Product discovery is not solved online

Reflecting on his Christmas Shopping Benedict Evans posted this tweet yesterday:

If “retail-as-discovery” is a problem for consumers it is an opportunity for entrepreneurs, and it’s one that we’ve invested in heavily over the last twelve months. The best examples are, perhaps, Thread and Stylect which help men discover new clothes and women discover new shoes respectively, but many of our partner companies are in the business of helping consumers discover and buy new products and services.

One of the big differences between retail-as-logistics and retail-as-discovery is that with discovery there’s a much higher trust hurdle. The consumer doesn’t know what they want in advance and so has to trust that the product is good, that the price is right, and maybe that the retailer will deliver on time and handle returns well. The trust hurdle is pushed higher by the fact that many discovery items are high value – like shoes and mens fashion.

When designing to deliver trust rather than designing to deliver logistics the user experience is more about reassurance and less about efficiency. That needs to be reflected in site design, brand values, returns policies and just about every aspect of the business. The need for efficiency in logistics doesn’t go away, at least not very often, but it moves from being prominent in the value proposition to somewhere in the background.

Getting all this right is tough. But then if it were easy, everyone would be doing it. Successful tactics include elements of human curation, striking visual design, and most importantly, a big focus on building great product.

  • miquel_gourmetorigins

    I read that twit and I actually replied to express my disagreement, at least that is my personal experience as far as Amazon is concerned. He replied me this is a minority position. Maybe, it is true that possibly my reading list os not exactly mainstream, but, for me it is precisely Amazon’s book suggestions and recommendations that get me browsing their site for hours, while I find new books and topics previously I did not know I needed (although I am a hevy Amazon user, Prime is not relevant to me since I mainly use Kindle, otherwise the regular postal service can deliver usually within a couple of days). I don’t think my Amazon wish list would anywhere near the 500 titles it has now if Amazon wasn’t amazingly good at “discovery”.

  • You make a good point. Amazon does a good job of product discovery in some areas. I don’t use their book recommendations, but when I need cheap household items I search on Amazon and buy products based on their reviews, and take comfort in the price because there are usually other similar items in the search results.

    In the post I was talking about high value items where the purchase is often partly driven by emotions – “will I feel good in this shirt?”.

  • Joseph Frater

    Interesting thought. How do you see user-generated content being a basis of a solution that addresses this issue?

    For example, YouTube is used particularly by beauty and apparel shoppers whilst looking for inspiration (interest research by Google on this) but this is highly unstructured. My start-up, Soapbox, is approaching this by bringing YouTubers, to address the trust aspect, to a new video platform that’s geared towards shopping discovery for beauty and apparel.

    An interesting question is: do you see the solution evolving on to a platform-based model (where users do the work) from the current practices used by the likes of Thread/Stylect?

  • Interesting thought, thank you. We dabbled in social discovery when I was on the board of Lyst and my conclusion from that is that the challenge with social discovery is finding groups of users with the same taste. That’s generally not a good match with friend groups.

  • Joseph Frater

    Good point. Agreed on that regarding friend groups; however, social discovery can be broader than personal connections. For instance, social for Soapbox is between the beauty/fashion Youtubers and their followers, who have that interest in common. The same idea of social discovery would apply to Twitch too, in my view.

    The challenge is: can you link an impartial product recommendation to a purchase? As you’re ultimately recommending something you’re selling. I think that, like you said, it comes back to trust.