Contingency Planning

Doomsday predicitions. Don’t we feel silly?

Evan Spence Evan Spence | 1998-07-17

It is currently vogue for companies to pay some lip service to their Year 2000 woes by making a statement in their annual reports. I’ve taken a grim glee in counting how many companies mumble about their Y2K project team preparing the company so shareholder value will be protected, and how the will be prepared for the next century. Costs are never mentioned, and progress is invariably summarized by stating that the projects will be finished in 1998, leaving 1999 for testing.

If everyone were to leave their testing for 1999, answer me this. Where is all this test capacity going to come from?

Be honest with yourself. It’s mid 1998. If you aren’t well along the way to finalizing your Y2K replacement/remediation/conversion projects, then now is the time for contingency planning. Yes, it’s time to cut your losses. Here’s how.

Start with the basics. What would you do if the power stopped flowing, and stayed stopped? For most companies, it’s simply not possible to haul a diesel generator up to their floor in a downtown high rise. But could you move some key processes to unaffected areas? Could you fall back on completely manual processes? Assuming you can work by daylight or candlelight, and without air conditioning and circulation, or maybe heat.

Can you budget for an extended period of non-productivity while your key supplies bootstrap themselves? Just get by on the legal minimum until systems return to operational. Remit your payroll, write your employees cheques (which they may or may no be able to cash), but otherwise go dormant. Can you afford the interruption? Your risk analysis should tell you if you have to start saving now.

Let’s assume the lights stay on. Go through each system—hardware and software—and ask yourself what you would do in the event of its failure or unavailability. This is not asking “Is it compliant?” this is “What if I can’t have it, for whatever reason?” Just assume some dependency has given way. Does paper work? Will a calculator do? If not, how about getting some redundancy now.

Same Thoughts, A Little More Organized

This is the process I use to get my mind turning around a potential Year 2000 failure and its consequences. To be a little more disciplined about the planning side of all this, follow the three steps of picking an objective, determining the criteria, and setting a timeline:

Objectives

Pick an objective for your contingency plan. Do you plan on operating just like you have been, with different facilities and systems? Would operating in a downgraded fashion suffice? How about stopping what you’re doing without hurting yourself? For many functions, operating in a downgraded fashion would be sufficient for a limited period of time.

Criteria

Under what set of circumstances should you invoke the contingency plan? Itemize the criteria by which you will know it’s time to act. Some good examples would be discovering, in late 1999, that your enterprise financial system wasn’t compliant after all, or having all your COBOL programmers fall out of your Death March remediation program.

Timeline

Contingency plans do not operate indefinitely. Predetermine the amount of time by which you will accept operations in contingency mode, after which normal operations can resume, or the downgraded operations are declared normal. By the same token, resumption of normal operations also has its own set of criteria.

Further Help

The real experts on contingency planning are the folks at the MITRE Corporation. In their Year 2000 section, they talk in detail about the numerous further steps for your plans. This is a rigorous process that contemplates the worst that can happen, from the very bottom of an organization’s physical infrastructure, right through the uppermost conceptual document.

Mitre impresses on you that designing such a plan is not a task for your spare consultants. The shock of contemplating downgraded operations, or actually proposing manual backup procedures to functions that have been automated since the Avro Arrow era, has to be steered through an organization from above.

Scaremongering again? Maybe. I’m good at it.

The best part about contingency planning is the development of a critical eye, that is used to evaluate the purpose, and importance, of every aspect of your business. I like to think of the Year 2000 as a singular opportunity to make worldwide businesses vastly more efficient.

What can I say, I’m an optimist.

Evan Spence

Friday, July 17, 1998

Appendix: Codetalker Contingency Plan

Section 4–H: Stockpile Inventory

pintday.org » Fresh every Tuesday.