Letting your customers know that there is a person on the other end of their online order who is grateful for their business is a beautiful thing.
That’s why it's common for budding ecommerce businesses to add a personal touch to a customer’s online shopping experience in order to stand out from larger competitors.
For example, startups like Frank & Oak include a hand-written thank you note (in the early days, the founders wrote the notes themselves) inside your shipment. Others might throw-in extra batteries or free product samples related to your purchase.
Nothing beats a value-add delivered by a real human being. But as your company grows, it can become difficult to maintain that same level of personalization – often because you’re just trying to stay on top of all of the new orders coming in, or you are too busy developing new business processes to manage rapid growth.
In the case of Frank & Oak (which has grown from a startup into a well-established fashion brand with over 1.5 million users) the thank you notes are still being hand-written by employees. But the founders no longer have the time to write the cards personally.
The good news is that technology is evolving and adapting to help your ecommerce business stay “personal” – even as it scales in size. And, in fact, consumers actually want you to use popular digital solutions to do so...so long as they don’t feel robotic.
Even in the case of handwritten letters, there’s now a website called MailLift that has industry experts on staff who will write a customized and personal letter by hand and ship it to your customer(s) on your behalf.
Let’s take a look at some of the trends and leaders in this space and see how they remain committed to personalization (both via a human connection and/or through machine learning) to cater to the needs of customers.
I’ll also discuss why it’s important to stay true to your internal culture and put your employees first – even when fighting external pressures to increase your bottom line as you grow.
Your Customers Want You To Get Even More Personal
Image via Monetate
A recent E-tailing Group Inc. survey revealed that “40% of consumers say they prefer buying from retailers that cater to their preferences.” And as the Monetate infographic above illustrates, the majority (75%) actually prefer that “retailers use their personal information to improve the shopping experience.”
So, as your online business grows and the volume of orders scales rapidly, there are ways to take some of your in-store or analog best practices online and provide solutions for customers based on their previous shopping behavior or browsing history with your website.
And there are some great examples of ecommerce companies who manage to stay true to their personal and high touch brand; even when they become mega-brands.
For example, companies like Zappos and ModCloth use digital marketing technologies to create a human and/or personalized connection with customers.
How To Personalize Your Digital Marketing Efforts
For years, Amazon has been using algorithms to make recommendations for similar books to the ones you just added to your shopping cart. But now with enhancements in mobile, social web and machine learning technologies, things are getting a whole lot more personal.
For instance, in 2014 Zappos launched a pilot program on Instagram – offering personal shopping recommendations to anyone who posted a selfie of their outfit of the day (OOTD) on the social photo sharing platform with the hashtag #NextOOTD.
Image via Instagram
The Zappos project required a person to manually respond to the photos –almost the same as when you buy a pair of pants in a store and the clerk recommends a belt or shoes to go with said pants. But as I’ll get to later on, machine learning technologies could help Zappos, and other retailers who want to launch similar programs, to scale and automate the process – all while maintaining a very personal shopping experience that caters to individual styles and needs.
Image via Monetate
Personalizing an automated email is another way to better serve a customer’s product tastes and get them coming back to your store. But the goal is to make customers feel like they are dealing with a human instead of a robot.
As I mentioned earlier, consumers expect retailers to collect personal information. As such, your email messages can be targeted to a user’s behavior and situation, including their location, purchase history and more (as outlined in the graph from Monetate above) – provided that you collect that information openly and protect their privacy.
Image via HubSpot
HubSpot has a great list on their blog of some leading personalized email campaigns. The screenshot from Modcloth’s email campaign showcases a great strategy that lets a customer know when a dress they placed in their shopping cart (but never purchased) is almost out of stock.
This tactic is very similar to when you place an item on hold in a retail store and the clerk calls you to let you know that someone else wants to buy it if you are no longer interested.
Not only does the email above convey a sense of urgency, but the tone (as HubSpot points out on its blog) of the email is tailored to the target customer as well. Although, ModCloth could have taken the message one step further by using the customer’s name in the message as well.
There are many other ways to personalize your digital marketing efforts as outlined in the infographic from Monetate. And in order to make the online shopping experience as seamless as buying in a store, retailers need to optimize the customer experience for mobile devices as well.
Tools To Help You Deliver Real-Time Marketing Personalization
While A/B testing can help you solve a small problem, like what headline to use on a blog, machine learning is valuable for helping businesses to solve multiple problem/solutions at the same time by helping you to discover the “customer rules” of personalization. Below are a few tools you might want to check out to help you achieve real-time marketing personalization and create deeper relationships with your customers.
