More customers, more problems? That doesn't sound right

Traditionally, customer acquisition strategies were guided by industry best practices and experiential knowledge held by an organization’s executives and senior-level individuals. Sometimes these strategies were successful, sometimes they were not.

Success was primarily determined by the return on investment (ROI) for various acquisition initiatives. While ROI is still one of the best ways to gauge the effectiveness of your efforts, the “trial and error” approach can have severe impacts on revenue growth, especially for small and midsize companies.

Machine learning, a subfield of AI, enables companies to take much of the guesswork out of the customer acquisition process, and it all starts with good data.

Extracting value from your data

B2C companies have access to an abundance of consumer data — data that enable companies to draw deeper insights about what their best customers look like, which leads are worth targeting, which types of promotions are likely to keep them engaged throughout the buying process, and just about any other outcome companies want to optimize.

As companies consider more data, they require more efficient means of analyzing it. That’s where AI becomes invaluable, as it allows companies to draw deep, data-driven insights and make accurate predictions about target outcomes. However, the key is implementation, leveraging these insights and predictions efficiently throughout the acquisition process.

3 applications of AI for B2C customer acquisition

1) Propensity modeling

The best way to optimize any customer acquisition initiative is to focus your resources on individuals with the highest propensity to convert on a target outcome. Why waste your precious marketing dollars on leads that probably won’t become customers, or worse, spend more on acquiring a new customer than they spend on your products or services?

Through various machine learning techniques, data scientists can build models capable of predicting — with a high degree of accuracy — a lead’s propensity to convert on a target outcome. These propensity models are especially helpful in optimizing lead generation and customer acquisition campaigns in two ways:

• Predictive targeting: Propensity models “score” individuals on the likelihood that they will convert on an outcome. Outcomes can be as broad as “make a purchase” or as specific as “bought product X and product Y, but not product Z.” Propensity scores become actionable when they’re fed back into a CRM, ESP, e-commerce platform, or any other marketing system. As you might imagine, targeting likely-to-convert audiences helps drive conversion rates and ultimately improve ROI on acquisition initiatives.
• Personalization: You achieve the greatest performance improvements when targeting propensity-based audiences with highly-personalized campaigns. Identifying the most influential variables in the modeling process helps you draw qualitative insights, which you can leverage to personalize and optimize qualitative aspects of your acquisition strategy: product photography, ad creative, web copy, etc.

2) Real-time lead scoring

The sooner you identify an individual’s propensity to convert, the better you can allocate your resources. Real-time scoring integrations feed new lead data into propensity models, where they’re scored and returned to various marketing systems. Real-time lead scoring can help you optimize your customer acquisition strategy in three ways:

• Immediately determine whether the lead is worth further marketing or sales engagement. If not, save your resources for high-scoring leads.
• Immediately enroll high-scoring leads into the right nurture campaigns. If a likely-to-purchase lead was captured on a particular product page, you might want to enroll them into a promotional campaign for that product.
• Immediately inform your sales team when high-scoring leads enter the funnel. This is especially useful when using multiple propensity models for multiple outcomes. If a new lead has a higher propensity to purchase product X than product Y, your sales team should know as soon as possible.

3) Location intelligence

Location intelligence situates business data in a geographic context. Identifying "hotspots" of high-scoring individuals can significantly reduce customer acquisition costs in two ways:

• Retail site and showroom expansion: Physical, or "brick-and-mortar," locations carry tremendous costs. It's no secret that location is paramount to a site's success. With scored geographies and careful analysis, you can identify the best (and worst) locations for new sites.
• Service area geotargeting: Whether the goal is to drive foot traffic to physical sites or to maximize ROI on direct mail or canvassing campaigns, targeting spatially-clustered areas of high-scoring individuals will increase conversion rates and lower acquisition costs.

Where to start

Regardless of which application of AI is most relevant to your business, start by acquiring rich, high-quality data, and lots of it. Good training data fuels the models that give you the insights and predictions needed to optimize your acquisition strategy.

When you have the right data, and enough of it, you’re ready to train your AI. You’ll need experienced data scientists and a solid machine learning engine to build models that actually work, but once they’re built and predicting accurately, you’re ready to start optimizing.

For a deeper dive into data discovery and machine learning, check out AI for the customer journey. The whitepaper lays out the four-step process that hundreds of B2C companies use to operationalize revenue-building AI, from customer acquisition to retention.

If you've got your finger on the pulse of your business, you know you need to adopt AI sooner than later.

While commercial AI adoption is on the rise, there's still time to catch up and get ahead of the curve with "off-the-shelf", or turnkey AI solutions.

Where do you stand in the AI landscape?

Consider the following questions:

• Are you brand new to AI? If so, where will AI have an immediate impact on your top and bottom lines?
• Are you experimenting with an enterprise-wide AI solution? If so, are you seeing the results you were expecting?
• Are you already developing AI in-house? If so, what can you do to optimize your existing resources?

