MarTech 2020: How to use AI to 2x Your Customer Engagement

Speaker:

• Josh Francia: Chief Growth Officer, Blueshift

Q: What is the Holy Grail of Marketing?

Josh Francia: Hi, everyone. My name is Josh Francia, and I'm the Chief Growth Officer at Blueshift. Today, I'm going to talk about how you can use AI to 2X your customer engagement.

I've been with Blueshift for about a year and a half. I was a former customer of Blueshift during my time at LendingTree, where I was the Chief Growth Officer. So I have a very strong passion for customer engagement, a strong passion for using data, and I love talking about these things.

Blueshift is a Smart Hub CDP. We believe that we are helping marketers with one of their biggest problems: figuring out how to use data to be more efficient. Today, we're going to talk about:

• What AI specifically works for marketers.
• Evidence of AI actually working and the results it produces.
• How you can get started with AI and make it valuable for your brand.

The holy grail of marketing is essentially figuring out:

• Who is likely to purchase (who to target).
• What content, product, or offer to recommend.
• When is the right time to engage.
• Where is the right place to engage.

This is the holy grail. It starts with having a unified view of your customer, but then the challenge is taking all that rich data and making sense of it to fill in these gaps. If we can figure out the who, what, when, and where, then one-to-one personalized campaigns at scale are not only possible, they are very easy to accomplish.

Q: What are the Hurdles to Personalized Marketing?

Josh Francia: There is a journey to get from where we are today—with many marketers stuck in static engagement—to intelligent, always-on, automated customer engagement.

We did a study about a year ago and asked over 500 marketers how many of them are truly personalizing at a one-to-one level. Only 5% of marketers said yes. This means 95% are not doing one-to-one personalization.

When we asked them about the hurdles, here were the top answers:

• Data wasn't collected or unified effectively. You must have unified customer profiles to do anything personalized.
• They couldn't run complex use cases efficiently, often getting stuck doing things manually.
• They were using outdated or obsolete tools built 10, 20, or 30 years ago that couldn't keep up with modern needs.

So, how do we get there? Our co-founder and CEO, Vijay Chittoor, says: "AI is the key that unlocks the door to scalable and sustained growth."

When we surveyed marketers who were using AI, 98% said they saw business improvements. And 81% of marketers using AI for personalization reported exceeding their revenue goals by at least 30%. So, while not everyone is using AI effectively, those who are are seeing great results.

Q: What is the Difference Between Static Rules and Dynamic AI?

Josh Francia: We have two paradigms: the old world of static business rules and the new world of dynamic AI.

Static Business Rules:

• Static: You set the rules once and have to manually change them.
• Unscalable: There are only so many rules you can write before it becomes unmanageable.
• Inefficient: It requires more human effort to manage the processes.
• Emotionally Driven: Decisions are often based on subjective feelings rather than data (e.g., "we're sending too many messages," "we shouldn't push this product").

Dynamic AI:

• Dynamic: AI constantly improves itself based on new data flowing in.
• Infinitely Scalable: It runs on computers, allowing for billions of records.
• Operationally Efficient: You use CPUs and servers, not humans, to run and manage the AI.
• Data-Driven: AI is not emotional. It looks at the data and makes scores and adjustments based on it.

When implementing AI, look for a solution that is:

• Transparent: Avoid black-box models. You should be able to understand the inputs and outputs.
• Simplifies Complexity: It should be easy to set up, train, and test, while still handling complex decisions.
• Actionable: It should allow you to activate and take action on the signals and scores it produces.

The co-founder and CEO of our partner, Made in Cookware, said: "AI and machine learning are the future of marketing automation." The holy grail of marketing is only possible when you use AI. You can't reliably do the who, what, when, and where without it.

Q: What is the Evidence of AI Working?

Josh Francia: We've heard about AI a lot, but does it actually work? Let's look at some examples of brands using AI to improve their KPIs and metrics.

• Skillshare (The Who): This online learning company used predictive scores to identify who to target. They saw an 89% improvement in enrollment rate. One of their marketing leads, Brooke Young, said AI helps their team "appear larger than it is" and allows them to speak to customers with a sophistication "like companies 10X our size." This shows how AI enables smaller teams to do more.
• Zumper (The What): This online apartment rental platform tested predictive recommendations to determine what content to recommend. They saw a 381% improvement in lead volume and a 128% improvement in click-through rates. The AI recommendation engine allowed them to better serve customers with targeted marketing and personalized campaigns at scale.
• LendingTree (The When): This financial marketplace tested when to message customers with triggered alerts. They saw a 35% lift in revenue by using engaged time optimization to predict the best time to send a message. This was a simple change (a checkbox click) that resulted in a significant revenue improvement.
• CarParts.com (The Where): This auto parts retailer tested predictive channels to figure out where to engage with customers. They saw a 400% engagement lift. Their CMO, Humlin, said the platform's flexibility makes it simple to deliver personalized messages and grow their channels.

These examples show that AI helps brands deliver the right experience, which in turn drives top-line business results.

Q: How Do You Implement AI for Your Brand?

Josh Francia: How do you implement AI for your brand? It really starts with data. You need to have all your data in one place. From there, you need to be able to understand that data, run predictions against it, and then orchestrate that data everywhere.

Blueshift, as a Smart Hub CDP, lives at the core of this. We unify your data, run predictions, and then allow you to orchestrate not just to normal marketing channels but also to every other part of the customer experience.

LendingTree & Zumper Implementation Examples

Here's an example of how LendingTree does this:

1. They score all customers based on which product they are most interested in.
2. They build segments based on these scores, biasing their campaigns toward higher-revenue products.
3. They push these segments to paid media destinations.
4. Their ads then reflect the product the customer is interested in, based on the probability score.

Zumper does the same thing, based on user-initiated behavior:

1. A customer visits their website and favorites an apartment.
2. The AI runs in real-time to figure out what other apartments they might like.
3. They message the customer with recommended apartments via email and push notifications, all based on a very simple signal (one wish list item).

This is a high-level view of how AI can work for your company and marketing. I'm a big believer in AI. I believe that when brands allow AI to lead with guidance and by looking at the outputs, they can unlock unbelievable growth and sustain it over time.

Q: Where Can I Find More Resources?

Josh Francia: We recently launched and it's now available, the "Ultimate Guide to AI Marketing." This is a comprehensive guide to help you understand how to get started with AI marketing and includes a lot of evidence of how it works. It's available today at blueshift.com.

Thank you so much for participating and listening for these 15 minutes. If you do want a free AI marketing consultation, you can reach out to us at hello@blueshift.com or go to our website. Thank you so much, and I wish you the best of luck in your AI marketing initiatives.