AI Can Help Marketing Get Personal

As sophisticated as we are in this age of digital age, we collectively lack imagination around using AI (Artificial Intelligence) in marketing. I am convinced that AI can enable dynamic, value-added, micro-engagements with every person of interest: from influencers to prospects, to customers. And AI can do it without requiring new platforms. it can work around and with your current platforms. AI has already arrived in some many other aspects of our lives. So why not in marketing?

We are Comfortable with Mediocracy

In this golden age of digital marketing, we have amassed plenty of data on every person who matters. Between the MarTech vendors (all 5,381 of them!) and many of us B2B digital marketing practitioners, we capture every click our customers and prospects make. We inhale a significant portion of their digital exhaust every day, and then we use it to profile them to death. Then we dutifully park all that great intelligence in our disparate, proprietary data lakes and turn to a mythical, God-like creature we call a Data Scientist to bestow an "aha moment" on us to help us run better marketing campaigns.

Then somewhere along the line, a marketer translates the "aha moment" (assuming our Data Scientist delivered one,) into a segmentation strategy to target specific groups of prospects and customers. And we construct these targeted segments with the premise that certain prospects belong together just because they happen to be in the same function or level, or they work in the same industry and their HQ is located in the same region. Our creative agency then builds a series of quasi-personalized emails and webinar invitations that we send in batches a few hundred thousand people at a time.

Of course, more often than not, the people we lumped together into segments are in dramatically different stages of the buyer's journey and have unique pain points and motivations that color their decision-making process. Yet we still lump them into segments. Wait, what? Really? The more I think about it.....segmentation is so last century and such a severe underutilization of the capabilities we have now!

As a result of all this dysfunction, we end up sending out emails that yield click-through rates of 1-2% and if we are lucky, maybe 0.5% of them end up as a lead. And we actually feel great about those numbers because everybody else also gets horrible results and we end up in "benchmark" territory! Come on now. We can do better than that. Where's AI in all of this? Why isn't AI all over this challenge?

Segment Marketing is Dead

Let's think about this. We all collect billions of bits of individual person-based intelligence, down to click-stream data, demographics, psychographics, and firmographics, past behavior and future predictions. So why are we still doing segment-based targeting? Why do we group individual people into lumpy, impersonal segments informed by function, vertical industry, HQ location, and the number of employees in their company? Segmentation glosses over all the beautiful texture and color we have on individual people we are dealing with.

So why don't we scratch segments all together and just build custom, dynamic marketing campaigns that are tailored to each individual that we track on our marketing radar? Instead of grouping individuals together, why don't we just build a profile of them specifically? Each and every one of them? This is obviously not a technology or data problem. We have plenty of vendors who do track, anonymize and profile people of interest for us. This is a scalability problem.

AI is all About Scale

There are just not enough people in marketing (or in the entire planet for that matter) to scale this idea of individualized marketing profiles and micro-campaigns to one person at a time. We can't possibly have enough marketers to manually study millions of digital activity signals and profile data regarding an individual's behavior, and then design micro-targeted campaigns that offer each of them the information they need, at the time they need it, and in the format that they want to consume it in.

Humans simply don't scale very well. On top of that, they don't like to perform repeatable mundane tasks, they get bored easily, they have career aspirations (most of them anyway), they demand some sleep and they even expect to take vacations. So you can't throw enough marketers at this opportunity. AI bots (MarkBots) can do this so much better!

AI Can Help us Scale Humans

AI can help us "micro-target" hundreds of thousands or even millions of individuals in a highly dynamic, tailored way without breaking a sweat. AI is the way to move beyond segment-based marketing and Account-Based Marketing (ABM), to People-Based Marketing (PBM.)

So what would that future look like if some super-smart MarTech vendors went after this opportunity to use MarkBots as a way to enable PBM?

A (Not So Distant) Future Example

Search behavior and content consumption trends are known to be a great early indicator of purchase intent. Many "predictive modeling" MarTech vendors already track individual search trends and consumption of solution-related content rolled up to a company level. They use that information to rate and rank prospective companies around these "demand signals" so you can prioritize your marketing and inside sales outreach to them.

For example, they'll say that company x is likely in the market for Cloud solutions because their search behavior and their consumption of cloud-related content has spiked to 3x this month compared to their average content consumption. That's interesting, but not sufficient. That doesn't give us high-enough confidence and rich-enough context so we can initiate our marketing outreach in a thoughtful way. Before we can outreach to a specific individual at that company, we'd need a bit more color. We need answers to questions like:

• Who are the specific individuals consuming the content in question (a name is not necessary - but their anonymized cookie-enabled profile can tell you a lot.) What do we know about their past behavior and how they interact with marketing materials and outreach?
• What's their function and level, and therefore their role in the decision making process? Do they just do the initial research or are they the one signing the check? Do they care about speeds and feeds or just the bottom line?
• What "buyer's journey" context can we gain by understanding the full text of the content they are consuming, as opposed to just the keywords contained in it? For example, they could be consuming top-of-funnel content like "Cloud for Dummies" or they could be taking deep dives at some very detailed feature/price comparisons and analyst reviews around deploying cloud infrastructures, which would suggest they are much further in their buyer's journey.

Having this depth of intelligence centered around the key individuals would have a significant impact on whether and how we engage them. AI-powered MarkBots can go for a swim into multiple, interconnected data lakes to piece together multiple data fragments into a holistic, and highly detailed profile for each person of interest. Once these thousands of individual profiles are constructed, one for each person, MarkBots can keep a constant eye on changes to individuals' digital body language, constantly learning and adjusting their profile so we are ready to trigger an action at any point we know they are most likely to be receptive to our outreach.

Such individual, dynamic profiles can then be used to activate a "marketing campaign of one", at the optimal time. A MarkBot can decide the precise moment (based on everything we know) to fire off an email, a tweet, a LinkedIn outreach, or an event invitation, using the right set of prepared messages to match their specific need, at that specific time.

And while your tireless MarkBots can execute hundreds of thousands of such individualized "micro-engagements" every day, they can also be on the lookout for the right reaction from the person you are reaching out to, and pass the baton to your sales team for that human touch at the time when it sense to be optimal.

So, in my view, a couple of things need to happen first before PBM can be reality:

1. It will first take a group of innovative and adventurous Martech vendors (challenge issued) to establish syndication agreements across their data lakes to allow MarkBots from participating companies to swim from lake to lake fishing for the right data fragments to piece together.
2. At the same time, a smart developer can build and "teach" the bots how to collect the fragments, build the individual segment knowledge, watch for triggers and then act. There are plenty of smart developers building special-purpose bots already for other fields like healthcare, manufacturing, and retail. Someone, please hire some of them and put them to work to build Markbots for us.
3. Finally, with GDPR on the way next year, we'll have to work with anonymized information and find clever ways to safeguard users' personal information until they are ready and willing to have a deeper engagement. We have to solve for this anyway. AI or not.

So, a challenge issued to all MarTech vendors. AI is the right technology, and marketing B2B marketing is the right arena. And if you decide to go after this, my team @DXCTechnology and I would love to be your partner.

By the way, "scaling humans", as well as the concept of using bots to work around existing systems, is a concept I borrowed from Sean Lane, co-founder and CEO of CrossChx; a company revolutionizing the use of AI in healthcare.