(This article comes from a newsletter we sent to potential clients over the last year or so, the information was current when the newsletter was published, but may be slightly dated if you are reading this for the first time today.)
Technology Planning 202
By Mitchell R. Sowards
A few months ago in another whitepaper (“Riding the Technology Wave”) we explored the philosophy of “staying ahead of” or “riding” the technology wave and avoiding the “hell” of “being swamped” by new technology innovations or the “perpetual purgatory” of trying to paddle up the inside of the oncoming wave. Now let’s take another step forward and explore some practical methods of implementing that philosophy and the benefits you can expect from doing so.
In that prior whitepaper, we advocated for a philosophy of seeking a technology adoption position somewhere between “early adopters” and “early majority” while avoiding the bleeding edge of “innovators” or the “laggards” of “mass market” adopters. The goal was to adopt new technology at a point where the price is acceptable and where you can benefit from the technology for the longest period of time with the greatest benefit.
Productivity is The Key
Adopting new technology for its own sake can have benefits for firm prestige and employee morale. But the real key to new technology adoption is in productivity. Imagine that your firm had suffered especially during the recent recession and was unable to make any technology purchases. And imagine that all employees are now working with computers purchased between 2004 and 2007 – 4 to 7 years old.
Since 2004 we have seen the advent of “Web 2.0” applications, two generations of Microsoft operating systems (Vista and Windows 7) and two generations of Microsoft Office (versions 2007 and 2010) all of which are much more CPU and memory intensive. The same is true for most applications including accounting, music and entertainment, and many others. So, those old 2004-2006 PCs and servers are VERY underpowered especially if you assume that your firm did not buy “top of the line” computers in those years. To put this in practical terms, look at the difference between PCs from the relative eras:
Year 2004 2007 2011
Intel Pentium D 0.8ghz (maybe with 1st generation dual core technology)
Intel Pentium Dual Core 1.6ghz (2nd gen dual core tech)
Intel “i3” 3.1ghz (with 3rd generation dual core technology)
512MB at a speed of maybe 533mhz
1GB at a speed of 667-800mhz
4GB at 1333mhz
Windows XP – 32bit
Windows XP – 32bit
Windows 7 – 64bit
Now imagine that all or many of your computer users are daily using a “modern” software application (say, QuickBooks 2011) and that running a typical report on a new 2011 computer takes only 10 seconds. Based on my experience with customers still using older computers, I would guess that the 2007 computer would take 20 seconds or more and the 2004 computer would take up to 40 seconds – or two and four times longer respectively to complete the same task. And some really intensive tasks will take much, much longer (perhaps 10 or 20 minutes on an old computer). Of course, users do not spend all day on computer intensive activities like calculating and running reports.
But even extremely simple tasks like pushing a “save” button or switching from one screen to the next will require twice or four times as long on those older computers. For example, 2 seconds or 4 seconds instead of only 1 second. Multiply this delay by hundreds or thousands of computer activities per day per worker, and you easily can arrive at hours of unproductive time. (Say, 2000 computer actions (“save” or “next screen”) at 4 seconds each per day translates to over 2 hours of time spent on a 2004 computer and only 33 minutes on a 2011 computer.)
So, if you compare a single worker using a 2011 computer (with equivalent new technology at the “server” and “network” levels) and a worker using 2004 technology, you might expect the worker with 2011 technology to be twice as productive as the worker with 2004 technology over the course of a day’s work. (This assumes that each worker only spends 50% of their day in computer related tasks and the other 50% performing non-computer tasks like attending meetings, answering phones, etc.)
Now extrapolate that math over a small 25 person office and you get a PRODUCTIVITY PENALTY of over 18 hours of work per day from using 7 year old computer technology!
Now, most computer users will know this intuitively because they become frustrated with all the WAITING and begin to agitate for newer computer tools. But it’s helpful to put all that waiting into numbers upon which business leaders and owners can place a dollar value. In the above example, if you assume an average hourly wage and benefits of $25/hour then the daily cost of hanging on to old technology is costing that 25 person firm perhaps $468/day in productivity penalty.
If you assume each new computer costs only about $900 (not unreasonable for a business class desktop today), then it takes only 50 days to recoup the investment in 25 new computers! Hanging onto those old computers for a full year (250 working days) represents $117,000 of productivity penalty!
Striking the Balance
In the above example, where computers were allowed to become very, very old, then clearly spending the money to purchase 25 new $900 computers (total cost of $22,500) is a no-brainer. There will also be costs for “back end” devices like servers and networking equipment and costs for IT services for installation and setup. But the ROI on those devices is similar and can be left out of the equation for the time being.
Now imagine that a 1 year old computer is only 25% less productive than a brand new one (takes 1.25 seconds to complete a task a brand new computer completes in 1 second.) Clearly it makes no sense to purchase a new computer every year or even every other year because the acquisition and transfer costs exceed the productivity gain.
Based on the above analysis, and on my decades of observation, the optimum time to replace a computer is after 3-4 years of productive use. And the optimum method is to simply replace 1/4th to 1/3rd of your computers every year in a constant cycle of replacement. This ensures that your workers maintain a level of productivity commensurate with costs of the technology tools purchased for them.
Three Technology Purchasing Tips
Having identified the return on investment (ROI) basis which supports the philosophy of adopting new technology in the “early adopter” or “early majority” stages, here are some very specific tips for making configuration choices when purchasing new computers (or servers) for your business purposes.
1. CPU Technology/Speed – Always select the current generation of technology (don’t choose “single core” when “dual core” is the latest) BUT choose a speed that is about 2 steps below the current top of the line. So, if there are about 6 speed levels to choose from, choose the 3rd highest speed. Unless you plan to keep your computers for an extended period (say 5 years or more), then there is little or no benefit in purchasing the top of the line speed.
2. Disk Capacity – As a rule of thumb, purchase at least 4 times the amount of disk capacity as you are using today. Data growth of 50% per year is typical these days and it is only accelerating. So, if your old server had 500GB of space, then purchase at least 2000GB (2TB) today.
3. RAM/memory – As a rule of thumb, purchase only what you need today BUT ensure that the system has the capacity to at least double WITHOUT discarding original RAM. So, it might be perfectly fine to purchase a server with only 16GB of RAM if the RAM can be upgraded to 32GB simply by adding 16GB more and not discarding the original 16GB. RAM prices generally tend to drop for the next few years and you can get more later at a lower price as long as you don’t wait too long. After about 3-4 years technology may have “moved on” and your “old style” memory might actually cost more!
This guidance is specific for persons or firms making purchases for “business” or the “business office” for non-profits or government or education. They don’t really apply to home users who might have completely different (lesser) needs for casual use like email or different (greater) needs for gaming. Similarly, organizations that routinely perform computing intensive tasks like architects, engineers, CPAs etc. should consider more frequent upgrades (perhaps every 2-3 years) and more powerful machines (perhaps one step below maximum). Organizations that routinely perform less intensive tasks can get away with less powerful machines or keep them longer.
ONE MORE THING: ENTRUST has a long, long history of providing technology planning advice to our customers and a track record of success in helping our customers bridge the technology wave”. We’d love to help you too.
Learn more about ENTRUST: www.ntrusts.com
Contact us at: [email protected]
Call us at: 866-863-4738