ig question is: how is this comprehension being used to actually improve the lives of customers? In smart retail it seems, the most popular way to personalise an experience is to begin a rapid process of decluttering; a spring-cleaning of all the redundant choices that appear to slow down or get in the way of customer decisions. After all, if you drill down to the stuff they really want, customers will keep coming back for more…right? While this approach may seem rationally sound, it makes a lot more sense for efficient machines than it does for human decision-makers. Here are three big psychological mistakes to beware of with personalisation: 1.Removing choice Most simply and profoundly of all, humans enjoy making choices. When we exert control and autonomy over a decision, our brain ends up pouring more value into it. Conversely, having our choices commandeered by an algorithm removes overall enjoyment and trust. This effect has been recently documented in a number of settings  – from psychologists refusing to use proven algorithms in diagnosis (Vrieze, S.I., & Grove, W.M., 2009) to the 49 per cent of Americans who purchased self-driving cars but still won’t activate autopilot. Even when an algorithm can objectively streamline and improve the quality of choices we make, our control-hungry brains demand their two cents. Consider the rising popularity of Flixable, a search engine that lets you peek beyond Netflix’s personalised ‘algorithm curtain’ and freely shop the full movie catalogue. While adding extra clutter and complexity into the browse, the psychological itch this scratches is nicely summarised in the following Reddit comment (with 39,000 upvotes): “I feel like [I]’m fairly stuck with the 50-100 titles shown to me on the homescreen, why can’t I browse their thousands of titles that they do they have outside of a search bar? [W]hy do I have to know the shows [sic] name to find it?” Beware: when we feel like a company is actively restricting our personal options – or offering up incomplete choices – it can make personalisation feel less like a tailor and more like a tyrant. 2. Complicating comparisons Remember that humans create value relatively, not in a vacuum. Whether choosing a first home or jar of peanut butter, we like to compare our options, efficiently weighing up features and benefits to work out which choice is ‘best’.  When Starbucks first introduced its 20-ounce Venti coffee, the size seemed daunting and somewhat undesirable. But after throwing the colossal 31-ounce Trenta into the ring, our choice to go Venti suddenly became a dainty compromise. When asked to choose between a short trip to a 3-star restaurant and a long trip to a 5-star restaurant, research subjects struggled to decide. However, when a third ‘decoy’ option was introduced – a painfully long trip to a 4-star restaurant – the 5-star journey suddenly became a clear winner (Huber, J., Payne, J.W. & Puto, C., 1982). In both of these examples, people are becoming more confident and happy in their choices by realising just how much better these are than clearly inferior alternatives.  Beware: when personalised retail strips away easy-to-make comparisons, and displays only options that are of similar value to a customer, it risks damaging both their confidence and desire to buy. 3. Hiding social proof While we undoubtedly trust our own good judgement and taste, there’s another intelligent trendsetter we look to with utmost reverence – the Herd. Since early human evolution, we’ve latched onto group behaviour as a signal of what’s right and what’s weird. Before picking which two-hour movie to commit to, ever paid a quick visit to Rotten Tomatoes? Or chosen a food stall in a bustling market because it had the longest queue?  This was your brain craving social proof – evidence of what other people like yourself are doing and loving. Beware: When you create an overly personalised retail environment, you extract the extremely valuable reference point of ‘other shoppers’, asking instead that customers make costly decisions based on their individual preference alone. What’s the takeaway Personalisation isn’t a bad aspiration for retail brands to have, as long as it aims to use data well, giving customers more confidence and joy in their decisions. However, when brands forget to take stock of the person in personalisation, no algorithm will save them. Dane Smith is Ogilvy Australia’s behavioural strategist and regional consulting partner.