By Harpreet Bushell, Group Chief Growth Officer, LAB Group
For decades, personalisation has been seen as the holy grail of marketing, with the term ‘relationship marketing’ first used in the early 1980s and the term Customer Relationship Management (CRM) officially entering the business vernacular in the 1990s. Even as far back as 2006, Time magazine named ‘you’ as person of the year in recognition of individualisation and trends like user generated content shaking up people’s expectations of how they interacted with content online.
30 years on however, most 1:1 personalisation marketing efforts fall short of being impactful or measurable, because we have come to over rely on the technology and customer historical data in these CRMs, for creating personas, predictive analysis and algorithms, instead of looking at ways to understand and serve individuals based on their current intent and motivations.
As historical data becomes less reliable and data privacy laws limit our ability to overlay cookies, and third-party data to supplement the gaps, it’s going to become exponentially harder to understand the make-up of our audiences through data and deliver relevant and personalised content to them. A Forrester report predicts that the industry will “say goodbye to third-party data and hello to zero-party data, data customers own and willingly provide to brands”.
Despite these obstacles, personalisation is still a no-brainer. Numerous research studies show personalisation can drive impulse purchases, higher basket values, and greater brand affinity leading to increased revenue and fewer returns. So, what can marketeers do to effectively personalise to anonymous audiences?
The solution lies in going back to the heart of personalisation and tailoring the experience based on the individual’s needs and wants in that given moment, much like we would face to face. Going back to my days as a sales assistant, it was second nature to tailor my approach based on what I could tell they were looking for and how they wanted help.
For example, working in the changing room I could tell from the range of things they were trying on (and how long they had been wondering around the store) if they’d be open to me suggesting the perfect accessory to match that top or showing them the colour options it came in.
At the other end of the scale, I could tell if someone rushed through the store and was impatiently tapping their feet in the queue the best thing would be to open another till to help them buy quicker rather than irritating them with the offer of 10% off if they opened a store card.
The same basic inferences from user behaviour can be applied online for personalisation. Research has shown ‘consumer habits and personality can be inferred from the way someone interacts online’ (Kosinski et al., 2014; Khan et al, 2008) and persuasive, personality-targeted marketing appeals have been proven to be significantly effective in contrast to impersonalised, typical marketing.
Furthermore, marketing appeals that were matched to people’s personality (extraversion or openness-to-experience) resulted in up to 40% more clicks and up to 50% more purchases than their mismatching or impersonalised counterparts (Matz et al., 2017).
Marketeers should think about 3 key stages to make the leap towards behavioural personalisation:
Stage 1: Map your desired conversions back to customer behaviours
In session data, including micro interactions enables us to segment users based on their current level of motivation and intention. With traditional personalisation we would map out all the actions and conversions we want to drive online and select what to display. This approach goes deeper, looking through the lens of how to nudge the user to complete those actions.
Stage 2: Identify metrics you can use to categorise these behaviours
Categorise these metrics of intention and motivation, inferring how much someone wants to browse, search and purchase. For example, I bought a few items of clothing recently off the back of a 20% off bank holiday offer. I quickly put a few tops in the basket and then continued to browse around new in, jeans, and cosmetics. From this you could infer I was a price-conscious shopper with high intention and motivation to buy.
Stage 3: Map out how you can adapt and personalise the user experience to nudge the next best actions for that user
As the shopper, if I had demonstrated high motivation and intent from my behaviours, we could make the calculated gamble to upsell similar items to in my basket, suggest colour variations or even tempt me into an entirely different category.
However, if I had searched specifically for a small, black, Nike t-shirt we could infer that although my intention was really high, my motivation to browse was not and similar to the offline example the best thing would be to make my purchase as simple and smooth as possible.
By looking at people’s actions on site and adapting the user experience accordingly in real time, rather than relying on rigid personas, demographic segments or inaccurate historical data we can not only overcome the challenges of anonymous users, we can also effectively use psychology for much more effective personalisation based on their motives, intentions, and emotional state.