In this week’s episode of Growth Interviews, we invite you to join our podcast conversation with Jim Sterne, internationally known speaker, author of a dozen books on advertising, marketing, customer service, email marketing, and web analytics, founder of the Marketing Analytics Summit (eMetrics Summit), the Marketing Evolution Experience and co-founder of the Digital Analytics Association.

Machine learning is changing how we look at data.

Here Are The Biggest Takeaways

  • Beginnings in information technology – 01.25
  • Big AHA moments and future technologies – 04.15
  • How machine learning is changing online marketing – 07.19
  • How efficient are machine learning and AI patterns – 13.09
  • Practical ideas on how to use machine learning for growth – 15.55
  • Mistakes in eCommerce – 22.02
  • Why are companies missing out on customer experience – 26.00
  • The road to becoming customer-centric – 34.25

Welcome to Growth Interviews!

Welcome to Growth Interviews, the fun, stimulating, and engaging series of conversations driven by digital business growth. Our mission is to provide valuable insights from the eCommerce arena, and each episode is a fascinating quest into the best-kept business secrets and money-making strategies of an insightful world-class expert.

Jim Sterne sold business computers to first-time owners in the 1980s, consulted and keynoted about online marketing in the 1990s, founded a conference and a professional association around digital analytics in the 2000s, and founded the Marketing Analytics Summit (formerly eMetrics Summit), chairing 100 conferences around the world. 

An internationally known speaker and consultant to Fortune 500 companies and Internet entrepreneurs, Jim has spent more than twenty years in sales and marketing most of that measuring the value of Digital media for creating and strengthening customer relationships. He’s authored books on Internet advertising, marketing, customer service, email marketing, web analytics, and most recently, Artificial Intelligence for Marketing: Practical Applications.

Jim was named one of the 50 most influential people in digital marketing by a top marketing magazine in the United Kingdom and identified as one of the top 25 Hot Speakers by the National Speakers Association. We had a wonderful talk that will give you fresh, new perspectives about the digital landscape, the history, and future development of technology.

Machine Learning is Changing Online Marketing

Illustration of people working on a large robotic head, with labels for concepts like artificial intelligence, virtual automation, smart learning, and self-correction.

Machine learning is capturing everyone’s mind as airplanes did over a hundred years ago. Technological advancements are perpetually fascinating humankind and the peak subject of our times is machine learning and AI. Jim Sterne, one of the most brilliant tech teachers of our times, told us in one of the most captivating Growth Interviews episodes that “First of all it [machine learning] is changing how we look at data.”

Our interesting times are first and foremost, fast. We sell and buy fast, we gather data fast and the databases are growing even faster. We are past the times when the data just fit in a spreadsheet and we could target our customers based on two marketing personas.

Analyzing Big Data with Machine Learning

Infographic titled "Big Data" highlighting five key aspects: Volume (huge amount of data), Variety (different formats of data from various sources), Value (extract useful data), Velocity (high speed of accumulation of data), and Veracity (inconsistencies and uncertainty in data).

Now we deal with big data and an incomprehensible number of permutations that are all saying something about our customers. We understand it but the numbers are overwhelming. This is the arena where machine learning technology comes in and analyses the incredibly vast amount of data with the scope of finding correlations that will later be interpreted by an expert analyst.

Customers are becoming more complex in their actions than ever before. They have a lot of diversity, a lot of impulses and they have become more demanding than ever, an aspect that raises different, more complex questions. Machine learning technology is changing the way we look at the data, the correlations and its interpretation, helping marketing specialists answer more complex questions.

More than helping with bigger, more sophisticated answers on the clustering, segmentation, and behaviors of our customers, machine learning technology is an extraordinary enabler of faster, more qualitative work helping all professionals augment themselves and become more efficient.

The Efficiency of Machine Learning and AI Patterns

Infographic titled "The Seven Patterns of AI" showing: Hyper-Personalization, Recognition, Conversation & Human Interaction, Predictive Analytics & Decisions, Goal-Driven Systems, Autonomous Systems, and Patterns & Anomalies.

