Zeta Global and AI Powered Email Marketing
No matter what your firm sells, you need to first qualify the customer who you’re trying to sell to. Take car salesman as an example. These people are largely a drain on our society, but even they know that the most important thing is finding out if you’re a serious buyer as quick as possible. They also do isht like try to keep you physically in the dealership for as long as possible so NEVER give them your car keys or you’ll be there until they close. Also, never talk to your partner in those little rooms they lead you into as they’re recording what you say when the salesman leaves to go “talk to the sales manager”.
So anyways, we were chatting a while back with some of the folks at Afiniti who said that we’re moving towards a society where facial recognition will identify you when you enter a store and immediately qualify you based on your income, current assets, and lifestyle. It will even go so far as to let the sales team know your political affiliation, sexuality, and hobbies in order to match you up to the salesperson that’s most likely to sell to your profile. While that’s a great idea for bricks-and-mortar situations, what about websites? How can we turn website visitors into customers? This is where companies like Zeta Global are doing some amazing things.
Founded in 2007, New York startup Zeta Global has taken in a staggering $310 million in funding at a valuation that gives them unicorn status ($1.3 billion to be exact). One of the founders of Zeta Global, John Sculley, is somewhat of a legend having served as the president of Pepsi and the CEO of Apple where he moved sales from $800 million to $8 billion in ten years’ time. This new venture of his does “customer lifecycle management” which sounds suspiciously like a business that uses information about what you’re doing with your life to sell you stuff. As it turns out, that’s exactly what they do. According to an article by CNBC that talks about a possible Zeta Global IPO, the CEO made this thought provoking statement:
“We would use our database of 350 million people. We could identify a cluster of 7 million people who look like your best customer. We would reach out to them via messaging, social media, and based on who interacts with messages and clicks stuff, the machine learning platform gets smarter and fine tunes who you are a looking for,” CEO David Steinberg told CNBC by phone.
Remember that example we gave earlier about being identified by facial recognition for targeted selling? Well, this database of 350 million people (roughly the entire population of the U.S.) is pretty much what you need to make that happen. You can bet that every single thing you do online that can be tracked ends up finding its way into this database and we’d bet that if you saw how much they know about some people, you’d probably be shocked. That’s because all the morons out there that make their social media profiles public and share every iota of their mundane lives with the world are pretty much making business models like this happen. We talked about the implications of big data privacy and the truth is, most people just don’t care enough which makes this a very lucrative business for Zeta Global which is now operating profitably and prepping for an IPO.
In order to understand what you might be able to accomplish using all of Zeta Global’s “big data”, we’re going to use Nanalyze as an example. Right now you’re reading this article and you could be a first-time reader or someone who reads our great content on a daily basis. While we don’t know much about you, we do know if you’re a first time reader (simple cookie stuff). If you are, we throw up a “lightbox” that asks you to subscribe to our newsletter, like this one:
Once we have your email, we’ll then do a newsletter blast to all our subscribers whenever we have about 5-8 articles. Typically about 20-25% of you open the email and 6-8% of you will click on an article. That’s a pretty consistent number we’ve been seeing over time and we don’t care too much about it since we’re so busy churning out insightful content for our lovely readers. However, now that we’re achieving some real momentum with our brand, we’d like to increase the click-through rate for our newsletter which would result in more traffic and ad dollars which we use to feed our MBAs and keep the fridge stocked with craft beer. Now that you understand the problem we’re trying to solve, take a look at what Zeta Global says they can do using all their big data and a tool called Zync:
Those numbers are absolutely amazing. There is no way that you could achieve those kinds of results using traditional A/B testing. If you’re unfamiliar with how A/B testing works, essentially you just try two things (A and B) and then choose the one that’s better. While that’s oversimplified a bit, it’s basically what you need to do in order to improve “click-throughs” and whatnot. It’s a painful process and it takes a long time. Zync shows us how using big data to understand your customer can have a tremendous impact on optimizing your key marketing metrics. Now, imagine if you combine that big data with AI? That’s exactly what Zeta Global had in mind when they acquired Boomtrain this week.
We previously wrote about 8 Artificial Intelligence Startups Improving CRM and Boomtrain didn’t make that list because we basically couldn’t include everyone. Now we’re wishing we had, as what they’re doing is frankly fascinating. Founded in 2012, San Francisco startup Boomtrain took in $14.77 million in funding from investors who are probably pretty happy right about now. Zeta Global acquired Boomtrain just this past week for “between $35 million and $40 million” according to an article by Tech Crunch. What Boomtrain does is to use AI for improving the effectiveness of online marketing. Remember how we keep saying that you can tell which startups are actually using AI based on the results they get? Boomtrain is one of those examples.
According to a report by Boomtrain based on data from over 230 million emails sent using the Boomtrain platform, the open rate for AI-personalized emails was a staggering 63%. It’s a process they call “AI optimization”, and it was used by 65 companies over a period of around 100 days to produce some astounding results. Here’s a look at the entire process (yellow highlight is ours):
That process of 1:1 personalization you see right there resulted in an average click-through rate improvement of 556%. Now imagine if you hooked those AI algorithms up to the big data set over at Zeta Global? That’s the intent of this acquisition as Zeta Global plans to use Boomtrain’s AI expertise and apply it across their entire CRM product offering. Just the fact that they can figure out the optimal time each person might open an email is a pretty cool idea alone, not to mention that they figure out what people’s interests are. We use a tool called AWeber that sends all the emails in one blast, then tells us what articles were clicked by how many people. Imagine if you could track that for each person individually and then use AI to customize every single email you send out?
We were so impressed with what Zeta Global claims they can do that we’re considering giving this a shot ourselves. We’re actually in the process of working with “John in Mumbai” right now to redesign our newsletter and he just sent us pretty much what we had before with a few subtle differences (rolls eyes). While he probably thinks that’s the easiest $200 he ever made, we’re going to have one of our MBAs who lifts every day harass him a bit. Hopefully she makes some progress with that, and in the meantime, we’re going to have a chat with
Boomtrain Zeta Global and see if maybe we can do a real world case study with them to share with our regular readers, the ones who click through that is.