To provide product recommendations you really only need two things: products and recommendations.
In the offline world recommendations are easy. You see someone looking at a nice skirt, you show them the matching top. If someone is browsing through videogame consoles, suggest your favorite game.
But what if you can’t see your customer? What if you don’t know how your customer is browsing your store? This is the problem online retailers struggle with, daily.
This blog post will discuss what you need to do in order to provide online product recommendations that actually work.
The Importance of Product Recommendations
The online shopper expects a certain level of personalization in their shopping experience. According to research conducted by the Keller Fay Group, “Personal recommendations are the number one driver of consumer purchase decisions at every stage in the purchase cycle”.
Companies who adapt to this change in consumer behavior are the ones who will succeed. If you’re reading this blog post, you’re on your way to becoming one of them.
There are many ways to calculate what products to recommend to a customer. Many consider Amazon’s product recommendations an industry best practice. According to Fortune, they use multiple factors when deciding what to recommend.
- Past purchases
- Items in shopping cart
- Items rated and liked
- Other customers’ similar purchases and views
This is a great place to start, but don’t take Amazon’s word as bond. No one knows your business better than you. Every business comes with its own unique customers and buying habits. Take the time to figure yours out.. Develop your own “recommendation algorithm”.
So, now you’ve decided to start recommending products your site. It’s not as simple as walking over and grabbing the recommended product from the store shelf.
Your website doesn’t know what your items are. It doesn’t know how they could be related. You have to provide this information. This is where product information management comes into play (and let me tell you, it’s important).
Product information is the map your website uses to find the items it’s looking for. If the data isn’t correct, you can’t make accurate recommendations.
This is the challenge that a Product Information Management (PIM) system is built to address. A PIM solution allows you to standardize your product information and make bulk edits to your products easily.
It may feel like overkill right now. It may be okay to manually enter your data for ten, twenty, or one hundred products. But, when you get to hundreds and thousands of SKUs, you simply can’t manage it all manually.
How to Provide Product Recommendations
If your product information is configured correctly, and you’ve designed your “recommendation algorithm”, the last thing you have to do it put your product recommendations into place. There are two ways you can do this:
- Download a prebuilt application
- Develop your own application
If you download someone else’s application, from somewhere like Shopify’s App Store or Magento Connect, it can be relatively inexpensive and easy. However, you may sacrifice some ability to customize the recommendations you want to make. You might even have to throw your recommendation algorithm out the window and start over. There are also enterprise level applications available, but unless you have a Fortune 500 budget, you might as well forget it.
If you decide to develop your own, you’ll need to be able to monitor your customer’s activity. You can do this with a cookie tied to a customer’s account. Or, if you want a more simple approach, you can track past purchases of an account and shopping cart inventory.
This approach can be more costly if you pay someone to develop it for you, or more complicated if you try to do it yourself. You have to decide whether completely customized recommendations are worth it.
The important thing is that you offer product recommendations, however you do so.. Today’s online shopper expects a certain level of personalization, and the companies who deliver that experience will win and retain business.
Once you get product recommendations figured out, you can move on to more complex issues, like buy online/pickup in store.