We live in historic times of the early days of machine learning and artificial intelligence when even the biggest companies are failing to obtain better results with the latest technological findings. The talks about the great opportunities these technologies are offering are much bigger and much louder than the actual performance and success obtained until now and many fail to recognize that these technologies are still in a phase of experimentation, given the vastness of their computing capabilities.

Although there are many breakthroughs and we see progress every day, through new tools that are getting on the market, all tech experts know that it’s a long and bumpy road that could lead in a thousand different directions and the critics are coming too early. As Jim Sterne stated during our interview “I’m going to give Google Analytics a little bit of slack, I’m going to cut them a break because they’re trying stuff they’re inventing.“

Chatbots: A Case Study in Machine Learning

Illustration of a friendly robot next to a smartphone screen showing a chat interface.

Like any other tool, technological tools are always great when they are used with a clear purpose and when their limits are considered and understood and not disregarded. During our talk with Jim, we opened up the subject of chatbots as examples of tools using machine learning technology. “They’re stupid, they’re a total waste of time at the moment, for most companies. But there are a couple of companies that have gotten it right and are doing such a good job,” said Jim.

It’s not a case of intelligence but a case of understanding the company’s need and capacity to understand, adopt, and properly use a machine learning-based tool. Jim gave us a great example: “I talked to a fellow at and they do all kinds of testing and all kinds of analytics there. 

They’re very happy to talk about the work they’re doing, in public, so it’s easy to find stuff on SlideShare and YouTube. I said ‘You know? What about chatbots?’ ‘Oh yeah, we’re using chatbots in a limited way because when people come to the website, there’s a handful of questions so we can answer them pretty quickly. And it’s resulted in several million dollars of uplift in a very short period.’”

Using chatbots at such a scale is resulting in gaining tremendous value not necessarily from selling more or better, but in reducing the cost of extensive support teams that always need continuous training due to personal change.

This is the case of which understood the capabilities and limits of chatbots and is using them to answer a set of questions that are particularly frequent and replaced the intervention of customer support with quick comprehensive answers available on the website, that are, in many cases, leading to a new sale.

Like any other new technology machine learning can be seen as good or bad, useful or not, but as practice shows the success of its use depends on the clarity of the intention of use, the resources involved, and the profound understanding of its capabilities, limitations, and capacity to fail in a particular direction.

Machine Learning for eCommerce Growth

Infographic titled "ML Use Cases in E-Commerce" highlighting various applications: search results ranking, building deals and bundles, price forecasting, selecting the seller, review and rating quality, product content quality, demand forecasting, and detecting malicious returns.

Probably the most popular branch of artificial intelligence, machine learning is a method of data analysis that automates analytical model building, being based on the idea that systems can learn from data, identify patterns, and make decisions with minimal human intervention.

What machine learning requires is searching through data to look for patterns and adjust program actions accordingly, learning from previous computations to produce reliable, repeatable decisions and results. As models are exposed to new data, they can independently adapt, creating new outliers, segments, clusters, and categories.

The Importance of Statistical Results

Diagram of Statistical Research Methods encircled by various types of analysis: Descriptive, Inferential, Predictive, Prescriptive, Exploratory Data, Causal, and Mechanistic Analysis.

A big supporter of machine learning, Jim Sterne outlines in our podcast interview the importance of the relevancy, the speed, and the amount of the new findings this technology delivers in comparison with the human effort, emphasizing the importance of the statistical rather than logical results that offer the analysts accurate information. “The statistical rather than logical is always fascinating because it discovers correlations that don’t make sense and yet are true,” points out Jim.

New, relevant, and complex information about customers and their behavior is, for any eCommerce, a goldmine. Knowing where to look more closely, and what interface, design, or sequence of actions to use is what every digital entrepreneur wants.

 So, working with machine learning tools that can give more information in a short amount of time and an organized, comprehensible manner should become the norm for any competitive eCommerce. “Machine learning is great at grouping, it’s great at finding the outlier, it’s great at segmentation. Ten different kinds of people come to your website so figure out ten different ways to treat them, to sell more stuff.”

