One of the thorniest challenges in today’s b-to-c ecommerce world is the shopper “cold start” problem. The term “cold start” originates from cars—when it’s really cold, an engine has difficulty starting up, but when it warms up, it runs smoothly.
Similarly, the best personalization tools excel at guiding consumers to highly relevant product options but have difficulty incorporating a visitor’s “newness.” It’s a bigger problem than many realize. A recent survey found that a whopping 72% of shoppers who prefer to use guest checkout are actually repeat shoppers who already have an account with the brand.
This is clearly a missed opportunity to properly connect with returning customers, and it isn’t a quick fix. There’s also time pressure building, because the cold start problem is being further exacerbated by the phase out of third-party cookies and persistent identifiers looming over marketers.
As brands will soon be unable to rely heavily on cross-site tracking with cookies, how can marketers begin to increase their understanding of anonymous visitors and, in turn, prepare for a future of even less visibility? The good news is, all hope is not lost: Here are some tips on how to create more structure with anonymous shopper data and learn more about these mysterious visitors.
Carrots are better than sticks
A basic tenet of economics that governs behavior states that higher incentives amount to greater levels of effort, which amounts to higher levels of performance. In the context of online shopping, this translates to customers being more likely to log in instead of using guest checkout if you make the process easy, especially if it saves them time or money.
It must be clear to the visitor that being logged in would equate to significant conveniences and perks, ranging from ease of delivery with prepopulated addresses and billing information to acknowledging a returning customer status with loyalty points or discounts.
Brands should find ways to incentivize the shopper to log in instead of using guest checkout—that will reap the lowest-hanging fruit.
Brand trust is key
Requesting access to shoppers’ sensitive personal data, such as payment information, requires trust. That trust is earned; it is not a given, nor should it be.
Brands need to make an effort to disclose their top-notch data security measures when requesting personal details. Trustworthiness should exude across all brand touch points, from customer service to website quality.
Doesn’t hurt to ask
For visitors who insist on guest checkout, there may be a way to motivate them to raise their hands and classify themselves. Some self-report tactics entail quick polls that offer perks, such as coupons, for those willing to share something about their background.
Hearing from each anonymous visitor about who they are and what their intent is will help you understand not only what experience to offer but also infer their customer potential (average order value, lifetime value and more).
Pieces of the puzzle tell a story (if you look for it)
To further gauge an unknown segment, brand marketers can categorize “new” visitors versus “possible new” visitors. One way to do this is by analyzing the split of new visits across devices or browsers versus a split of “new” visits within a single source.
Take the time to examine buyer behavior closely—it can identify surprising patterns and result in unveiling valuable information with ramifications for your marketing assumptions and program.
A concierge approach drives intel
Another way to engage with and glean more information about your anonymous shoppers is to deepen the interaction. The idea is to approach online shopping in a similar way to how a store clerk would analyze in-store shoppers for clues to deduce their intent.
For example, if you have a shopper standing in the golf section in the middle of summer and they ask where they can find the gloves, the associate has a few clues to suggest they are looking for golf gloves and not ski gloves. This is a scenario where AI can come in handy.
Deep learning models can be trained to act like a sales clerk based on insights derived from the customer’s engagement history, mined from clickstream data coupled with a list of products viewed in session. But what’s more is that with the latest advancements in AI, it is also possible for recommendations to be personalized in real time depending on the consumer’s actions within that shopping session alone.
On the hunt for more clues
While AI and machine learning technologies can delve into these types of personalization without much historical data, A/B testing can help identify the most successful product pairings for a given brand.
Say a shopper has women’s shoes in their cart. A brand can try recommending a pair of men’s or children’s shoes as well to see if there’s interest to add a purchase for a family member.
Taking a bird’s-eye view
A big-picture view of what the potential ramifications are for the business is in order to determine the scope of the problem specific to the organization. In the case of the cold shopper, it can be tempting to hone in on one data set or assumption, but when we are too granular in our approach, we may miss out on a larger and more effective strategy.
Today’s shoppers expect individualized experiences like the ones they get from brands like Netflix and Amazon. They want to feel seen and that their future needs are also anticipated, whether or not they share their personal data. Brands that are working to make this happen now for both loyal, repeat customers and new visitors waiting to be won over are future-proofing their marketing strategy.
