. A new post will be published every Monday. The posts are taken from Building Successful Information Systems: Five Best Practices to Ensure Organizational Effectiveness and Profitability. 2nd edition, by Michael J. Savoie. Business Expert Press, 2016. BusinessExpertPress.com
When Is the Right Time?
If I have the right data and I send it to the right place, but it arrives 2 days after I need it, do I have a good information system?
Ask FedEx, ranked 73rd on the 2011 Fortune 500 list, why they are successful. Rob Carter, the CIO of FedEx in response to the question “What business are you in?” said the following:
I believe we engineer time. I believe that as the world shrinks and changes, we offer solutions that allow you to engineer time to make things happen along time schedules that weren’t possible.
(Taken from an interview with Rob Carter, CIO of FedEx by Geoffrey Colvin, FORTUNE senior editor at large, March 20, 2006)
Engineering time is no longer just a FedEx thing. Now every company must engineer time to be successful. Seconds count. If I can turn around a quote to a customer while I sit in their office, how much better chance do I have to close the deal than if I have to leave and fax or e-mail the quote later on?
How Quickly Do We Want It?
Instantaneous access to information is what we desire. People want to access information no matter where they are. The question is how to provide this while ensuring the quality of the information, and without breaking the bank.
The right time deals with two criteria:
- When is the information needed?
- How “fresh” does the information need to be?
When Is Information Needed?
To determine when the information is needed, ask yourself two questions:
- Do we operate in a 24 × 7 world?
- Should we?
The answer to the first question seems obvious to most. The business world, and indeed the world as a whole, operates on a 24 hour per day, 7 days a week, 365 days per year cycle. There is no such thing as “downtime” anymore. However, does that mean that we have to operate that way?
The key question that needs to be answered for a successful information system is not “do we?” but “should we?” In other words, how important is it to our organization that information be “always available”? This may seem like an obvious answer, but when you consider the risks associated with always available data, the answer becomes less clear.
There are many examples of companies (Chik-fil-A, Hobby Lobby) that have chosen to operate on a less than 24 × 7 cycle. These companies seem to be doing just fine. Many universities limit the access to web portals to working hours (usually 8 to midnight on weekdays and 9 to 5 or less on weekends). This is done to minimize the possibility of undetected intrusions, such as malware or hacks. Why then, do we push ourselves to be “continually on”?
When we talk about the timing of data—getting the right data to the right place at the right time—we need to understand the timing of our organization. Just as FedEx sees their job as engineering time, we too must understand the importance of timing when delivering data.
The most obvious example is when data arrives late. Consider your own company. Do you want accounting data delivered to you every minute? What about every hour? Most of us feel like we’re drowning in e-mails. How bad would it be if you received detailed corporate data pushed to you every second of every day? Would you be able to use the data to make a decision? Note that we're not suggesting you don't have access to the data when you need it, rather that the data should not be pushed to you on a continuous basis.
How “Fresh” Does the Information Need to Be?
Fresh is defined as how close the reported data is to the real-time data. You would generally think that everyone would want the freshest data possible. However, there is a trade-off. If we just feed raw data directly into the information system, it comes in unfiltered. There may be missing data, invalid fields, and other errors that can cause serious domino effects throughout the network.
As an example, let’s say that a numerical control (NC) machine on a shop floor is feeding data directly into the shop floor control system, which is linked to quality assurance (QA) and accounting. Now, the machine operator improperly calibrates the machine prior to a run. The run is completed with 500 parts being made within specifications.
However, the machine, since it wasn’t calibrated correctly, reports that the parts were made outside of tolerances.
- QA is notified that the parts are bad and need to be rejected.
- Finished goods inventory is notified to remove the parts from inventory and place them in the recycle bin.
- Sales is notified that there are no parts available for sale.
- Accounting notes that all materials associated with the 500 parts need to be expensed against zero revenue.
And it goes on, even though the parts are actually fine. The only way to fix the problem once it occurs is to manually inspect each part and then override the system to put the parts back in. Not a good use of time, money, people, or the information system.
There Are Costs Associated with Timeliness
There are two primary areas where increasing costs are associated with timeliness of data. First, the sooner you want the data, the more of your information system’s power you will need. You may need additional hardware, bandwidth, and networking capabilities to support faster data transfer and faster processing. These costs can quickly add up.
Second, the sooner you want the data, the faster the verification process must be. This requires additional safety and security measures, additional personnel, additional monitoring software, and possible hardware upgrades. It is imperative that no data, no users, and no access occur without going through the verification process. At the same time, the verification process must not hinder the business performance of the organization.
It is a very fine line that balances these two requirements.
What Are the Benefits?
The biggest benefit is that the system, and thus your data, is always available allowing processes to function 24 × 7 with full access to the necessary data to support the business functions. Note that by “system,” we don’t mean the entire network. A correct strategy will include downtime, maintenance, and upgrades that occur at different times for different parts of the system. Therefore, shop floor control may be available 7 days a week, with upgrades in the evenings, whereas accounting may go through upgrades at mid-cycle or on the weekends. Sales, especially online sales, will operate 24 × 7, with maintenance occurring in the background or offline. All of this is dependent upon the business operating parameters.
What Are the Drawbacks?
There are two primary drawbacks in operating 24 × 7. As stated earlier, the risks associated with critical system failure go up as the opportunities for maintenance go down. There are many options for addressing the risk but none for eliminating it. Total Preventative Maintenance (TPM) programs should be created based on the “always-on” scenario. These programs address the ongoing maintenance of the system, contingency planning for local outages, and business continuity planning should a catastrophic failure occur.
The second drawback has to do with validation checks on the data. When information is flowing to everyone all the time, there is little time to check validity or data integrity. Too often, decision makers in a company will act on information that may be inaccurate, or more likely incomplete. In systems that operate 24 × 7, validation checks must be done as early as possible. Strong validity verification on input fields is a must. Drop down menus should replace typed input whenever possible.
For instance, dates should always be entered through a calendar pop up and states through a drop down menu. In addition, there should be clear rules on access by person and position (this is discussed in more detail in a later post). Finally, checks should be placed in the system whenever possible that allow for internal validation of the data request. These range from simple things such as attempting to generate reports with future date ranges to time and location restrictions.
Should we operate in a 24 × 7 world?
This is a strategic decision that you should make for your organization. There is no question that the world operates 24 × 7. However, that doesn’t mean you have to. If you choose to operate 24 × 7, recognize that information systems, like other pieces of equipment in your organization, need downtime for maintenance and upgrades. Build that into a strategic IT plan that covers all aspects of the five rights.
The key decision is determining the trade-off between importance of decision and speed of decision. In many instances today, executives make the assumption that the more important the decision, the greater the need to make it quickly. So, based on that assumption, the desire to have critical data delivered instantaneously becomes the goal.
However, instantaneous access to information, while a key driver in an organization, must be balanced against the need for the data to be valid and correct. The goal of the information system then is to provide information as quickly as possible without compromising the validity of the data.
Next Week: How do we Identify the Right Person?