Developing and managing an online store is hard work. Getting consumers to your online store requires money and effort, and when they get there, you need another bunch of euros to convert them into paying customers. Read how to increase average order value by taking strategic approach to cross- and upsell.
Read also:
Part 1: Four different types of impulse purchase.
Part 2: How to apply impulse purchase insight to online
Boost sales with variety of stimulus
In the previous chapter, you learned the different types of impulse purchases. You might have started to see, how the product offering in your online store would fit into those different categories. Good! Now the next thing would be to understand how to apply this insight to digital.
Compared to brick and mortar, online stores have a lot of advantages, for example in terms of targeting. Only the sky is the limit for refining concepts triggering impulse purchase behaviour.
In user-centered designs, the user personas are tools for designers to better understand the end customers. One aspect of user personas is to provide insights into shopping behaviour.
Roughly speaking, there are two types of shopping behaviour:
- Task/goal-oriented
- Experiental/hedonic
The task/goal-oriented buyers get a bang from the outcome of the shopping, whereas the experientals get pleasure from the full shopping experience. From functional and visual design perspective, both appreciate well-functioning website and intuitive categorisation, but nice visuals and high-quality content bring extra value for experientals.
Understanding these two insights come in handy, when designing customer journeys for product recommendations.
So, a consumer is looking for a balcony table and you’d like to increase the value of the shopping cart by cross-selling additional products. It makes sense to show the matching chairs in context of the table, but you can be smarter than that.
Let’s first remind ourselves of the types of impulse purchases:
- Pure impulse
- Suggestion impulse
- Reminder impulse
- Planned impulse
For pure impulse, the most fruitful context to trigger impulse purchase would be inspirational content; images of a balcony table set for two or a larger group, nice atmosphere - maybe a darkening August evening - candles and lanterns casting their glow to surroundings, and throws readily on the chairs to warm the crowd when the sun goes down.
So, while setting the mood and selling the moment, you have a chance to sell all that is needed to create that atmosphere.
Showing in the context the wood oil (reminder), which undeniable is an important maintenance product for a wooden table, would ruin the moment. When you’re triggering impulse purchase behaviour in the context of inspiration content, you’re targeting the experiental buyer. The moment and sense of belonging is more important than the product itself.
The product page is of course the most common place for product recommendation. You can delight the goal-oriented shopper by making the shopping experience effective; present relevant products complementing the balcony table. Even the throws – remember, a goal-oriented buyer does not need inspirational content to make a purchase decision.
When a customer navigates to their shopping cart, you can make recommendations based on what is already in the cart. This would be a good opportunity to remind about the maintenance, and perhaps also about winter storage and recommend some storage bags (reminder), or show an offer for some practical outdoor tableware (planned impulse purchase).
Leave the details to machine
In brick and mortar, you still need some physical labour when optimising the product offering in the different placements in the store.
In online store, the machine can do the hard work of optimising.
So, it is still your job to create a model for cross-selling (by utilising the insights on different types of impulse purchase), and design and implement the concepts catering for different types of shopping behaviour. However, you can let data drive the details and optimise conversion on a product level.
If you have your cross-sell model in place and enough metadata to implement the model as logic, you don’t have to manually link products to other products. Instead, let a recommender system to decide, based on all the available data, which individual products to show. The data could simply be the best profit margin, customers' browsing history, or some prediction of which products are most likely bought together.
