The most valuable customer data marketers can have is data that gets consumers to convert. That data is rarely broad demographics alone. More often, nuanced behavioral and attitudinal attributes underlie the best performing campaigns.

In fact, 56 percent of U.S. consumers say they’re more loyal to brands that “get me” as an individual, according to a study by marketing agency Wunderman. And, 79 percent of consumers polled in that same study say that before they’ll even consider making a purchase they expect brands to show they “understand and care about me.”

How can anyone but marketers with the biggest budgets and largest team of analysts succeed in that environment? By tapping lightning in a bottle; that is, by building precise custom audiences using powerful artificial intelligence (AI) tools.

Technology has helped level the marketing playing field in myriad ways. One of the most formidable is in using AI to enrich and optimize customer data. No longer do you need the marketing budget of a Fortune 500 behemoth to build the rich audiences, personas, and segments that normally require a team of expert analysts.

When it comes to getting the most from customer data, one challenge for many marketers is that their first-party data is limited. Another is that they may have a relatively small number of variables or criteria, or they lack the budget or analytics team needed to build comprehensive customer personas and segments.

Using sophisticated AI tools, those marketers can build the kind of precise custom audiences, quickly, that are otherwise available only to the largest organization. These tools build distinctive audiences using millions of variables or criteria, rather than the 10 or 20 most marketers have in-house. The resulting audiences and segments are much more dynamic and complex than would be available through a traditional approach — except when marketers have a huge budget and reams of data. Many marketers don’t have either. Even marketers from large companies such as CPGs are often disintermediated from consumers.

The best of both

Lacking resources, marketers typically have to make tradeoffs between speed and complexity when it comes to building audiences.

When marketers want speed, they tend to use prebuilt segments created from just a few variables, such as life stage cluster or financial cluster or criteria gathered from first-party purchase data. When marketers custom build segments, they may get richer ones, but there’s a significant time lag from idea to implementation while analysts buy data to create models, build models, run regressions, and the like.

Lacking resources, marketers typically have to make tradeoffs between speed and complexity when it comes to building audiences.

Using a more modern approach, one built on AI, computers do the work. There’s no need to trade speed for complexity, or vice versa. Marketers can have both. AI is fast and can select from more variables in a short time. AI tools for building custom audiences help marketers find the best prospective customers at scale, which is where the real power comes in.

Plus, using the right AI tools, marketers can self-serve. A marketing planner or media buyer can create segments themselves. Once marketers understand their audience, they can choose variables best suited for specific campaigns. They don’t have to engage an analyst or data scientist to do it. This puts power in the hands of even the somewhat data-phobic marketer.

As a result, non-analysts can make better data-based decisions. AI tools for building custom audiences allow marketers to do what they do best — such as build out strategy and create campaigns — while the AI helps define the audience needed to make that strategy or campaign actually work.

AI in action

The ideal AI engine for building custom audiences will access a large, dynamic consumer graph — representing millions of real people with millions of diverse data points about them.  The data should be diverse, allowing the computer to select from the most relevant and predictive behaviors, demographics, past purchases, interests, affinities, and the like to develop a truly unique and custom audience segment. It should allow marketers to understand and target these audiences themselves through a self-service tool, without engaging a team of data analysts or scientists.

These AI tools enable marketers to take a deeper dive to find characteristics that they can include, or exclude, to build truly unique audiences. Marketing planners and media buyers can work collaboratively to get the audiences that they jointly need to drive the best campaign performance. And, they can get their segments in a day.

Even non-analysts can make better data-based decisions. AI tools for building custom audiences allow marketers to do what they do best — such as build out strategy and create campaigns — while the AI helps define the audience needed to make that strategy or campaign actually work.

Companies large and small can benefit from AI informed targeting. One example comes from a leading transportation supply manufacturer, who uses AI to build and reach distinct audiences. It may be a global company, but it didn’t have the budget to go broad with marketing to increase its presence in the U.S. as it changes its brand image from a stodgy manufacturer to “an innovator that cares about the environment.” Instead, these marketers are targeting receptive, relevant audiences at scale. Using an AI tool designed to build custom audiences, its marketers dig into nuances to find consumers who appreciate innovation and green issues. It can be tough to get people excited about tires, but this company is doing just that. The resulting audiences have been performing well and the company is seeing a change in consumer behavior.

Sure, marketers can go to a social media platform and do something like this, but are often confined within a walled garden. Walled gardens have two inherent problems: One is the wall; if marketers want to run a consistent campaign across multiple platforms it can be tough to do if the data they need is only available inside the wall. In this scenario, marketers lack control of the data. The other is the condition of the garden. The data inside the walled garden may not be relevant or may not offer a wide diversity of data to leverage multiple dimensions in constructing complex audiences. Modern customer segmentation that’s independent of these walled gardens can go wide, fast — allowing marketers to execute audience-specific campaigns synchronously across multiple channels. This is especially important considering that 87 percent of consumers polled by customer feedback platform provider Kampyle say brands need to put more effort into delivering a consistent experience across channels.

Marketers who want to improve campaign performance and hit their engagement targets with fewer impressions, and who want costs to drop while click-through rates outperform benchmarks, should look to AI tools to build custom audiences. Marketers already using AI to build custom audiences are seeing improvements like 60 percent higher click through rates, while enjoying 30 percent lower customer acquisition costs.  Who wouldn’t want more engagement while serving fewer ads and reducing spend to gain more customers?

With consumers expecting relevance before they’re even customers, marketers need to focus more on strategy and use AI to make customer data actionable at scale. Traditional customer analytics are still important, but AI brings democratization of analytics — allowing marketers who lack big marketing budgets and a deep bench of analysts to get the same wins as much larger competitors. With AI tools for building custom audiences, can marketers release the lightning in a bottle that will charge up their campaign performance.