Africa Needs to Get Big on Big Data

I once came across a Forbes article that said “small data analysis is like looking for needles in haystacks. Big data analysis is about turning the hay into needles”. Now, I am not sure how tempting an actual stack of needles is, but I believe what the writer was alluding to is the power that big data has to turn dormant potential into useful resources.

In this article, I will be defining the various kinds of datasets, with a focus on big data and explaining its practical uses in sectors across the world. I will draw attention to issues we currently face with gathering data in Africa and pose arguments on why it is important to focus more on how past behaviour can help us take advantage of future opportunities.

What have we got here?


‘The world’s most valuable resource is no longer oil, but data’ was the title of the May edition of The Economist released in 2017. It spoke about a new commodity that was dominating markets and prompting regulators into action, describing the front runners in this market as a new wave of services that appeared free from financial commitment and without cost, but, in reality, demanded a different sort of payment - data.

Data refers to facts and statistics collected for analysis purposes. Facts and statistics could be as straightforward as the time you wake up every morning or as complex as your nation’s daily contribution of co2 to the atmosphere. The point here is that if you are counting and/or keeping a record of facts for analysis, you are collecting data. The difference between the two datasets ‘big data’ and ‘small data’ is characterised by 4 V’s. Volume, variety, veracity and velocity of processing. All of which combine to make big data a bit more difficult to manage, but more accurate. Big data involves larger quantities of, usually unstructured, facts and information while small data involves smaller precise metrics. There is also a wider scope of variety in big data sets over small data sets, which usually means that big data is collected and analysed in batches while small data can be processed more quickly and in real time.


So what’s the hype about this new commodity?


A basic rule of thumb I have for identifying the most promising sector of any generation is looking at the most valuable firms on the Fortune 500 and noting what emerging industry most of them fall into (market valuation is not based on current revenue but potential revenue which is dictated by weighted indicators). Currently, Amazon, Apple, Google, Microsoft and Facebook sit at the top of the list. The interesting thing here is that, although they are all technology-centric, these companies belong to traditionally different industry categories. For example, Amazon is an e-commerce site while Apple is an expert on hardware. However, their focus on using data to drive their operations has brought about the creation of a new supersector.

According to the aforementioned release by The Economist, three years ago Google and Facebook accounted for almost all the revenue growth in digital advertising in America. Today they maintain this dominance with rollout of products like an assistant that can schedule an appointment for you over the phone, customised suggestions on google maps and even a sentence completing feature on your Gmail. All of which underpin their demand for more data. With over a billion active users across its seven products, Google has become a bank of information for what people want, and this has made it into one of the world’s most well-regarded brands. It’s privacy policy states “We use the information we collect from all of our services to provide, maintain, protect and improve them, to develop new ones, and to protect Google and our users. We also use this information to offer you tailored content – like giving you more relevant search results and ads.”

So in more direct words, Google offers you free access to its tools, then uses your data to better advertise to you which allows it to record $137 billion in revenue. As hard as it may be to believe, especially for African countries like Nigeria that currently still makes $143m a day from oil barrel sales, times have changed. The world has moved on and cars now run on data. Read more about autonomous car manufacturers like Waymo and Zoox who are gathering data to help navigate the next generation of vehicles.

Apart from ushering in technological advances, (big) data is also used to solve problems that have existed for centuries. For example, in healthcare, it has revolutionised the way doctors diagnose and treat diseases. By understanding trends over a long period of time, doctors can now plot what the potential end result of diagnosis is. This can be fed into machines or algorithms to quicken the process of gaining insight (supervised machine learning). This value chain is consistent across every industry that data drives: Collecting information - analysing said information to understand trends (the general direction in which something is developing or changing and causation) - using insight to make better decisions.


Africa could benefit from better decisions.


Policymakers across the continent are being urged to use evidence-based research to inform decisions around the development of policies. But this is impossible without a diligent approach to collecting and disseminating data. For example, how can a government build schools without knowing how many children need to be enrolled, or pass and measure the effectiveness of its policies if it does not have the development indicators in place to accurately depict the current status? The same is the case in the private sector. Investors need to know what resources are available in a given country, businesses want to know the behaviour and number of consumers in a given industry and donors want to know whether their aid is changing lives.

Currently, a lack of an entrenched culture of data use across the continent means that big data cannot be compiled. Think about this, if every store/kiosk/stall on the side of the road and those in more structured environments meticulously collected information on sales and customer behaviour, a central body could easily compile this data nationwide and gain a deeper understanding into a given industry - leading to better decisions made with policy and investment. This, however, is not the case in the retail industry and many other sectors across Africa. In 2014, according to a BBC world news release, Nigeria became Africa's biggest economy overnight. Literally overnight. This happened because it’s GDP was calculated with a new re-basing method - a review that is supposed to be carried out every 3-5 years but was not done for decades, suggesting that, for a very long time, decisions in and regarding one of Africa’s largest economies were based on inaccurate and untimely data.

There are some encouraging signs of a culture shift. Businesses and governments across Africa have gotten wind of the potential value of data-driven processes. A few days ago, Kenya became the first African country to launch a blueprint for a digital economy, indicating its mission to build a digital infrastructure that enables it to gather data that aids decision making. Visualise No Malaria (VNM), an initiative in Zambia is already using smart data gathered from digital tools to reduce malaria-related deaths. M-kopa is supplying solar energy instruments to rural East Africa and using IoT and cloud technology to generate data, manage its solar panel devices and upsell to customers. The World Banks’ Open Data Impact Map lists that over 38 organisations and startups across Africa are using big data to provide information on how farmers can either grow or sell produce. Ghanian based Farmerline is one of these businesses, providing best practice information to farmers on managing farms and increasing yield. In Senegal and Ivory Coast, Flowminder is using anonymised call data records to create likely paths of disease spread. Its maps were used to check the Ebola epidemic.

There are many more examples, but the truth is that growth is still slow. Many of these industries are private - which means they are not responsible for passing policy that could cause greater change. In addition, there is a shortage of skills and expertise required in digital transformation, data analysis and strategy.  

Tolu Oni is a Strategy Lead at Thread Strategy.

Thread Strategy