Let's get back to the basics. What is native advertising? What does it look like? and when/ why should we use it? Whether your trying to get client buy-in or you simply need a refresher, you've come to the right place.

Predictive Analytics|Leveraging data for ultra-targeted content

Predictive analytics enables ultra-targeted content and ads by leveraging data sets from multiple sources. Learn the basics and what the future has in store

zz-xcBy Ze Zook
Integrated marketing expert, author, lecturer and digital consultant
Connect

 


1.    What Exactly is predictive analytics?

Predictive analytics is a form of advanced analytics that examines data or content to answer the question “What is going to happen?”- or more precisely – “What is likely to happen?” (Gartner, N.D)

In this article, we’ll discuss the concept and promise predictive analytics holds for native advertising. Then, we’ll show you a real-life example of how predictive analytics is being leveraged today, through a closer look at retail brand Footasylum and their work with the digital marketing agency RedEye. Lastly, we’ll look at future applications of predictive analytics and its potential for reshaping the entire marketing industry.

2.    Why Is Predictive Analytics important?

Predictive Analytics offers great promise to native advertisers and marketers alike, as it offers them the potential to deliver the right offers at the right time and price to the right prospects—all with little effort. This means more time tweaking and optimizing content for conversion and less time getting bogged down in the process testing data. 

3. How does it work?

Predictive analytics is – first and foremost – rooted in Business Intelligence  (BI) software.

Screen Shot 2019-08-23 at 12.14.39

The aim of BI is for the marketer to be able to make better and more precise decisions. This has become all the more critical in today’s consumer landscape, where customers make use of multiple devices prior to responding to a call-to-action or making an actual purchase (Omnichannel marketing). Adoption of predictive analytics and business intelligence is said to be paving the way toward a new and cutting-edge era of marketing, complete with anultra customerfocus.

The mechanisms which are increasingly enabling this shift are Customer Data Platforms (CDPs), which work by creating a unified customer database that is accessible via other systems.

These platforms “create a comprehensive view of each customer by capturing data from multiple systems, linking information related to the same customer, and storing the information to track behavior over time” (CDPI, 2019)

The magic ingredient in this element is its real-time capability.

Once the data has been given a common structure and format, what is referred to as – Online Analytical Processing (OLAP),  insights from data mining, semantic or text mining applications can then create customized reports, which can be integrated with CRM systems. This seamless harmonization provides deeper customer insight and ultimately a competitive advantage.

NAD_Grafik_600x150

4. Examples of Predictive Analytics

In the competitive world we live in today, the adoption of some kind of Customer Data Platform (CDP) is almost essential. When working this way, though, we need to pay extra mind to issues surrounding data capture – such as content – establishing transparency with prospects from the start.

         Screen Shot 2019-08-23 at 12.14.54                                                                                                  (Footasylum 2019)

Case example: Objective setting for Predictive Analytics

To put all of the above into perspective, we’ll now take a look at a recent and successful use of predictive analytics, carried out by digital agency RedEye and their work with the online shoe retailer Footasylum, in which they set up a predictive Modeller Framework with targeted campaigns.

Footasylum’s first objective was to set establish a total of 24 identifying points. The essence of this framework is that it models a complete view of customer interactions addressing elements such as:

  • Personal /demographic, Age, D.O.B, etc
  • Channel Engagement, Email, social, etc
  • Onsite behaviour, Page views, Page Visits, etc
  • API’s & Data Feed, Social logins, Reviews, iBeacons
  • Transactional and datasets, LTV, online purchases

The second objective was through analysis, to understand and assess the individual and persona behavioural level of engagement from the Footasylum website. Because predictive analytics functions by the use and dissemination of datasets, the model frame needed to be refined further. Examples of this could be to reduce churn, increase the purchase frequency of high valued customers or, to convert customers from one-time to repeat purchasers.

In this light, predictive analytics becomes the glue that enables marketing to operate seamlessly in the background with data in the Customer Data Platform being interoperable and updated automatically, adjusting to external influences and new behavioural patterns, as well as allowing marketing content to be communicated at specifically targeted segments at the relevant stages the recipient was at in the buying cycle.

Of the persons in their database, Footasylum knew 2 things.

  1. They identified a segment whereby the likelihood to purchase was high – at 0.19% (segment A)
  2. The segment group whereby the likelihood to purchase was low at 0.06%% (segment B)

With Likeability being important in marketing, Footasylum set out an alluring Email campaign with a specific offer – appealing to both segments with a limited time frame for action (Purchase) to be taken.

They sent two high impact visually appealing emails to segments A and B, with a heightened sense of caution so as not to bombard the targets and a subtle reminder was sent to those who failed to respond after a set time period. 

When the results came in, the response was amazing. From those with a high likelihood to purchase, which was anticipated; there was a 27.5% higher conversion rate among the most likely buyers. It was those with a low likelihood of purchase who brought the greatest shock as it resulted in a 115% sales increase among the least likely buyers.

Footasylum knew from the onset that their objective was to leverage the relationship of prospects visiting the company website and not purchasing.  Part of the secret as to why their approach was so successful was the fact that data being pulled from one source point. If an objective is set to pull data from say, social media platforms, and banner adverts, there are different options available and this is where the CDP as a Data discovery application, comes into full effect.