This tool is designed to allow marketers to understand a large amount of context behind each user on-the-fly - without the help of your IT department. For example, if a user is already subscribed to your email newsletter, Commerce Sciences can help you identify those users so that they no longer need to see a promotion requesting them to sign-up for that program ever again. You can also set rules for showing promotions based on whether a customer is brand new vs. a repeat customer, and so much more.
Companies like Conductrics help marketers to convert more users via their technology which blends A/B Testing with machine learning to deliver optimal experiences for each user automatically. And you can test any digital event (via a website, app, email or even call centre) to later be optimized based on real-time user behaviors.
Ensighten enables you to collect user data across all of your customer touchpoints in order to help you create 1:1 personalized optimizations across all of your marketing and/or advertising channels. This tool offers an “open marketing platform” which integrates with a lot of “best of breed” applications in your marketing stack.
Questions To Ask Internally Before Using Machine Learning For Marketing Personalization
Matt Gershoff, CEO of Conductrics argues that real-time personalization, in its current state isn’t perfect and isn’t always necessary. “Personalization is going to help you half of the time,” he says. “That’s why it’s important to first identify when it is most necessary - so that you don’t misallocate resources or over-complicate the experience.”
For example, do you need to collect data on a user’s personal shipping preferences to predict whether or not to offer them free shipping? Probably not.
But Omri Yacubovich, Head of Marketing at Commerce Sciences says that “you should identify what country or region a customer is coming from in order to determine whether your free shipping offer applies to them personally. If, say, you only offer free shipping to your U.S. customers, then don’t upset your Canadian customers by displaying the promotion to them as well.”
Here are some questions that Gershoff suggests your team should be asking internally prior to allocating resources to collect data for personalization and prediction modelling:
Does this seem like a problem that can be solved with personalization?
What is the value of personalizing this particular customer experience? Is it worth it?
If your answer is yes to the first question, but you don’t know the answer to the second question, then it’s time to test the value of that experience using existing data - or tapping into a tool like Conductrics to collect the data efficiently and then model user attributes (i.e. their geo-data, the time of day, etc.) to various marketing and/or customer experience options in order to determine which option (say the decision of serving up a Las Vegas holiday content page vs. a family vacation content page to a user) has the best conversion rate or provides the most value to the end-user.
Looking Ahead: Machine Learning And The “Super Concierge”
Thanks to enhancements in big data and machine learning technologies, the tactics that I described earlier (and then some) are about to get even more sophisticated through curated and customized shopping experiences.
This New York Times article does a great job of explaining how startup ecommerce businesses like Stitch Fix are using your shopping preferences and profile data about other shoppers (who have similar fashion tastes as you do), to predict and recommend curated outfits that you might like. Then, a stylist at Stitch Fix hand picks and ships you a box of clothes to try on, and you just send back whatever you don’t like. The company then adds that information to their database to learn even more about your personal shopping style and preferences.
“’…there’s way more data science at work here than people may realize,’ said Bill Gurley to the New York Times. He is a general partner at Benchmark, a venture firm that invested in Stitch Fix. ‘There’s a 15-page profile, there are over 66 characteristics tracked and there’s a predictive heat score for every single item against every single user,’ meaning a way to determine the likelihood that a customer will keep an item.”
A recent whitepaper published by the product intelligence company Indix states that “by 2020, personalized and enhanced shopping experiences will become a reality attainable by those with smart devices.” In fact, future shopping experiences may involve interacting with
“a super concierge that attends to our product needs and wants. This concierge, an Ambient Shopping Assistant (ASHA), might bear a likeness to digital assistants used today, such as Siri and Cortana, but it will offer much more customized and intuitive services.”
This “super concierge” concept is not far off from what high-end brick and mortar fashion stores do already – which is get to know their best and most loyal customers’ personal tastes, clothing size and their personalities to recommend products, or even put something aside for them to try on when they are back in the store.
Online stores are already trying to blend digital experiences with traditional brick-and-mortar best practices. In fact, some pureplay online stores like Frank and Oak, have recently opened up retail locations where customers can come not only try on their favourite styles in-person, but they can also learn more about the retailer’s brand experience via in-house personal stylists, barbers, cafés, and regular live events.
“But in the future, personalization will involve a lot more native website components - allowing you to offer many different experiences for each of your customers,” says Yacubovich. “And cross-channel marketing will take a more dominant role which will allow retailers to integrate all of the insights that they capture about a customer (from both online and offline channels) to make the right decision, at the right time, and push the right call to action to convert that user.”
