Between those who believe in the value of data science in helping brands make better decisions and those who undermine its ability to predict consumer behavior,  Brands, today, have access to advanced technologies allowing them to leverage big data and extract insights and better promote their brand with their target audience and enhance their decision making. The Berries interviewed Christos Solomi, Executive Director of Programmatic at Omnicom Media Group (MENA), to get his acumen on how can brands extract value out of Big Data .

BB: A significant change to strategic brand management has taken place as consumer trends drastically evolve. The large increase in customer interactions with brand touch points has created a huge amount of data. What does this mean for brands?

CS: In the face of an enormous amount of data, it is often difficult to filter out the right signals from the noise. If you have not yet done so, then invest in building the right talent within your organization that have the skills to do this, or seek external partners that can help. In any digital-first business this talent sits at the core of the organization. Data Scientists, developers and digital analytics experts are a given at companies like Uber and AirBnB, so it is no surprise that they thrive in the digital economy. Organizations not routed in digital have had a more painful and slow transformation. These specializations are no longer the fringe; it is the core of every business. Once you have access to this talent then the opportunities will become not only apparent but achievable.

BB: The era of Big Data has arrived, yet, few brand marketers seem to properly understand/make use of it. How can data become a primary source of competitive advantage and future earnings for brands?

CS: If you do not have the right talent within your organization or right partners you won’t be able to leverage the power of data. Often your first hire here will be crucial. Look at bringing onboard a CTO of CIO as they will be best placed to identify talent and partnerships.

In programmatic advertising, we use data, technology and rules to show relevant advertising to consumers with a relevant message. In the process of doing so, both we and our clients often start with a preconceived notion of who a “relevant consumer” is. The danger of being too sure of this notion and so targeted that you don’t allow your brand to be discovered by anyone outside of this initial group of “relevant consumers”. We try and avoid being blinkered by this. When you manage a programmatic campaign, or indeed any media campaign for that matter, you will no doubt have some incredibly targeted elements to your campaign but the smart marketer will blend in broader segments. By doing this, and within these broader segments you allow your brand to be exposed to new audiences that may be very interested in your products and services but were not initially obvious to you as a marketer. Our post campaign analytics often unearths such insights, which go beyond media targeting strategy and can be used for product development.

BB: Can you give examples on brands where big data helped them make better decisions? How can Big data be used to create effective personalized customer solutions?

CS: Creative used to dictate media buying decisions. I’m seeing, albeit slowly, a change in this dynamic. In the addressable/programmatic media space, we will often build and select the audiences we wish to target at the campaign planning stage. This information can be used to build a creative brief for the advertising agency so that audience strategy informs the creative messaging. We are already seeing examples of this happening in our region, at least at our agency. So, you may have hundreds of different ad copies, graphic elements, and call-to-actions built separately by the creative agency. Then at the execution stage, Dynamic Creative Optimization (DCO) technologies build creative on the fly so that the message and creative each consumer sees is unique and tailored to them. This brings us a little closer to the old mantra of right person, right message at the right time.

BB: How can brands use big date to reshape its brand experience?

CS: Let me illustrate this by way of an example. One of the most common ways a consumer experiences a brand is via their website or app. If you have a data layer in place on your site and you are capturing detailed information about visitors you can begin to create dynamic bespoke experiences for them. A logged-in user with a purchase history visiting your site can be shown bespoke content versus a first-time visitor. This is not new; it happens every time you visit a site like Amazon. However, you do not need to be an Amazon to deliver customer experiences like this anymore. If you have the right team in place they can select the right technology partners to deliver on this.

BB: One of the most important questions brand marketers face is what to do with big data. Can you share some tips on how to approach big data?

CS: The best advice is to either onboard a data science and analytics team in-house or partner with an agency that can fulfill this role. They will allow you to focus on what you do best, which is understand your customers’ needs and deliver products and services your customers want. At Omnicom Media Group, our data science division Annalect empowers our clients with data and analytics.

BB: What’s Omnicom’s guide for brand marketers when dealing with Big data? Does collecting more data always mean a better understanding for consumers? How can brands look at big data with critical thinking?

CS: It’s easy to get carried away with data but we need to remember that correlation doesn’t imply causation. The more data you have, the more correlations you will see but only a few will turn out to be true. It’s in our nature to want to believe correlations, it gives order to the world so they can be quite seductive. When taking any decision marketers need to have access to someone with a background in statistics and data science to help weed out false correlations. If you don’t then test, learn and avoid making drastic changes off the back of some data you have collected or seen. Ask yourself if the data makes common sense. If it is telling you something new about your customers, check if it resonates with what you know about them. I like the adage if something looks too good or too bad to be true then it probably is. If you are seeing incredible poor results and low conversion rates on a campaign, you probably have a tracking problem rather than a sudden dislike to your product!

BB: The Trust Mess: In your opinion, how can brands come up with an ethical framework when using customer data?

CS: First of all, let us distinguish legal from ethical. The first is important as there is now growing legislation around the world that governs how companies can use consumer data. If you are a company based in the Middle East or Asia but use digital tools that store data in the EU, then you better consult with a lawyer that is a subject matter expert here to ensure you are complying with EU law. If you are not, then you may be liable to some hefty fines. As more governments start to embrace legislation around data, marketers will have to work more closely with their legal teams to ensure that they’re not breaking on laws.

Ethical, on the other hand, is something totally different. You can be legally compliant but within the confines of the law still be perceived to be acting unethically. I would advise you apply common sense here and remember that a brand is in the game of winning over hearts and minds, not court cases.

Look at and understand your data collection and usage policy. Imagine that your own data is part of your organizations database – do you feel comfortable with how your own organization is using your data? Is it providing utility by giving relevant information to consumer for example? If the answer is no, then you may want to consider changing the way you operate. Consumers are becoming a lot more savvy and vocal. It only takes a small number of people to flag a brand’s inappropriate usage of data and have a significant impact on the brand perception of the wider, silent majority.