Why your BIA bogs down, and how to unblock it

The usual pattern

You send a questionnaire to every process owner. You collect a hundred and twenty responses. Each claims its process is critical. The exercise ends six months later on a table nobody can use, and the tacit conclusion is that "the business exaggerates".

The business is not exaggerating. It was asked the wrong question, in the wrong order, with the wrong instrument.

The sequencing error

ISO/TS 22317 describes three levels of analysis, and the order is not indicative. The strategic BIA asks the executive committee which products and services must absolutely continue, and in what order — a two-hour workshop, five to fifteen lines of output. The tactical BIA then identifies the activities supporting those priority products; all others drop out of scope, which cuts the workload by two thirds. The operational BIA finally applies only to the twenty-five to forty retained activities.

Starting at the bottom amounts to asking everyone to rate their own importance. The result is known in advance.

The instrument error

The questionnaire is excellent for collecting verifiable facts: volumes processed, applications used, headcount, seasonality, dependencies. These are things the respondent knows and has no reason to distort.

It is unusable as soon as arbitration is involved. Comparing two activities, setting a duration, accepting that a service will stop: these are decisions requiring adversarial debate, and therefore a workshop.

The question that unblocks

Asking "what is your MTPD?" invariably produces "four hours". Reframe it:

"It is nine o'clock on a Tuesday. Your activity is completely stopped. Who calls first, and at what time?"

The question moves the subject from feeling to observable fact, and the answers become usable. "Nobody before tomorrow morning" gives an MTPD of at least twenty-four hours. "The regulator, at eleven, because the filing is due at noon" reveals a three-hour threshold effect. "Clients, straight away, but they can wait a day" describes inconvenience, not harm.

Three levers against universal criticality

The ranking constraint: instead of scoring each activity independently, ask the committee to rank them by recovery order. You cannot have three activities in first place.

The recovery budget: "you have twelve people for the first forty-eight hours, allocate them". Scarcity forces arbitration far better than a scoring scale.

The explicit trade-off: "a four-hour RTO implies an active-active site, roughly four hundred thousand euros a year; do you confirm?" The question is not rhetorical. Sometimes the answer is yes, and that is valuable information.

The format that holds

A light factual questionnaire, fifteen questions, ten minutes. A two-hour workshop per division. Cross-cutting playback at executive committee with disagreements made visible. Formal, dated validation.

Six to eight weeks for a first complete BIA. Beyond three months, the exercise goes stale before it is finished.