The Indix whitepaper argues that Netflix has already built the machine learning algorithm technology to enable retailers to offer an optimal personalization experience for customers. The whitepaper explains that “Netflix not only built the database, but also made it actionable by tagging each piece of content with a vast library of attributes, resulting in the formation of 76,897 micro-genres. Such a rich database enables Netflix to make personalized recommendations, and at the same time, it allows subscribers to look up movies using a combination of attributes.”
So, if retailers can use that same technology in the future to learn as much as they can about their customers, they will become increasingly better at making product recommendations and personalizing individual experiences to keep users coming back for more.
Closing In On A Future Personalization Panacea
In the future, personalization should be able to blend our individual and subjective interpretations of the world around us with machine learning. But the technology available today is not up to snuff.
This TechCrunch story contributed by Jarno Koponen, eloquently outlines the current challenges with machine learning technology that need to be resolved in order to paint a more accurate picture of customers and deliver an optimal personalization experience.
“Because of the personalization gaps and internal paradox, personalization remains unfulfilling and incomplete. It leaves us with a feeling that it serves someone else’s interests better than our own,” says Koponen via TechCrunch.
Below is a quick summary of the 5 gaps that he outlines in the article.
The Data Gap
A personalized offer or message is only as good as the amount of data a business has collected about you. Likewise, a recent eMarketer report found that “80% of marketers worldwide said they didn’t understand their customers beyond basic data such as demographics and purchase history. And unifying the data they did have was even harder, as 96% said it was challenging to build a comprehensive single view of a customer.”
The Computing Gap
Current computing power and machine learning technologies have many limitations. “Today’s fastest [and most advanced] systems become too slow when they try to understand the complexity of an individual on their own terms…[and they] haven’t yet provided a way for computers to seamlessly learn from us and adapt to us,” says Koponen.
The Interest Gap
There are many cogs in the personalization machine; including users, platforms and marketers; each of which have conflicting interests. The challenge for today’s machine learning technologies is to determine “whose interests and preferences are prioritized when deciding what you can see and do (e.g. being served an ad)?”
The Action Gap
Koponen argues that a user’s “actions are simplified to fit the [ecommerce] environment’s limited feedback loops.” In other words, there are certain actions that you may wish to take (or wish to exclude from your experience) on a website or app, but they simply don’t exist. This shortfall makes it difficult for retailers to learn from your true customer experience.
The Content Gap
It’s unlikely that the platform or application that you are using will consistently “have content that serves your exact intentions or needs. It also means that the diversity of the served content might be very limited.”
Getting Personal With Your Employees Has A Ripple Effect
Ever heard the phrase “it’s what’s on the inside that counts?” The same belief can be applied to a customer’s perception of your brand.
If your employees are happy and motivated from the within your company, they will be more likely to provide great service and your customers will keep coming back for more of the same great shopping experiences they have learned to expect from your online store.
So, it’s important to make your employees feel like a human, instead of a cog in a machine.
I know that I mentioned Zappos earlier in this post – and it could just be that I really love shoes – but I wanted to use this company as an example again. In this case, I want to discuss how Zappos is focused on “delivering happiness whether it’s for customers or employees.”
“Our whole belief is if you get the culture right, then most of the other stuff, like delivering great customer service or building a long-term brand or business will just be a natural by-product,” said Zappos CEO Tony Hsieh at the 2014 CinemaCon at Caesars Palace in Las Vegas.
This philosophy has enabled the company to project massive profit growth. And Zappos is still committed to staying true to its startup culture – even after being acquired by Amazon and generating billions in revenue.
Some of the Zappos “happiness” strategies include collaboratively writing a company culture book which includes contributions from employees about their vision and ideas about team culture.
In addition, the company has created ten core family values that every employee must embrace in order “to live and deliver WOW.” Hsieh also wrote a book called Delivering Happiness which beat out Oprah on the New York Times best sellers list.
And the Zappos “personal happiness” vision and culture is so important to the company that, according to greatplacetowork.ca, after an employee’s first week of training, they are “offered payment for their time, plus $2,000 to quit and leave the company before they fully assume their roles.” That’s because Zappos only wants to keep employees that strongly believe in its long-term vision and culture.
So, if a behemoth retailer like Zappos can commit to preserving its family-focused culture with employees – while continuing to delight customers with fantastic, personalized service (although it was a sad day when Zappos announced that they would no longer ship to Canada; their reasons for doing so can be saved for another post) – it’s clearly attainable for other budding ecommerce businesses to stay focused on people as they grow.
Have a tip on how to maintain a strong personal connection with customers – even as an ecommerce business scales in size? Please share in the comments section below.
About The Author
Andrea Wahbe is a freelance B2B marketing strategist and corporate storyteller who writes about Canadian SMEs, marketing and digital media trends. Follow her on Twitter.