Late evidence is the most expensive kind

Every insight leader knows the meeting. Fieldwork closes on Friday, the debrief is booked for the following fortnight, and the leadership team made the call on Tuesday. Nobody was reckless; the business simply could not wait. The research was rigorous, on budget and, for practical purposes, irrelevant.

Speed and quality travel together

The usual assumption is that speed is bought with quality. The evidence suggests otherwise. In McKinsey’s survey of more than 1,200 managers², organisations that made decisions quickly were twice as likely as slow ones to report high-quality decisions¹. Only 37% said their decisions were both fast and good, and 61% said most of the time they spent on decisions was used ineffectively¹.

Bain’s decade-long research programme across more than 1,000 companies points the same way⁴. Decision effectiveness, of which speed is one part, correlates with financial results at a confidence level of 95% or more, and the most effective firms generated total shareholder returns nearly six percentage points higher than the rest³.

Jeff Bezos gave the practitioner’s version in his 2016 letter to Amazon shareholders: most decisions can be taken with around 70% of the information you would like, because a wrong call can be corrected, while “being slow is going to be expensive for sure”⁵.

Start from the decision date

If timing matters this much, research should be planned backwards from the moment of decision, not forwards from the start of fieldwork. In practice that means three things.

• ·

  Name the date. When will the decision actually be taken, and by whom? Not when the research would ideally finish.

• ·

  Stage the evidence. An early read from existing data, desk research and well-governed AI tools; then fieldwork aimed squarely at the uncertainty that remains; then confirmation where the stakes justify it.

• ·

  Make every stage usable. An interim answer that arrives in time is worth more than a perfect one that arrives after the meeting.

What has changed

None of this was impossible before; it was simply expensive. Desk research, questionnaire drafting, data processing and first-draft reporting once consumed weeks of a timeline. AI now absorbs much of that work, provided someone senior is designing the study and checking the output. The time it releases can go where it matters: the design at the start and the interpretation at the end.

The result is not a faster version of the old process. It is research built around the calendar the business actually runs on. If a decision in front of you has a date attached and a research plan that does not quite reach it, we would be glad to take a look.

References

1. ·

   McKinsey & Company (2019). Decision making in the age of urgency. McKinsey Global Survey, April 2019. https://www.mckinsey.com/capabilities/people-and-organization/our-insights/decision-making-in-the-age-of-urgency

2. ·

   McKinsey & Company (2019). Three keys to faster, better decisions. May 2019. https://www.mckinsey.com/featured-insights/leadership/the-organization-blog/three-keys-to-faster-better-decisions

3. ·

   Blenko, M. W., Mankins, M. and Rogers, P. (2010). The decision-driven organization. Harvard Business Review, June 2010. https://hbr.org/2010/06/the-decision-driven-organization

4. ·

   Bain & Company. The five essential steps for making better business decisions (originally published in AMEinfo). https://www.bain.com/el/insights/the-five-essential-steps-for-making-better-business-decisions-ameinfo/

5. ·

   Bezos, J. (2017). 2016 Letter to Shareholders. Amazon. https://www.aboutamazon.com/news/company-news/2016-letter-to-shareholders