Source: amazon.com/samples
Will Amazon succeed with brand sampling rooted in machine learning?
Through a special arrangement, presented here for discussion is an excerpt of a current article from the blog of Kiri Masters, founder and CEO, Bobsled Marketing. The article first appeared on Forbes.com.
Amazon.com is experimenting with free product samples as a paid advertising service for CPG brands.
According to the program’s landing page, the samples are “like Amazon’s product recommendations, but real,” available for Prime and non-Prime members. The effort is separate from Amazon’s Prime Sample program, where Prime members pay to receive a box of product samples.
A job listing found by Axios for a senior software engineer describes the “Targeted Sampling” program: “Free samples of new products are sent to customers selected using ML (machine learning), thus ensuring a higher likelihood of conversion than display ads. The program has a challenging mix of problems involving targeting, fulfillment, customer and vendor experience, and cross-campaign learning.”
And thus underlies the strength of a sampling program designed by Amazon — an additional advertising option for CPGs eager to harness the reams of consumer shopping and intent data that the e-tail giant holds. Product samples are widely used by CPG brands to activate new customers. But Amazon is able to pair fulfilment of sampling with its proprietary shopper data and make sure that only highly interested, “in-market” customers receive the samples.
Amazon is already the third largest digital advertising platform in the U.S., according to eMarketer. Adding product sampling to its array of advertising options for large brands is almost a no-brainer — it drives more sales on the platform, locks brands into marketing spend on the platform and presumably brings in advertising revenue of its own accord.
Managing the logistics of product samples is also something that the Google-Facebook advertising duopoly just can’t match.
Still, some brands may be concerned that Amazon may choose to prioritize its own private label brands ahead of third-party CPGs in the future.
Amazon also owns the customer. Relative to the vast array of shopper information that Amazon collects, vendors and sellers on the platform are able to access very little customer information. Sellers and vendors are expressly prohibited from contacting customers outside of specific customer service requests or remarketing to them
The program may also face resistance from consumers who recognize that their browsing and shopping data is being used within Amazon’s advertising programs.
Discussion Questions
DISCUSSION QUESTIONS: Does Amazon’s free samples program hold more benefits than risks for CPG brands? Is it a win-win-win for Amazon, consumers and brand?
Poll
BrainTrust
Lauren Goldberg
Principal, LSG Marketing Solutions
Jennifer McDermott
Consumer Advocate, finder.com
Amazon’s product sampling program helps bridge the gap between e-commerce and brick and mortar. Samples have long been part of grocery stores and other retailers and can be a good way to introduce customers to a new brand. Using samples in e-commerce will give customers more of an in-store experience where they can handle and try the product. With its vast shopper data Amazon can easily identify the “in market” customers more likely to convert, the tricky part will be the costs involved in packing and shipping.
Any opportunity brands have to engage, entertain and gather new insights is a win, and in this case a big win for Amazon as well. Leveraging machine learning and AI to truly know your customers, their preferences, and shopping history to drive a personalized experience is the right strategy to employ.
For the brands, this has historically taken place in the wholesale and brick-and-mortar channels. L’Occitane en Provence, Sephora, and other beauty brands always provide free samples, with the hopes that the consumer will return to buy these products. Yet, now with the power of machine learning, Amazon will be one step ahead of the competition and know what similar or complementary products will drive additional engagement.
Yet as it is described in the article, from a brand standpoint this is a “black box”: give money and free product to Amazon and you *might* see increased volume, but whether you do or don’t there is no transparency or audit trail. Anything I’d like to learn as a brand — What kinds of consumers were targeted? In what browsing/search/buying situations were shoppers sampled? How many sample sessions led to conversion? — none of that is going to be available. Shoppers may be delighted with Amazon to be sure, but at the expense of the brands.
This is really advertising taken to the next level and is, essentially, another way for Amazon to make money by leveraging its vast logistics capability and its massive network of shoppers.
I guess if customers are already using Amazon to purchase CPG products there is little harm in exposing them to samples of national brands. However, for general Amazon customers not buying CPG online there is some danger in bringing them into Amazon’s web. Once they’re there, Amazon can target them to switch to its own or other brands as best suits.
It’s the age old dilemma. Amazon brings great benefits, but there are also great dangers in becoming too reliant on it as a sales channel.
This could be a big “win-win-win” for all involved. Amazon has so much data and can provide samples in a very targeted way, as opposed to the brick-and-mortar model of anyone who walks by. However, to make it a real “win” for the CPGs, Amazon must be willing to share some of the data to help understand the ROI.
Amazon posted a job opening for a senior software engineer, defining the targeted sampling program as: “Free samples of new products are sent to customers selected using Machine Learning, thus ensuring a higher likelihood of conversion than display ads. The program has a challenging mix of problems involving targeting, fulfillment, customer and vendor experience, and cross-campaign learning.”
Amazon is America’s third largest advertising platform. With compelling data regarding not only “likes” or “searches” or “preferences” but actual purchases, the addition of physical product sampling is a powerful tool and a major competitive advantage.
Sampling has a long and respected history in the retail world as an effective form of persuasive communications that is welcomed by shoppers. Within stores, its pretty much a “spray and pray” proposition with fairly low conversion rates. Think of the food sampling inside Trader Joe’s and Costco.
Why wouldn’t Amazon seek to use its delivery channel to provide a more targeted sampling service? With millions of boxes being delivered daily filled with air pillows, it would be a simple matter to insert sample products on the fulfillment line. The ride-along costs would be near zero.
Targeting would make straightforward use of Amazon’s customer data to match manufacturer samples with likely consumers. Machine learning may be overkill for this, but it would make sense to leverage the existing engine, whatever it is.
From the shopper’s perspective, well-considered samples and coupons could be perceived as a welcome surprise – in New Orleans they call this little something extra a “lagniappe”.
Amazon doesn’t make many (any?) moves without the data to support it. I think they’ve found a sweet spot here to deliver more value to both consumers and the brands. And it does create a platform to compete, in some cases with Amazon’s own brands. If Amazon creates fair competition by getting good products at good value in front of the consumers, let the best brand win. But Amazon doesn’t always play fair.
Amazon’s Product Sampling Program for CPG Brands will be a great enabler for new product adoption but considering the past history of similar other maneuvers by Amazon it could also prove to be a double edged sword for CPG brands, as there is probability of Amazon launching its own private label of similar other successful products. Therefore CPG brands should tread carefully in this direction.
This is definitely a new success stream for Amazon. Sampling leads to conversion, greater brand loyalty (mostly for Amazon than the products being sampled) and higher total value of purchases. Not only will they be able to make money off selling the program and access to highly targeted consumers to retailers but also, the customers who will sample-then-buy. Genius.