
Online sales have been dramatically increasing for a while: their compounded annual growth rate in the U.S. between 2010 and 2014 reached 9.2% [1]. However, they still make up a small proportion of overall retail sales. Now, compare that to the 85.5% compounded annual growth rate [2] of online-influenced sales made in-store, in the same time period and in the same market. A clear trend is visible: people are influenced by the online channels, but rely on the offline media to complete their customer journey.

A question emerges: how can one leverage the power of the offline in creating world-class, omni-channel customer engagement? The data confirms that brick-and-mortar points-of-sale have to remain a key element in any marketing effort that embraces both online and offline channels.
The bar for what constitutes great customer engagement is set very high. It is a personalised, data driven engagement that leaves the customer with a better impression of the brand. The brand, at the very least, met every single expectation of the customer (e.g. in terms of speed, product availability or service level) and potentially did more than just that. The digital natives like Amazon, with their predictive models suggesting the things you never knew you wanted (not necessarily needed!), or Apple, with its intuitive and simple design, are setting the standards. Why shouldn’t we expect the offline channels to follow the lead and, instead of random “spray and pray” strategies, engage in more data-backed targeting? Additionally, the Internet of Things (IoT) brings ever-increasing opportunities for data collection and the test-and-learn approach of digital marketers can be easily applied to the in-store customer experience design (thus enabling the concept of Real Time Retail to turn into reality).
So, how can we weave offline channels into the omni-channel customer engagement?
There are different levels of complexity involved in a comprehensive omni-channel customer engagement strategy. The easiest solution features in-store digital displays (essentially a couple of TVs of different sizes, from 10-inch to 100-inch) with content controlled from a central location. This content might include, for example, ads suggesting specific items, or promotions that are likely to take into account the specific store’s sales behaviour, the time of day or weather conditions. For example, different ads can be shown in a Sydney store vs. a Melbourne store, not only at 8AM and at 4PM on a given day, but also on a sunny or rainy morning at 8AM.
Many companies that are currently experimenting with the digital screens in-store tend to lack the ability to assess the real commercial impact of such solutions (measuring extra sales or better brand perception). This means that many of them are dipping their toes in the water, trying to “keep up with the Joneses” (“everyone is putting in screens, let’s put a couple in…”), rather than making investments driven by quantified commercial impact. The key to quantifying the impact is, normally, an ability to work with the transaction-level data that allows for analysing relationships at the granular level of detail.
A slightly more complex solution uses facial recognition technology to detect different demographic segments of customers (e.g., male vs. female, child vs. young adult vs. adult vs. senior) and serve targeted messaging created specifically to those groups. The same technology (albeit in a much more elaborate form!) is currently used by the custom services in Australia and New Zealand in an automated border processing system (called SmartGate). The face of an individual crossing the border is compared with the image in the e-passport microchip, and his identity is thereby verified.
Among the various biometric techniques, facial recognition is frequently challenged on the basis of reliability and efficiency, as it performs best under certain conditions (light, positioning of the individual, facial expression, etc.). However, for the purposes of marketing in-store, the current accuracy is more than enough. The accuracy of technology is outweighed by its main advantage – that it does not require a direct engagement of the customer (unlike asking people to sign up to email lists or loyalty programs). Facial recognition tools can be deployed as a part of a digital screen solution, or used independently to get insights into customer behaviour; they are especially effective as tools for companies with less developed loyalty programs.
The most complex solution of offline stimulation involves mobile applications using beacons, which enables a two-way communication channel with the customers both in-store and online. The basic value of this approach comes from engaging the customers by “activating” an idle mobile application (that has been downloaded but perhaps not extensively used before). Beacons enable sending targeted and highly personalised offers based on location as well as the entire customer shopping history. Moreover, the use of loyalty apps in that context can drive additional reward frameworks for various actions taken in-store. Most importantly, the use of beacons enables retailers to get valuable information on customers’ preferred locations and product preferences, all in real-time.
There are a number of requirements to keep in mind while “designing” effective offline channels, and those requirements might become more challenging when you decide to enable broader omni-channel customer engagement. Firstly, in order to improve the granularity of the engagement, you need to scale your content libraries at the same time. Secondly, your technology infrastructure needs to be connecting multiple channels – something that becomes easier every day with scalable PaaS solutions (enabling multiple systems communicating across their APIs). Last, but not least, an aspect you need to consider is customer privacy: it remains crucial to assess whether improvements in customer engagement will entice enough customers to give up some privacy as a part of the process. Each of those topics deserves its own discussion.
[1] Own calculations based on: Internet Trends 2015 – Code Conference, KPCB 2015 (based on Forrester, data for the U.S.)
[2] Own calculations based on: The New Digital Divide: Retailers, shoppers, and the digital influence factor, Deloitte Digital 2014 (data for the U.S.)