Rivalry2

HTML Diff

3 added 4 removed

Original 2026-01-01

Modified 2026-03-10

1 Ok, so you’re at the point where you’ve built out a pretty sophisticated email marketing strategy. So… what’s next? What’s on the horizon?

2 The buzz in the marketing world today is all about artificial intelligence and machine learning.

3 Marketers seem eager to go all in on AI, but there seems to be a difference in opinion around how the terms “machine learning”, “artificial intelligence”, and “data mining” are being used.

4 A number of companies that claim to make use of artificial intelligence are actually really marketing machine learning techniques.

5 - By 2020, 85% of customer interactions will be managed without a human.<a>Gartner</a>

5 + By 2020, 85% of customer interactions will be managed without a human.<a>Gartner</a>

6 - Artificial intelligence will replace 16% of American jobs by the end of the decade.<a>Forrester</a>

6 + Artificial intelligence will replace 16% of American jobs by the end of the decade.<a>Forrester</a>

7 Confused? Let’s break down the true differences:

8 <h3>Machine learning</h3>

9 When you think of machine learning, think about basic automation where you have a list of steps that you have taught a computer to follow. With machine learning, you have a set of defined questions that you are looking to answer, with different known variables that ultimately help you figure out what matters.

10 An example of machine learning would be calculating the optimal time to send a campaign or offer a discount based on data you already have. While it would be possible for a human to pull the data points needed and make an assessment, repeating this process 1,000, 10,000 or even a million times would not too feasible to scale. Machine learning has the ability to automate this process.

11 <h3>Artificial intelligence</h3>

12 With artificial intelligence, there are two pathways forming: narrow AI and more general AI. Narrow AI is more commonly what we see today in marketing technology and AI at large.

13 Narrow AI could best be described as having a specific question in mind that you’re looking to find an answer to, such as “How can I improve the revenue of my campaigns?” or “When is the best time for our company to send emails?” While machine learning is built on basic automation of identified data points, narrow AI can take into account things that you might have missed earlier or overlooked previously (event tagging, for example).

14 General AI falls more into advanced technology like Apple’s Siri, Amazon’s Alexa, and IBM’s Watson. Each of these technologies, dubbed digital assistants, can answer a seemingly endless amount of questions. But not every question (digital assistants can’t cook you dinner or take your dog for a walk).

15 <h3>What AI for Ecommerce truly looks like</h3>

16 At Klaviyo, we can boil that down even more. Unlocking AI for ecommerce will come when it’s possible to tell marketers what they should do next to get the best results for their business without having to know which questions to ask.

17 Here’s what that means. Instead of focusing on how to improve certain data points, like click-through rates and opens, artificial intelligence for ecommerce will automatically let you know what exact actions you should take to improve revenue and even goes so far as to tell you how much revenue you’ll see for each action. Imagine logging into your marketing automation platform and being automatically alerted to the top actions you could take to increase revenue for your business. AI for ecommerce would allow marketers to spend less time poking around in their data. Instead, they’d enjoy being proactively told the exact actions they needed to take to help their business.

18 - Right now the only company that is really at this point of sophistication in the sales and marketing world is <a>Salesforce</a>. The technology they have built allows sales reps to know which leads to prioritize using their artificial intelligence tool, <a>Einstein</a>.

18 + Right now the only company that is really at this point of sophistication in the sales and marketing world is<a>Salesforce</a>. The technology they have built allows sales reps to know which leads to prioritize using their artificial intelligence tool,<a>Einstein</a>.

19 And the reality is, while there are an ever-growing number of marketing technologies that claim to be AI-powered, the ability to really take AI and effectively apply it to ecommerce marketing is something that is still limited to a very small number of companies. Why? Because it requires a lot of time, money, and manpower. A small group of marketing technologists claim to do this, but the current state of A.I. and predictive analytics is underdeveloped.

20 In fact, the current state of AI and ML as it applies to ecommerce email marketing is not unlike where email marketing itself was not all that long ago: it’s expensive, takes an army to implement, is overly complicated to use, and vendors promise more than they can deliver.

21 In order to get to 100% for ecommerce marketers, the technology needs to not require heavy lifting and be able to reassess and learn what your subscribers need with each interaction.

22 -