The collective intelligence of Digital Cream
Category : Blogs
Station10 were proud sponsors of eConsultancy’s Digital Cream event last month. If you haven’t been to Digital Cream before, it’s an annual invitation only event for senior marketers to get together and compare notes on a range of topics and challenges that have come to the fore in today’s digital age.Digital transformation has happened and marketers need to find new ways to embrace recent changes. Digital Cream offers an opportunity to explore these new ideas with likeminded souls.
Analytics and Optimisation
Throughout the day, a range of topics were examined: from emerging trends in data analysis to best practice in creating digital experiences.The agenda is based around table discussions; each hosted by an expert on a particular subject. Our founder and MD, David Ellis chaired a round-table discussion on Analytics & Optimisation – Tools & Techniques for Improving Conversion Rates, exploring practical ways to gather and employ data to optimise your customer’s experience and, in turn, increase conversion.Conversations centered around two key challenges that were being faced across the board:
- How do I join up different customer data-sets collected from disparate sources?
We explored solutions, including creating mini segments to enable you to combine many small data sets incrementally and build up a wealth of insight.
- How do I help my organisation to become more data driven? We covered the importance of organisational change and why it’s essential to get senior buy-in to deliver insights. It’s all about your company’s approach to both collecting data and responding to those insights.
Of course all dialogues were kept under Chatham House Rules (what happens in Vegas stays in Vegas!) – but, back at the Station10 stand I was meeting and talking with some really interesting figures in marketing from a host of different organisations.
We discussed different strategies for collecting insight and why real data-driven decision making has become the cornerstone of successful businesses today.
The wisdom of the crowd
I also hosted a competition at the stand that centred around a jar of sweets and asked attendees:1. How many sweets are in the jar?
2. How many calories are in that jar?The player whose answers were most accurate won a new FitBit. But this game went beyond its winner. It was a tangible example of the statistical phenomenon known as the “Wisdom of the Crowd” in which the collective opinion of a group is more accurate than that of any single expert.This theory was established in 1907 when statistician and psychologist Sir Francis Galton asked 787 villagers to guess the weight of an OX to win a prize. The Ox was not in one piece, which made this estimation even more of a challenging. But what he discovered was that the crowd’s average answer was accurate within 1% of the Ox’s true weight:
- Ox Weight – 1,198 lbs
- Crowd’s average estimation – 1,197 lbs
This is taken as a classic demonstration of the power of collective intelligence, and Galton’s story is told and re-told with endless variations on the theme. My competition being one such adaption.What my adjustment emphasised was the volatility of variables. By including a second question I revealed a much larger chance of error in the crowd’s wisdom. It poses an additional challenge; which is more to do with people’s lack of knowledge about calories and also brought the theory around the wisdom of the crowd into question.
The wisdom of Digital Cream
I was able to speak to around 229 guests on the day (not as many Sir Francis’ 787 villagers but more than I had expected.) I was grateful for how willing people were to participate (although I did appreciate that this small a data-set might poke holes in Galton’s theory. I’ll come to this later).So, how much wisdom did this crowd have? In answer to the first question – “How many sweets where in the Jar?” – I was given various answers from 1,100 to 112 sweets. But, despite some wild answers the crowd as a whole where able to answer within 5% of actual number of sweets within the Jar:
- Number of sweets: 259
- The averaged estimation: 272
Here Galton’s theory rang true.However, when tackling question 2 – how many calories are in the Jar? – the crowd’s wisdom suffered (as did Galton’s theory). The average answer missed the mark by 45%:
- Jar’s calories: 6480
- Average guess: 9,385
OK, so Galton’s theory fell short here and as expected people lack of knowledge regarding calories push this theory to it limits.
The idea behind this theory is that in any crowd a range of people will give a range of answers. Some will guess too high; some too low. With enough participants these errors will balance out and bring the average closest to the truth.The value to be found in “the wisdom of the Crowd” is that you can make up for inaccuracy. Individuals may be subject to human error when it comes to predicting on their own, but as a collective they can forecast quite accurately. Of course, as I showed, multiple variables and the size of your data set will effect your results. Two heads are better than one. But, a multitude of brains at work; well that’s even better.As one of my hero’s said:
Not everything that can be counted counts, and not everything that counts can be counted.
— Albert Einstein.The value of your insight will largely depend on the questions that you ask. Ask the right questions; find the right answers. Only by truly understanding your data will you be able to deliver the insight that drives the most value.