Why does it matter if you choose the wrong search engine?
Regardless of the channel used for purchase (online or offline), consumers have the same expectations: they want to find a particular kind of product, explore your range and find bargains.
In a physical store, customers will walk along the aisles allowing themselves to be guided by the shelving and signage until they find the product they are looking for. If they don’t find it, they will go to a salesperson to ask for advice or directions.
Your customers approach your ecommerce site in the same way. First they use the menu to navigate and if your site structure doesn’t take them to what they want they will almost certainly turn to the internal search engine. It is also increasingly common for customers to use this search engine directly.
To use an analogy, your site structure is the equivalent of your store layout and your search engine is the salesperson. Its role is to facilitate a sale by making it easier to explore your site and find your products.
The big difference with a physical store is that on your ecommerce site the product offering can be infinite. You are not limited by floor space and you would do well to make the most of this!
Just imagine the confusion of a customer who enters a warehouse the size of Amazon’s and tries to find their way to a particular product by following signs. A search engine should be designed to avoid this situation. No matter what the customer’s level of familiarity with the purchase process, the search engine must be able to understand their question immediately and to respond accurately.
Whether the customer is searching for a particular category of product or for a specific item, making an impulse purchase or browsing the full product range, the internal search engine must be able to help them. Unlike structured navigation, it should not be limited by the classifications that have been decided by the merchant on behalf of the customer.
Give yourself every chance to satisfy your customers
If a salesperson in a shop were to reply “I’m sorry, I don’t understand your question”, it goes without saying that the customer would lose no time in leaving and going to a competitor.
The same is true on the web: if the internal search engine doesn’t answer the question, the internet user will go immediately to a competing ecommerce site.
If the customer has come to your site, or used your search engine, it is because they think that you have what they want. If you don’t respond, you miss a sale.
We know that every customer has their own knowledge of products and their own set of vocabulary that goes with it. Just like a salesperson, an effective search engine knows how to respond to this by putting forward a specific product when the request is specific (for example, a video game such as Call of Duty, FIFA or Assassin’s Creed) and to help narrow down the search if the term is more broad (e.g. a type of product such as “dress”, “water” or “television”).
The search engine needs to guide the customer to a set of products which include the one (or ones) that meet their requirements. Don’t forget that a customer who uses your internal search engine is a customer expressing a willingness to buy!
An internal search engine must be able to…
Understand the words that have been typed
As we have just seen, customers have their own set of vocabulary and it is rare that they will use any of your technical jargon when searching for a product. Who types “power tools” to find an electric drill? It’s up to your search engine to adapt, not only by correcting the words if necessary (such as common errors and typos) but also by subscribing to a shared language.
It must recognise the phrases that have been typed and use these to guide the customer towards the right result. For example, if the customer types “tracksuit”, the search engine will understand that it should also bring up the leggings from your catalogue in the result.
Take the search context into account
When a customer alludes to a product, the salesperson knows how to interpret the request and work out what to offer them. An internal search engine must also be able to do this, by deducing which products are suitable from the lexical field of the request.
For example, if the internet user types the words “wedding dress”, the search engine is able to expand the search results to include formal dresses or evening dresses - if the site does not have wedding dresses of course!
If your search engine isn’t able to understand the customer’s use of abstract concepts (e.g. “little black dress”) or generic terms (“evening dress”) it should nonetheless be able to put forward a result that is relevant (dresses, in this example).
Interpret what the customer wants
Even before they have spoken to a salesperson, customers transmit visual signals that can be interpreted. A salesperson can spot these signals and deduce the customer’s needs in order to respond to them. This is known as “cold reading.”
For example, if a salesperson notices that a customer is in the electronics department for some time, picking up equipment aimed at children, they might reasonably approach the customer to show them the hi-fi system that they have just bought for their son’s birthday.
A good salesperson would also remember that their customer recently came in to buy a Bosch sander and a drill. They might offer the customer a screwdriver from the same range, or at least know which brand to put forward first.
Your customer will be sending out as many, if not more, signals throughout their visit to your site.
Your search engine must be able to take advantage of all the available clues in order to put forward the most suitable products.
● For Richard, a “shirt” is for men, quite sporty, often blue, and from Pepe Jeans or G-Star, regardless of price.
● For Pauline, “jeans” are for women, not too expensive, skinny, and grey, regardless of the brand.
● ...
This data exists and it is important that your search engine uses it correctly, because the first results that are returned can make all the difference. These are the ones which determine whether a customer continues to use the search engine to shop with you. As we know, a customer who gives up on the search engine is a customer who gives up on the site.
Don't miss the following episodes:
an internal search engine must be able to:
- meeting the challenge of relevance
- help the customer to explore your product offering (suggestions, auto-completion, navigational aids)
- guide the customer to the right product