Subscribe to our Newsletter

Decisions not data - how the 'crowd' helped me

There is a lot of talk now about data, big data, unstructured data etc. However, if I might be so bold, talking data first is putting the cart before the horse. Data is only valuable if it’s useful. As a data strategist and analyst, the focus of my work is on decisions not data, and it's not just me whose focus is shifting. PwC last year did a report on Big Decisions and this year CapGemini wrote a great article summarising their research and stating the case for excellence in data driven decision making (rather than data analysis or collection!). It’s great to see the focus moving back to decisions, where it truly belongs.

I invite you to consider what the important decisions are that you face, and then consider how you could you harness crowd intelligence as one of the data feeds to that decision? I turned to a new platform called almanis where I could ask a well organised crowd the questions that could help me improve my decision making. Having accurate forecasts on which to make decisions in business is critical. But history has shown the flaws in our traditional sources of foresight into the future.

“Human beings are not perfectly designed decision makers. We generally have less information than we’d like. We have limited foresight into the future. And we often let emotion affect our judgement.” –James Surowiecki, Wisdom of the crowds

We look to experts, for their superior knowledge and insight in a field to provide the best judgement and most accurate forecasts available. But research dating back to 1906 has repeatedly proven that the collective intelligence of the crowd is better. In essence, if you have a decentralised, independent and diverse group of people; the average forecast of the group will be better than that of the smartest expert.

This is collective intelligence. This is the incredible power of crowd prediction.

I work with large data sets, sometimes ambiguous data, big data if you will, sometimes small data, sometimes very messy and hard to use data. And in many cases, it seemed to me that the greatest information asset in most organisations is its people, and it’s being wasted. Utilising a crowd intelligence platform enabled me to harness the world’s people intelligence and transform it, into a highly refined signal. It’s backed up by rigorous science, lots of practical examples (in the intelligence community for instance) and the crowd on almanis - the prediction site I use reports a 94.7% accuracy so far on closed questions.

So in practical terms how can we apply this new technology when working as a data scientists, or experts within our own fields?

Forecasting product demand:
Here’s a thought. Let’s say that I run a supermarket in Australia and I want to forecast the demand for powdered milk. Those who know the Australian market will know that demand for powdered milk is far greater than domestic consumption would suggest, because every time there is a food scare in China demand surges. So, if I want to forecast demand and work out how much to put out on the shelves, it might be worth asking a question about the chance of a China food scare that makes a paper of record in a Western country in the next quarter.

Forecasting regional budgets:
The classic case on this is a question on the lifting of Iran sanctions. That’s a tough call to make, in point of fact that almanis crowd was highly volatile on the subject. But the commentary and the degree of volatility would be helpful to anyone whose job involves making decisions based on the likelihood of the lifting of sanctions happening (it did happen by the way). If nothing else, the way the almanis crowd processed the information on Iran indicated that
A. It was highly uncertain right up until January
B. There were some great resources available within the community with some really exciting opinions.

If I was trying to forecast a Middle East budget as an analyst, the data from almanis would help me document the key assumptions and would support me in assigning high risk to any numbers I put forward. And, it would be an invaluable sense check. It’s actually a bit of a no brainer!

Planning my next trip:
What about my overseas trip? Let’s say that I am planning to travel to a rather high risk area for malaria. We ask health questions all the time. You could ask whether the incidence of malaria for 2016 for country x will exceed y confirmed cases (or whatever measure from the Global Health Observatory you want to pick). The difference between a collective intelligence site and most statistics is that in asking the crowd, it provides a leading indicator. When you need a signal to make a decision not an ex post explanation of why what you did was wrong, it's incredibly helpful.

Assessing liability risk as an insurer:
What if I work in insurance and am heavily exposed to the US East Coast. I may well want to get a different opinion to my in-house experts on the likely severity of hurricane season. Pose the question, sit back and wait for the signal in the noise.

What decisions do you need to make today? I can’t help you there. Only you know. But the ones that matter probably keep you awake at night, so why not get the crowd to help you out? Next time you can’t sleep because of a question you can’t answer, ask yourself if you can get the community to work on it?

Almanis is a publicly available free crowd intelligence platform. Visit to join.

Lisa Schutz, Managing Director, InFact Decisions

Views: 66

Tags: Collective, big, data, decision, forecasting, intelligence, making, predictions


You need to be a member of BigDataNews to add comments!

Join BigDataNews

Sponsored By

On Data Science Central

© 2021   TechTarget, Inc.   Powered by

Badges  |  Report an Issue  |  Privacy Policy  |  Terms of Service