Global Marketing Alliance

4 ways to personalise your ads and wow customers

ad personalisation data

Marketers have spent the best part of two decades collecting as much data from consumers as they physically can; more data than they can use. In theory, all this information can be translated into data-driven decisions. This is the real benefit of these swathes of data, but many haven’t taken advantage of this.

Perhaps this over-hype comes from the sheer magnitude of observational data the “big three” hold.

〉 Google receives more than 4 million search queries per minute from the 2.4 billion internet users around the world and processes 20 petabytes of information per day.

〉Facebook’s 1.3 billion users share 2.5 million pieces of content each minute.

〉Amazon has created a marketplace with 278 million active customers from which it records data on online browsing and purchasing behaviour.

Regardless of why we collated all this data, it is our responsibility as marketers to make the most of all of it to drive both more success in marketing campaigns, and more value for consumers.

4 ways to utilise and personalise data

1. Three-tier granularity

Data provides the opportunity to deploy highly personalised marketing that is customised to an individual’s tastes or behaviour. The availability of data makes it possible for three levels of granularity in personalisation:

Facebook is a prime example of an advertising platform that offers different levels of granularity. For instance, its interest and behaviour targeting allows advertisers to deploy segment-level personalisation, while the dynamic product ads allow for individual-level targeting based on the products they have viewed on the website.


2. Online reviews 

In online retail today, the inclusion of consumer reviews has become a common practice, allowing brands to be more competitive in the market.

A recent study by Chong et al (2015) investigated the impact of online reviews and promotional marketing, such as free delivery and on product demand. With web crawling datasets from Amazon.com, they used technologies like neural network modelling – which can capture and represent complex input/output relationships – to predict the demands of electronic products.

The study showed that positive reviews were strong predictors of product demand and when coupled with promotional marketing, could lead to an increase in sales. Importantly the volume of reviews and number of answered questions were essential predictors of product demand in their neural network model.

Whilst this study has limitations – it only examined electronic products, the sample size was around 30,000 records and Amazon.com was the only marketplace analysed – it nonetheless shows the value of analysing data like online reviews, to predict product demand and tailor the marketing strategy accordingly.


3. Cross-channel analytics

Cross-channel interactions generate large sets of data, and marketers can make smart use of this data to gain valuable insights into which online and offline touch points influence the customers purchasing behaviour.

For instance, Joo et al (2013) found that television ads affected the number of branded-related search queries online. Dinner et al (2014) investigated the cross-channel effects of display, search and traditional advertising and found that they had a significant impact on different channels, particularly from online advertising to offline sales. These studies highlight the convergence of different media (television, display and search) and the resulting spillovers on campaign results.

Marketers should observe any carry-over effects from offline channels to predict their impact on online channels at a granular level. The importance of cross-channel strategy has been discussed for a while, but some marketing managers rely on experience and gut intuition to decide how to divide advertising budgets across media. But be wary, ignoring the quantifiable impact of these cross-media effects may put you at risk of underspending on offline channels and overspending on online channels.


4. Clickstream data

Clickstream data records which parts of the screen a user clicks when visiting a website or using a software application. Analysing this data is very useful for pattern matching between customer and non-customer behaviour, which helps firms identify segments for behavioural targeting.

A seminal study by Moe et al (2003) used clickstream data to segment visits as a buying, browsing, searching, or knowledge-building visit based on observed, on-site navigational patterns, including the general content of the pages viewed:

Of course, this is just one study, but ultimately understanding these categories of website navigation behaviour can allow marketers to identify likely buyers and design more effective and tailored promotional messaging.


The personalisation challenge 

These examples show how data provides great opportunities for marketers to deliver profitable campaigns, however, these opportunities do come with their challenges.

Tracking the customer journey

One challenge of data in marketing is the ability to generate and leverage deep customer insights. Most digital marketing reports are good for creating single channel reports, but it can be difficult to track the whole cross-channel customer journey.

For instance, customers may first see a brand on a billboard, read about it on desktop and make a purchase through mobile. Having the tools and skill set to bring together multiple sources of data and turn each individual touch point into actionable insights can be challenging.

For big data to truly shed light on the customer’s journey from awareness to conversion, marketers will need access to an understanding of tools that can bridge the gap between offline and online audience experiences.


Handling privacy issues

As more customer data is collected and ad targeting advances, privacy and security have become critical issues for big data in digital marketing.

According to a survey in 2015, less than 20 percent of smartphone and tablet users aged 13 to 54 in the US feel that companies don’t do the best job in protecting their digital privacy.

Moreover, more than half felt that companies “somewhat” protected their information but lacked transparency on their privacy policies. The Facebook-Cambridge Analytica scandal is testament of how the personally identifiable information of millions of users can be misused.

Additionally, the Yahoo data breach in 2016 that affected 3 billion user accounts is also an example of how privacy laws and security technology have not kept pace with data collection, storage and processing technologies. As a result, governments will increasingly enact strict privacy laws to protect their citizens, giving them more control over their data and the recent implementation of the less than 20 percent of smartphone and tablet users in the EU is evidence of this.

Strict privacy laws will limit how data can be used for marketing purposes, as less than 20 percent of smartphone and tablet users will need to be anonymised and advertisers will need to acquire consent from consumers.

Regardless, respecting customers privacy is good business practice and helps businesses build relationships with customers.

A study by Tucker supports this notion as they found that the click-through rate on personalised ads doubled when a website gave consumers more control over their personal information. Therefore, it may be plausible for marketers to focus on new software solutions that enhance security and give consumers more control over their data while still maximising on personalised marketing opportunities.

 

How are you meeting the challenge of utilising data? Feel free to share your experiences below.

 

Exit mobile version