Big data for marketing – tackling the challenges

Yves de Montcheuil (pictured) has tips to help you gain advantage for your business.

Marketing has been one of the first business departments to appreciate the value of big data as a tool for commercial advantage.

Marketers were quick to recognise the benefit of getting instant insight into customers and using that understanding to influence behaviour at the point of sale. They were also ahead of the game in seeing the potential of analysing vast volumes of data to drive analysis and segmentation of different groups and achieve more accurate targeting of customers and prospects.

Understanding the theoretical benefits of big data is one thing. However, translating the theory into real commercial benefit is quite another. Here are five top tips, outlined to help marketing departments achieve this shift and deliver successful big data projects:

1. Don’t worry too much about volume – Remember, big data is diverse in origin, style, consistency and quality. Size is not always important. Some organisations have to handle massive quantities of data. Others have smaller data sets to manage but more sources and formats to deal with. Always make sure you focus on the ‘right’ data. Whether you are looking at social media feeds, CRM records or sales performance: identify every relevant source and don’t get too worried if it’s not necessary to instantly expand your processes to manage vast quantities of data.
2. Don’t waste the potential of your data – Some of the data needed for big data projects is easy to identify, such as transactional data used or generated by CRM or performance measurement tools. Much more is hidden on servers, log files or desktops and is often neglected. Some even goes to waste in the ‘exhaust fumes’ of IT. Typical examples are activity or sensor data that are not processed until errors or exceptions occur, but which can provide great insight even in normal operating conditions. All of this data is potentially relevant. Don’t limit your project to the first group: record it all and deploy collection mechanisms so that it adds business value.
3. Don’t move everything – Too many marketing departments focus on breaking down data silos and bringing all the data together in one central location. Remember, it’s not always necessary to duplicate everything. Businesses need to think about making distribution as easy as possible beyond the data processing phase. You don’t need all the data physically in the same place in order to get a single view of it, as long as a logical layer exists that enables this data to be accessed transparently.
4. Don’t focus on storage alone – Software frameworks should not just be repositories for big data. They also need to give marketers the opportunity to extract meaningful information from that data. Unfortunately, today’s businesses are not making full use of the processing tools and capabilities available to them. In order to draw out the intelligence needed, they should be applying the latest technology to harness that data, through data processing and analytics.
5. Don’t treat big data in isolation – Sandboxes for testing technologies work well for proof-of-concepts, but when big data projects for marketing go live they need to be managed as an integral part of the business architecture rather than an isolated project. Your business will need to integrate big data applications with other systems – both upstream and downstream – and at the same time ensure big data is part of the business’s overall IT and information governance policy. Marketers may only be one constituent of this overall strategy, but it will have a big impact on them.

Forward thinking

Interest in big data for marketing is growing rapidly. In line with this we are seeing more companies rolling out strategies that address this core business need.

Leading-edge technologies are lowering the adoption barrier, making it easier for marketing departments to get started. Yet, moving pilot projects into mainstream IT requires more than just technology. If marketers take note and follow the five tips above, they should ensure their big data projects get off the ground and help drive success for the business as a whole.

Yves de Montcheuil is VP Marketing Talend.