Dust off your Y2K playbook: 10 strategies to drive a successful XP migration

As the clock ticks down to end-of-life for Windows XP, several tricks from Y2K conversion projects have become relevant again.



Remember the great Y2K conversion? Even though it's been 14 years, the Y2K playbook included several excellent strategies you can apply to large-scale migrations off Windows XP. Like Y2K, Windows XP's end of life has a drop-dead date: April 8, 2014. So if you're watching the grains slip through the hourglass, these key plays from Y2K initiatives might keep your aspirin on the shelf.

1: Accept disruption

Anytime you have a large user base and a definite end-of-life date, there is bound to be disruption, because not all migrations are smooth or entirely predictable. Organize a phased migration plan that proceeds either by application or by user area. Ensure that everyone involved has visibility of this plan and your progress against it.

Keeping your plan visible and updated takes a lot of the mystery and the anxiety out of the process for end users and IT staff. You should also have a "Plan B" failover strategy in case an application migration fails: Continue to run the app on XP (even once XP is unsupported) until you can safely migrate the app over to its new operating system. (Note: At least the XP Plan B beats Y2K's, which for many IT departments was a set of one-way tickets to Rio!)

2: Don't get fancy

Stick with just doing a straight app conversion and avoid the temptation to enhance an app or add something new that is custom. If you don't keep your code base the same on both sides of the migration, you'll complicate the process. Custom changes can introduce their own sets of bugs.

3: Get upfront buy-in on the budget

It costs money to migrate. Businesses don't like to pay for projects that return little value because all that is accomplished is that the same apps you've always run now run on something else. During Y2K, CIOs faced major hurdles obtaining the budgets they needed to do the full Y2K job. Until enough "risk" articles appeared in magazines about Y2K, many CEOs had trouble believing that a little date routine could put their entire business at risk. Budgetary support eventually got there, but CIOs had to fight for it. If you have a budget issue with your XP migration, make sure your higher-ups understand that you might have to run certain areas of the business on unsupported software for a time — and what the risks are in doing that.

4: Have a risk management/failover strategy in place

If your XP migration is exceptionally large, you will likely experience some migration failures (or at least complications). Define a failover methodology and allocate IT resources for failover so that your staff can quickly migrate an application back to the original XP platform if that becomes necessary.

5: Don't count on vendor support

Most vendors were great during Y2K — but some weren't. In fact, there were those who didn't even warn you about a Y2K date bug they already knew about until your conversion failed. Y2K folks quickly learned who these vendors were and didn't count on them for support when they had to resolve an issue. The XP migration team should adopt the same self-reliant attitude.

6: Identify your must-have apps

If you have a large XP migration comprising many applications, sit down with your users so you can all agree on a prioritization of which apps should be migrated first. These apps should be the most mission-critical to the business. You want to include your end users in this exercise because you need business buy-in. If everyone agrees to the game plan, they will be much more understanding if time runs out and you couldn't complete a total migration of all applications.

7: Choose total deconversion where it makes sense

One of the best things about Y2K was that it gave IT departments (and the business) an opportunity to throw out the trash. As IT and end users worked though the library of enterprise applications, a number of apps were discovered that had been installed but not actively used for years. In other cases, apps were so archaic that it was pointless to convert them. XP should present similar housekeeping opportunities.

8: Consider the cloud as an alternative

Migrations are a great time to consider whether you want to move an app to a cloud vendor forever —or if you're up against the deadline, move an app to the cloud until you can migrate it internally. Cloud is one weapon IT'ers didn't have during Y2K.

9: Communicate, communicate, communicate

One of the greatest moves from the Y2K playbook was the focus on daily project communications to end users, to IT, and to critical stakeholders in the enterprise, such as shareholders and boards of directors. After all, a failed Y2K conversion could break a business. The stakes may not be as great for XP conversions, but plenty of concerns can still be mitigated if IT regularly communicates status to interested parties.

10: Share the pain and the solutions with others

Widespread collaboration on problem resolutions and tricks of conversion between companies was a trademark of Y2K. It might have been one of the few times when companies that normally were fiercely competitive with each other opened their doors and shared both pain and solutions, because everyone had a common objective to beat the clock with their date fixes.

Organizations with major XP conversions can do the same thing. Sharing information on fixes and problems saves time and frustration. And it lets you know that you are not the only one out there who is wrestling with the migration.

Also read

Good times…

Were you involved in Y2K conversions? What other strategies from those days are playing into your Windows XP migration plans?




Mary E. Shacklett is president of Transworld Data, a technology research and market development firm. Prior to founding the company, Mary was Senior Vice President of Marketing and Technology at TCCU, Inc., a financial services firm; Vice President o...


Thanks for the article and expert comments from community. Sometimes we have limited choices if the client insists and  client applications run or to be tested in MS based systems. It is a time for companies to pilot the different alternatives.


