Personalized messaging and promotions are a tried and true strategy, but most effective at the bottom of the purchase funnel when a shopper is already conceptually sold on the product. To actually encourage undecided users to buy without dropping the price, the reason behind why a user should buy a product needs to come to life.
Likewise, affinity-based recommendations tend to outperform showing popular items or other basic recommendation strategies, and are at the heart of sophisticated personalization efforts.
However, even the most sophisticated affinity-based algorithms are missing the most valuable data input: Qualitative data about products. No matter how much data they have about a user, how can algorithms decide what to recommend if the only thing they know about a product is an SKU?
The why behind the buy: Boost commerce margins
A better strategy is to connect products with all of the content created about them online – especially content created by expert reviewers at publishers who have put hundreds of hours into testing and reviewing a product category. This content can provide the “why” behind the “buy”.
Here’s a practical example of how to do that:
Let’s say you’ve identified a visitor to your website who you would like to present with an individualized experience. Here are a few of the facts you know about them:
- A frequent shopper
- Has never bought an item that is more than $300, but has added several more expensive items to cart.
- Has viewed the product description pages for bedding and pillows more than three times
As it turns out, they have been traveling extensively for business and enjoying the deluxe bedding at higher-end hotels. On the flights home, they have browsed many different bedspreads, but haven’t quite found anything worth splurging on. Luckily, your merchandising team has just stocked a killer duvet cover that is akin to sleeping on clouds. Unfortunately, it is outside of their usual price threshold.
At this point, most conventional personalization efforts would resort to the tried and true strategy of offering a customized promotion. Since this item costs $400, trigger a 25% off coupon to bring this product comfortably into a price range that is more palatable for this user. Getting more sophisticated, you might even choose to create an overlay offering a $100 gift card instead of a refund to incentivize buying the item at full price AND give them a reason to return to your site.
Unfortunately, those tactics eat away at your margin. What if you could make the product worth paying full price for by really bringing the je ne sais quoi of this hotel-quality comforter to life? After all, price is only ever an issue in the absence of (perceived) value.
Luckily for merchants, commerce publishers have already done the hard work of writing in-depth reviews of your product catalog. Let’s take a look at a write-up of the aforementioned duvet cover in The Strategist:
If our buyer is still on the fence, this eloquent endorsement from two experts might be enough to recall memories of nights spent in ultimate comfort.
Power personalization with product characteristics
Furthermore, even if the price point is still too high, feeding qualitative product characteristics extracted from these expert reviews into your personalization algorithms will help you identify products aligned to their affinity at a more palatable price point.
Maybe there is a $250 duvet cover in stock that also delivers an “air of elegance” – to choose one example of a product attribute that could be extracted from the review above. With a standard collaborative filtering model, you would be restricted to recommending products that similar users had clicked on. With qualitative data such as this, you can actually recommend items that align to your shoppers’ taste.
As commerce content has become a lynchpin of successful media business models, publishers have doubled down on writing expert reviews across every major retail category, creating a massive digital library of qualitative data.
Collecting a myriad of data points to power a single view of the customer is at best, half of the equation.
The other half is the data behind the products. With this complete approach, you can unlock strategies to boost commerce margins – without driving down your bottom line.