Regardless of your position in the AI landscape, you can quickly get started with AI or optimize your existing resources by implementing proven, turnkey AI solutions that address the low-hanging fruit.

What are the low-hanging fruit?

It's all about automating tasks that have an immediate impact on revenues and costs.

Rather than using AI to automate existing jobs, use it to improve efficiency and productivity within those jobs.

After working with hundreds of B2C companies in a wide range of industries, we found that the customer lifecycle is an ideal framework for discovering highly predictive data, applying machine learning, and developing predictive models for a variety of use cases.

Let's take a quick look at a couple popular use cases to get you thinking about how they might apply to your business.

Increasing conversion rates and reducing acquisition costs

The most effective way to optimize your spend throughout the customer acquisition process is by targeting individuals with the highest propensity to convert.

So how do we identify individuals who are likely to convert?

• The first step is to enrich your lead and customer data with additional demographic, psychographic, and property-based attributes, giving you a vibrant picture of who your leads and customers are, not just what they purchased or where they came from.
• The next step is to use that enriched data to train the AI that will ultimately make predictions on future data inputs.
• Once trained, the AI can make data-driven predictions on who your most likely buyers are, where to reach them, and when to reach out.

We've seen projected sales cost reductions of over $300,000 when applying this approach.

Proactively preventing customer churn

Certain companies have a notorious problem with customer churn, especially those with subscription-based business models. Using a similar approach as the one mentioned above, companies can use AI to identify churn-prone customers before they even begin showing signs of churn.

This can help minimize marketing dollars spent on these churn-prone individuals, and focus more on customers with a higher lifetime value (LTV).

A deeper dive into use cases for turnkey AI

Hopefully, you're thinking about how you could leverage a turnkey AI solution to optimize outreach initiatives across your customer lifecycle.

For a deeper dive into the opportunities available with turnkey AI, check out our free whitepaper, AI or die: Understanding the inevitability of AI in business.

For more information on to use AI to optimize your customer lifecycle outreach, download our other whitepaper, AI for the customer lifecycle: Making the most of your data.

You've probably noticed the growing buzz around AI and ML. You've likely done some research into the field, read some AI success stories, and believe that artificial intelligence is more than just a buzzword.

Sound like you? Good, you're totally on the right path!

So is all the hype true? Is AI really as promising as it seems? Have most companies already adopted AI? Or even worse, are you too far behind to be competitive with AI?

Luckily, you're not too late.

Hype versus reality

In a survey of over 3,000 senior executives, McKinsey found that only 10% of respondents said they have actually operationalized AI at scale.

That said, another 10% claimed to have operationalized AI in at least one core business function, and about 40% are experimenting with some form of AI.

Couple those numbers with the massive increase in AI-related investments over the last few years, and we can see the time to get serious about AI is undoubtedly now.

Where's the interest, and how much is there?

Over the last five years, investments in AI have more than tripled. In 2016 alone, total investments (internal and external) ranged from $26 billion to $39 billion, with over 75% of that coming from tech giants like Amazon, Apple, Google and Baidu.

These cash-rich giants have mostly invested internally — on R&D and deployment — but they've also invested significantly in hard-to-find talent, primarily through major acquisitions.

Source: MGI Artificial Intelligence Discussion paper.

How can I possibly catch up and get ahead?

These massive investments can feel deflating, and rightly so if your plan is to go head-to-head with the giants by building up your own AI tools in-house. The good news is that you don't have to, and often shouldn't develop AI from scratch.

A wealth of AI startups have caught some serious traction by developing fit-for-purpose solutions that address focused sets of business challenges.

Whether those challenges rest in IT and security, finance and accounting, or revenue growth and marketing, organizations can easily adopt and operationalize these "off-the-shelf" solutions in a fraction of the time, and at a fraction of the cost of alternative "one-size-fits-all" solutions like IBM's Watson.

The key is to automate tasks that have an immediate impact on your top and bottom lines. AI technologies can't replace your entire marketing team, but they can arm that team with powerful customer insights and predictions, simply unattainable without powerful machine learning algorithms.

Some use cases

For more on this, and to read some specific use cases for off-the-shelf AI solutions, download our free whitepaper, AI or die: Understanding the inevitability of AI in business.

This post is part 4 of our 4-part Customer Lifecycle Optimization series.

First, identify your most predictive customer lifecycle data

B2C revenues are often tied directly to customer lifecycle outcomes. Targeting likely-to-convert leads, identifying good up-sell candidates amongst your existing customers, and proactively engaging churn-prone customers are all good ways to boost revenues while reducing overall customer acquisition costs.

The key to identifying those likely-to-convert leads, up-sell candidates, and churn-prone customers is mapping out your customer lifecycle and identifying predictive lifecycle data to better predict future behavior. With data-driven predictions, you can optimize your outreach initiatives by targeting better audiences at each stage of the customer lifecycle.

Motivating transitions between customer lifecycle stages

Whether you want to increase conversion rates from lead generation campaigns or reduce churn from existing customers, properly targeted outreach is essential to engaging and motivating the right leads and customers to take a desired action.