5. The Future of Predictive Analytics

We are just on the cusp of unraveling what predictive analytics can bring. Why? Because it is driven by AI and – as we know – this subject matter is just in its infancy. Not to mention, the 2018 European Directive General Data Protection Regulation (GDPR) has been a bit of a hiccup when it comes to data collection, making it difficult to grasp a comprehensive view of customers, through data, and impacting the effectiveness of predictive analytics, which concerns how this data all comes together.

In reality, predictive analytics is not an entirely new concept. Maavi, a private Turkish high-fashion clothing manufacturer and retailer, founded in 1991, has been making use of predictive analytics for over a decade. They used it to locate groups of people, with distinct product preferences, in their database before putting them into product-based clusters and sent them seasonal targeted marketing campaigns over a fixed period. Using these clusters, Mavi was able to reactivate 20% of lapsed customers into purchasing new items.

As technology continues to progress, however, so too will the mechanisms through which predictive analytics will be carried out, moving toward the software-driven machine and deep learning that will enable the extrapolation of information – invisible to the naked eye – from seamlessly interconnected datasets.

A point to note is that predictive analytics is not solely a domain to be explored by large companies. While it may seem daunting to Small Medium Enterprises (SME’s), particularly in the wake of the new GDPR 2018 laws. Regardless of the size of the company or agency, the industry as a whole should be embracing the new practice and becoming champions of best practice and responsible usage.

A critical issue which might dissuade SME’s from making use of predictive analytics is, of course, cost. Presently, agencies and companies are still cagey regarding costs in relation to set up and implementation of predictive analytical campaigns.

While it may seem obvious, there are various factors – both internal and external that contribute to whether a company is willing to allocate financial resources for predictive analytics. It should be obvious to business owners of both SME’s and large companies, though, that this area of marketing should be perceived as a worthwhile investment. 

Tech-oriented small companies may be slightly more likely to be first movers when it comes to leveraging data and making use of predictive analytics. This, of course, hinges on whether directors and senior management have the correct mindset and persistence to make the approach work. Exploration and perseverance will be critical to its success. On the whole, if customers are treated with respect and the marketing industry places emphasis on trust, these are the building blocks of not only good business but essential customer relations.

Any business operating today should be holistically seeking out methods to transform and leverage the in and outflow of data, to provide ultimate value to their current and potential customers.

References

  • Artun, O. Levin, D (2015) Predictive Marketing: Easy Ways Every Marketer Can Use Customer Analytics and Big Data, Wiley
  • Customer Data Platform Institute: https://www.cdpinstitute.org/
  • Customer Data Platform Institute: CPDI (2018) Predictive Analytics & Customer Data Platforms: https://www.cdpinstitute.org/
  • Extract Transform Load (ETL) https://www.webopedia.com/TERM/E/ETL.html
  • Footasylum & RedEye discuss using Predictive Modelling to improve marketing results: https://bit.ly/2GiOr3c
  • Footasylum: The latest trainers & apparel: https://www.footasylum.com/
  • Gartner: https://www.gartner.com/it-glossary/predictive-analytics-2
  • Gartner: https://www.gartner.com/en/conferences/na/data-analytics-us
  • What is Predictive Analytics? https://www.predictiveanalyticstoday.com/what-is-predictive-analytics/
  • N.D (2014) Big Data and the Transformation of the Gaming Industry: https://www.experfy.com/blog/big-data-transformation-gaming-industry
  • Watson, H and Volonino, L (2002) Customer relationship management at Harrah’s Entertainment, Research Gate, Conference Paper (PDF Available) https://www.researchgate.net/publication/262161156_Customer_relationship_management_at_Harrah’s_entertainment

Want more insights?

Join us at Native Advertising DAYS 2019 for industry insights from 500+ practitioners. Don’t wait, get your ticket today to save BIG!

NAD_Grafik_600x300-NY

 

Leave a Reply

[data-image-id='gourmet_bg']
[data-image-id='gourmet_bg']
[data-image-id='gourmet_bg']
[data-image-id='gourmet_bg']
[data-image-id='gourmet_bg']
[data-image-id='gourmet_bg']
[data-image-id='gourmet']
[data-image-id='gourmet']
[data-image-id='gourmet']
[data-image-id='gourmet']
[data-image-id='gourmet']
[data-image-id='gourmet']
[data-image-id='gourmet']
[data-image-id='gourmet']
[data-image-id='gourmet']
[data-image-id='gourmet']
[data-image-id='gourmet_bg']
[data-image-id='gourmet_bg']
[data-image-id='gourmet_bg']
[data-image-id='gourmet_bg']
[data-image-id='gourmet_bg']
[data-image-id='gourmet_bg']
[data-image-id='gourmet']
[data-image-id='gourmet']
[data-image-id='gourmet']
[data-image-id='gourmet']
[data-image-id='gourmet']
[data-image-id='gourmet']
[data-image-id='gourmet']
[data-image-id='gourmet']
[data-image-id='gourmet']
[data-image-id='gourmet']
[data-image-id='gourmet_bg']
[data-image-id='gourmet_bg']
[data-image-id='gourmet_bg']
[data-image-id='gourmet_bg']
[data-image-id='gourmet_bg']
[data-image-id='gourmet_bg']
[data-image-id='gourmet']
[data-image-id='gourmet']
[data-image-id='gourmet']
[data-image-id='gourmet']
[data-image-id='gourmet']
[data-image-id='gourmet']
[data-image-id='gourmet']
[data-image-id='gourmet']
[data-image-id='gourmet']
[data-image-id='gourmet']