Machine learning applications are valuable and complex, and demonstrate a lot of potential in all areas of digital, from advertising to loyalty and retention. The most important aspect though is not what the machine learning technology can do for a business, but what are the needs of the business that machine learning can help with. 

“The question is what’s the most important thing for you to accomplish at the moment? What is your most critical key performance indicator, what is the problem you’re trying to solve at the moment, and if you have enough data and if you have tolerance for failure and experimentation machine learning can help you.”

Figuring out where a company needs help and improvements through new technological developments is the primary problem for the management of a digital business and, in Jim Sterne’s vision, it consists of three jobs only. What problem do we want to solve? Which data should we give the machine to think about that problem? Does the output make sense?

Biggest Mistakes In E-commerce

Illustration of an online shopping scene with a shopping cart, credit card, and product images on a tablet screen, highlighting e-commerce activity.

E-commerce is one of the fastest-growing industries in the world and there is no sign that in 2020 its evolution will slow down. But despite its growth in dimensions and the complexity of its processes, eCommerce entrepreneurs tend to repeat the same mistakes over and over again. 

The first biggest mistake, in Jim Sterne’s experience, is high expectations and chasing after “the shiny new object” which is, of course, every new trend and technological tool that seems to solve all the problems of the business. eCommerce is so vast that in many cases it is almost impossible even to compare two online stores selling the same merchandise in two different countries. 

The best an entrepreneur should do is collect the best cases, analyze the conditions on the market, and extract and use what is valuable. In the rush of earning new clients, new market share, and getting ahead of the competition, many entrepreneurs are adopting other success recipes without analyzing the conditions. Jim Sterne advises that before investing in the next hype thing everyone should make a proper test and find if it’s working for their niche.

“I’m going to say number two is not combining marketing processes. So I’m thinking now as a buyer and it just drives me crazy that I buy something on Amazon and then Amazon advertises that thing to me for the next month. I mean retargeting works. It’s great. It’s powerful but for heaven’s sake, I bought it from you already. Don’t advertise to me again. And why? The e-commerce database doesn’t talk to the advertising database and that just drives me crazy.”

Retargeting has proved to be one of the best marketing tactics for several years now, and a pretty cheap one, too. All digital companies, in B2C and B2B alike, are using it as a common practice already, but unfortunately, in too many cases, the retargeting includes the recent buyers of the advertised products, creating discomfort and negatively impacting the customer experience.


Machine learning technology is creating a lot of opportunities for improving eCommerce as we know it. Although the tech is here and its applications are many, the most important question is not “How will the machine work?’ but “With what purpose does the machine work?” 

We hope you enjoyed our video interview with Jim Sterne!
For more valuable insights, make sure you come back to check out our next Growth Interviews as well. 


What are some practical steps for a business to start integrating machine learning into its eCommerce platform?

To start integrating machine learning into an eCommerce platform, businesses should first identify the specific problems they want to solve, such as improving customer segmentation or personalizing recommendations. Next, they should collect and organize relevant data for training machine learning models. Partnering with experienced data scientists or using machine learning platforms can streamline the implementation process. 

Additionally, businesses should continuously monitor and evaluate the performance of their machine-learning applications to ensure they are delivering the desired outcomes.

How can machine learning improve the customer experience in eCommerce?

Machine learning can significantly enhance the customer experience in eCommerce by personalizing shopping experiences, predicting customer preferences, and providing relevant product recommendations. It can also optimize search results, streamline customer service through chatbots, and detect fraudulent activities. 

By analyzing customer behavior and feedback, machine learning algorithms can identify patterns and trends, allowing businesses to tailor their offerings and improve overall customer satisfaction.

What are the potential challenges businesses might face when implementing machine learning in their eCommerce operations?

Businesses might face several challenges when implementing machine learning in their eCommerce operations, including data quality and availability issues, the complexity of integrating machine learning models with existing systems, and the need for specialized skills and expertise. 

Additionally, there can be resistance to change from staff and management, as well as concerns about data privacy and security. Businesses need to have a clear strategy, adequate resources, and a commitment to continuous learning and adaptation to overcome these challenges effectively.