Driving Value By Monetising Data
“Data monetisation” will become a major source of revenues, as the world will create 180 zettabytes of data (or 180 trillion gigabytes) in 2025, up from less than 10 zettabytes in 2015, according to IDC”
I’m writing this from Cape Town, where I have been invited to speak to the African Telecoms market at the HTTF conference. Opening on the first day I talked about three things:
- Data strategy
- The Internet of Things, and
- Data monetisation vs. data commercialisation
This blog is about that final bullet item: Data Monetisation vs. Data Commercialisation (quite a mouthful!)
Defining Data Monetisation and Data Commercialisation
These are my definitions of the two, here goes:
- Monetisation (Internal): using the data you currently have today to create new products / services, enhance revenue streams or streamline processes / operations. Essentially “cost-saving” or “revenue generating”. Examples of Data Monetisation: Segmentation / Targeting, Customer Acquisition, Customer Retention, Customer Churn and Fraud Management.
- Commercialisation (External): using your data to create “data services / products” that may provide richer data sets to other industries or providing a portal to specific partners that provide them with trends / patterns etc. Examples of Data Commercialisation: creating an API of anonymised and aggregated CDR (Call Data Records) and selling the service to insurers who can hook into the data as part of their credit scoring process.
How Companies Get Value From Their Data
Predictive Maintenance – Monetisation
A Media and Telecoms company had a real challenge with maintaining their satellite nodes. TV companies would broadcast programmes over their network, and if there was any downtime, a large fine would need to be paid as agreed service level agreements were not achieved, and broadcasts were interrupted. A specific monetary example told by one of the executives, was a node being down for 3 minutes, and the penalty the company had to pay was £2m. Ouch! Engineers would then have to go out to fix the affected node as quick as they could, which became very reactive and very costly for them.
With a better understanding of the data, a ton of questions about the process etc., we went to work building an algorithm that would be able to predict which nodes were in danger of going down, providing the engineers with this knowledge, becoming more proactive in their maintenance. The algorithm learnt over time and after a few tweaks, predicted the top 10 priority nodes in different locations around the UK that needed to be dealt with immediately. This helped engineering management plan their teams, and service those priority nodes before another disaster happened, saving a lot of time and a huge amount of money. The insights from the algorithm were provided through a visualisation tool which made it easier to interpret the raw data in a story board format, which could be easily interrogated by the teams to understand the issues across the network. Ensuring that the company were able to protect revenues and save on the cost of reactive maintenance, plus those fines.
Location Based Targeting – Commercialisation
A large advertising company who had fixed digital billboards in many European cities struggled with targeted campaigns to ensure their customers were getting the right exposure and conversions on their advertising campaigns. The CFO of the company was getting rather frustrated by a lot of customers asking how they could become more targeted in their advertising using their billboards, as opposed to using different channels such as social media advertising. The obvious view is that you have more expense on the physical nature of the digital billboard then you would say on advertising to your audience on Facebook, which gives you deep insights into the audience that you want to pitch to.
Discussing this and delving into the data that they had internally, there wasn’t enough to provide the insights to customers that would allow for a “minority report” style advertising campaign, not targeted at the individual because of privacy rules, but a group of similar people. After much discussion, the idea put on the table was to source two types of data
- Location data – sourced from a Telco provider which would provide where they lived only at a postcode level, and using the GPS, network and wi-fi data how they commuted from their homes (outside of London) to their jobs in London. So, you could generally see a mass of aggregated groups travelling from say Oxford to London Paddington or London Marylebone etc.
- Experian data – sourcing lifestyle data again at a postcode level to give an understanding of customer segments. An example of this is the following classification for City Prosperity (see image below)
Based on these profiles and location based data, the company now had the ability to track a group of aggregated and anonymised customers from specific postcodes to their destination which would typically be a large London train station. Once that was done, the groups with similar classifications would then be targeted by ads and messages that would appeal to them, and would lead to an interest or purchase of the thing being advertised. Using one city to pilot this and to see the results, they were then able to use it in other European locations and scale quicker once country data sets were acquired and combined.
So how can you start your journey?
This would be a very long blog if I mapped that all out, therefore, I’m going to be brief and boil it down to points
- Consider all potential stakeholders: customers, suppliers, partners etc. (internal and external)
- Understand their key pains addressed by your data or possible analytics products you could provide
- Create an inventory of your data – where is it, who owns it, how unique is it, is it granular enough, is it reliable i.e. good data quality etc.
- Make sure you think about the relevant security / privacy / regulatory rules before using it for commercialisation purposes as you don’t want to be fined big money
- Ensure you look at processes, skills and the culture of your organisation as data monetisation doesn’t just happen by giving people tools and data.
Make sure you have a sound data monetisation strategy, are committed to getting your data in order and making the right investments to support your revenue goals.