Is your website like an untrained, uninterested shop assistant?
You invest a lot in marketing and click-through fees to get visitors to your website. Those potential customers start searching and browsing before – hopefully – they buy. A key part of that journey will be driven by the search results on your e-commerce platform, and by the products highlighted and recommended to these site visitors. Most e-com platforms offer search and recommendations and may claim advanced features for these. You may even have paid for additional plug-ins or SaaS integrated into your site to enable search and recommendations.
But all this will be ineffective, and you will lose sales and customers if your site makes common mistakes. It is like an untrained, unthinking, and uninterested shop assistant. You use mystery shoppers and staff training and motivation to aim for a consistently good customer experience in your physical stores. Yet on your website, you may inadvertently be making significant mistakes.
Here are five things your site must avoid:
- Making the same recommendations to all customers: It’s like a poor shop assistant not listening to what the customer is telling them or spotting the clues as to what product would suit them best. Your e-com platform needs to identify what each customer is looking for and adapt recommendations to best match that.
- Fixed recommendations: This means that the same associated and recommended products will be shown on a product or category page, regardless of other information. Your website must avoid this, and instead have dynamic recommendations. For example, that would mean not recommending a product that is out of stock or has just had a poor review. Related to the point above, your e-com platform should recognize each customer journey and modify product recommendations.
- Rules-based recommendations: Some rules can be useful, e.g., to promote specific products agreed by a supplier. But every rule needs to be set up and maintained. The performance of each rule should also be monitored, so that unsuccessful recommendations are quickly modified, to make the best use of valuable screen space. So, for the most part, rules-based recommendations should be avoided, as they are likely to quickly become out-of-date or ineffective. They also take a lot of manual work to maintain.
- Dumb search results: Not recognizing misspellings, phrases or other linguistic clues as to what the customer is searching for. While many search bars can pick up misspelt regular words, it’s surprising how often brand names aren’t recognized when spelt incorrectly in a search.
- Fixed search results: Like recommendations, you want to avoid showing the same products in the same order every time in response to a search. The results should be modified according to factors like the stock position, which products are selling best and preferences the customer has shown in the current browsing session or previous browsing or purchases.
Any one of these errors can cause visitors to leave your site without buying. The perfect products may be there and in stock, but the customer never sees them. Worse, they may find the experience frustrating and view your site as a waste of time, and they will be unlikely to come back. Again, the comparison is with the physical store experience – as a retail executive or as a consumer, you wouldn’t tolerate poor, general advice; you expect retail staff to take an interest in what you want, understand the product range and advise you accordingly.
Fortunately, it’s straightforward to review your site and see whether it has avoided the pitfalls above. If you realise that your customers may not be getting a great web experience, it represents an excellent opportunity to increase conversion rates and basket size. The tools to do this are easy to implemented and generating results within weeks.