If you're migrating from XP then you should seriously consider getting off Windows altogether.

Look at Linux and MacOS but treat Chrome as the joke that it is.


My business just finished our conversion to 7.   

I am not sure how I would feel if the question was only about home PCs.

In the business realm I would worry a lot if I had not already converted.  Not because XP will magically stop working, but because of the security threats after that day.   I do not love MS, but they do react to exploits and zero day threats with timely updates.  If I was going to release a new threat, I would definitely hold it until April 9th.  

That was one of our main reasons to drop XP.  I definitely would not gamble my business network security solely on the likes of Symantecs or McAfee.   

I am retiring in two years. Otherwise MS products would not even be in the short list of replacement contenders.



If You Have A Volume License Key You Can Install XP In Any New Machine, Even If The Old One Dies.

If Drivers Are The Problem, Buy Motherboards With Enough Slots, And Purchase Legacy Cards With XP Drivers - Be It Graphics, Sound Or Network.

I Have An External USB Network Adapter And It Came With XP Drivers - It Saved My Bacon A Few Too Many Times.



Why mention the cloud? or this another way that Microsoft is trying to divert attention from IE and its sieve like security issues? If they can't fix problems with IE who in their right mind would trust them with the cloud?


The clues is in the lady's background,  Marketing, she should goo and  get a job with Microsoft and join the happy band of double speak who are technically illiterate.


Alternatively just ignore the hullabaloo. Don't panic and stick with XP until your hardware eventually dies. After all, it's been your firewall and your antivirus software that's been keeping  you protected all these years, not Microsoft updates. If you haven't felt the need to update yet then there's no need to get in a tizzy now. Keep calm and watch the deadline sail by. The sky is not falling.


I am already down that road, I have selected a Linux distribution so that I can dump that boat anchor Microsoft junk and selected a European supplier for the computers, laptops, tablets etc. who will load any distribution of Linux I want at no extra charge and I won't be paying the hidden Microsoft tax. I am already getting a sigh of relief from all and sundry who have been laboring under Microsoft (you fill in the blank i.e. xp, vista, 8 and so on...) for years.

Robi Jf
Robi Jf

It seems like everyone is in trouble for not getting an update. I did mine about four years ago: I migrated to Linux - sorry, fewer viruses, all the apps I need are readily available, I do regular updates and major upgrades. No hassle, no pain, but gain here.

It won't apply to all of you folks out there. But for personal computing, there are plenty of solutions around. Should I opt for anything else? No I shouldn't! But that is good for me and me alone. I regret others seem limited to W$.

0:Why did you wait until the last minute?


First, you say Don't get fancy; then you suggest users Consider the cloud as an alternative ... really ? migrating an application to the cloud is not fancy/complicated ?

Then you say if you're up against the deadline, move an app to the cloud until you can migrate it internally ... which does not jive with the earlier statement that It costs money to migrate. Businesses don't like to pay for projects that return little value because all that is accomplished is that the same apps you've always run now run on something else ... so why would users try for 'two migrations that return little value' ? ... and is 'two months from the end of support for XP' the time to really explore cloud alternatives ?


"Like Y2K, Windows XP's end of life has a drop-dead date: April 8, 2014."

Huh?  This shows a lack of understanding regarding the old Y2K problem.

The Y2K problem was due to programs and data that used a 2-digit year or didn't accept 4-digit years beyond '1999'.  Once the drop-dead date was reached, they would either return erroneous results or flat-out crash.

When April 8th gets here, XP isn't going to stop working.  The apps running on it aren't going to stop working.  The output they generate is still going to be valid.

The planning points are valid, but that opening comparison is way out to lunch.


You can buy all the legacy cards and peripherals you want.  You'll do okay finding some things (NICs, video, hard drive controllers), but good luck finding new, unused, warrantied versions of everything you want to replace, especially in a business environment.  If you're just going to install old gear on a new mother, why bother?

Running XP on an existing system is living on borrowed time, but it's a low 'interest rate'.  Slapping XP on new mobos crammed with legacy parts and apps isn't a viable long-term business solution.  If you're using that approach as a long-term deployment strategy, the cost of ownership will quickly become prohibitive.  Who builds systems for a workplace environment any more?  The labor cost alone of ordering, testing, configuring is more than buying off the shelf, much less the cost of the parts to replace newer stuff that is already on most mobos anyway.  Most new applications and peripherals won't run on it, and younger employees will have to learn how to  navigating it.  It may be a business necessity in the case of legacy apps; it may even be an interesting hobby, like maintaining a vintage car.  Long-term?  Bite some other bullet: Linux, the cloud, even using that volume license to cover downgrading to W7 would be better.

XP, FOREVER (on systems it's already on).

Editor's Picks