Outbound communication is the strongest and most versatile intervention at an organization's disposal to compel progress and therefore expand revenue. This includes individually targeted digital advertising, a form of direct outreach.

Consider the following examples of outreach initiatives:


Having identified a stage transition you'd like to motivate with an outreach intervention, the question becomes, "Who do I reach?" Regardless of the desired transition, the general technique is called audience expansion, also known as "lookalike" audiences.

3 examples of audience expansion

The following table contains three examples of audience expansion. In each case, an existing audience is "expanded" by identifying individuals with similar attributes in a larger set of candidates.


Note that in the third example, retention, we're looking for candidates (current customers) likely to join the indicated audience (lost customers) thereby entering the destination stage (reactivation) so that we can act to prevent that transition.

Audience vs candidates

To apply the audience expansion technique, we must always identify the audience and a set of candidates with similar characteristics and attributes.

An audience is a sample of the customers you're trying to find more of. This can be as simple as matching a lifecycle stage (i.e. retention and expansion), or as specific as trying to find more high-value customers.

Candidates consist of the universe of possible recipients of your outreach within which to expand your audience. Using lead generation as an example, your candidates would be your current leads, while the corresponding audience would be your current customers. Your audience (customers) is expanding into these candidates (leads) with motivation from your outreach initiatives.

Applying your predictive groundwork

With these groups defined, the next step is to apply your predictive groundwork. This could involve using patterns you identified in your data to look for similar opportunities among your candidates, or in more advanced cases, using artificial intelligence to build a predictive model trained to discriminate between likely and unlikely transitioners.

Finally, you will be left with a well-defined group of candidates likely to transition into a desired lifecycle stage when reached with relevant content.

Putting it all together

As mentioned earlier, this article is the fourth and final part of our customer lifecycle optimization series. To summarize the CLO practice, it's useful to recall our original motivation.

We employ AI-powered predictions to build and defend revenue by improving outcomes within our customer lifecycle. We know this means motivating transitions from one stage to the next.

1) Map your customer lifecycle to understand its stages and transitions.

2) Inventory your customer data, including indications of transitions.

3) Identify patterns in the data leading to transitions.

4) Leverage these patterns to guide outreach.

For more information on anything discussed in our customer lifecycle optimization blog series, download our free whitepaper: AI for the customer lifecycle: Making the most of your data.

This post is part 3 of our 4-part Customer Lifecycle Optimization series

Expanding on your customer lifecycle map

In our last article, Lifecycle mapping: uncovering rich, predictive data sources, we discussed the importance of mapping out your customer lifecycle to better understand where your most predictive customer data is hiding.

Lifecycle mapping is the first step to using artificial intelligence (AI) to optimize your customer lifecycle marketing initiatives.

Now, we'll pose some questions to help identify your predictive customer attributes and lifecycle events, pinpoint where that data is located, and recognize patterns to predict outcomes for future prospects, leads, and customers.

Step 2: Data discovery

Data discovery is the second stage in the customer lifecycle optimization (CLO) process. The primary task of this stage is to expand on your lifecycle map to identify authoritative data sources that establish progress.

As a reminder, your customer lifecycle map should look something like this:

<a href="

At each stage, prospects, leads, and customers will complete certain events that will individually or collectively trigger a transition in or out of that stage. Individuals will also have different attributes, which help determine whether or not they belong in any given stage.

Events typically emerge from data that represents behaviors and change. For example, your e-commerce system records purchases, your CRM records wins, and your ESP records email engagement. When individuals click a link in an email, sign up for a demo, or make a purchase, they're completing transition-triggering events.

Attributes exist in the data that describes the customer themselves. For example, your CRM will contain contact information, lead status, and possibly physical addresses. If you service a specific geographical market, a determinant attribute will likely be the individual's household address. If they live outside of your market, they will fail the litmus test.

Go through each cell of your customer lifecycle map and think about which events or attributes in your data could be used to trigger a transition or pass the litmus test. Think about where you can find that data.

Step 3: Predictive groundwork

Now that you've formalized your lifecycle and its representation in data, you can start recognizing patterns and eventually predict outcomes.

Imagine loading up your converted leads (customers in "retention and expansion") alongside your stale leads that never converted. What differences can you find?

This comparison analysis is especially effective when you've added depth and breadth to your existing data (see our previous article for more information on data depth and breadth).

As you experiment with this type of pattern recognition, you'll quickly realize that it's the kind of thing that computers do very well. That's where machine learning comes in handy.

Machine learning refers to a class of artificial intelligence techniques in which previous outcomes are algorithmically analyzed to uncover patterns, a process known as "training" a predictive model. The model can then be used to predict likely outcomes given new data by looking for indicative patterns.

What's next?

Once you've mapped out your customer lifecycle, identified where to find your most predictive data, and developed models to facilitate future predictions, you're ready to use those predictions to guide your outreach.

For a deeper dive into the four step customer lifecycle optimization process, download our free whitepaper, AI for the customer lifecycle: Making the most of your data.