Unlocking Untapped Revenue in Your CRM

Deep learning and AI can help you uncover unidentified sources of revenue within your customer database. Here's how.

Matt Engstrom

Head of Marketing

Time to read8 min read
March 21, 2022

Finding Demand Through Acquisition Is Costly

New customer acquisition is indeed essential for growth. After all, there are no doubt many potential customers out there who want to buy and are able to buy the things you are selling. 

However, acquisition is challenging. Firstly, it’s difficult to get a potential buyer to change behavior. If they haven’t bought from you before or are currently buying from a competitor, you are essentially asking them to venture into the unknown and change behavior. This challenge is exactly why marketing costs are so high.

You have to find potential customers and use the right marketing tactics to get them to convert. Not to mention, the rising costs of data and increased targeting challenges due to increased privacy regulations means that acquisition is expensive.

So, where else can marketers locate untapped demand? 

The answer lies within your first-party data. Your CRM contains a potentially huge amount of demand that you are likely simply just not tapping into. And to locate this demand and unlock the revenue within, all you need is the data assets you already have access to. This could be customer data, transaction history, digital data, and more!

It’s also a misconception that you need your data to be structured in a sophisticated way either through your own internal systems or within an CDP. Although having structured data makes it more usable and actionable in a lot of ways, finding untapped demand doesn’t necessitate sophisticated data structures

There is revenue on the table, right now, within your CRM

People are complicated and different, and make purchase decisions based on many different factors depending on the product and the customer. But, the reality is that in order for someone to buy, they have to have a need or desire for the product and the capability to purchase. Finding customers who fit those two criteria is the tricky part.

It’s challenging because you can simply ask every customer what they would want to purchase from you. 1) that isn’t scalable and 2) some people might not even be aware that they are in the market for an offer you are selling. The trick is you need to let them know about it… in some cases showing them offers they aren’t even aware they want yet!

Identifying demand is challenging but a huge opportunity for your business

Think of untapped demand as the total potential revenue you could generate if you were able to deliver an offer to every customer within your database who is in market for that offer minus the demand you are currently capturing through your CRM marketing efforts. 

The reality is you are likely missing out on a ton of potential revenue.

There are products that many of your customers would be willing to buy they simply don’t know about or many of your offers have potential buyers but they are under marketed.

Also, demand shifts with time. The potential buying audiences for a specific offer today is not going to be the same as it will be tomorrow. Which means that demand is not static; it’s fluid. Luckily, you are already collecting tons of data that enables you to find demand even as it shifts over time.

Your data can tell you everything, but not all signals are created equal

Showing each individual customer every offer and asking them what they’d want to buy simply isn’t scalable; however, your customers are sending signals to you all the time that can tell you just what you need to know and they don’t even know it!

Many CRM marketers traditionally rely on explicit signals to identify demand. These include past purchases, website visitation, search behavior, or abandoned carts. This data is incredibly important but these signals might not always point you in the right direction. 

Explicit signals rely on small data points to make big assumptions. Finding customers who have purchased handbags in the past and delivering campaigns for your new line of handbags is an intuitive strategy, but you aren’t able to take into account customers who no longer need handbags nor are you able to find customers who haven’t purchased them in the past but have a new interest in handbags in the future.

Implicit signals are actually much stronger indicators of intent and more powerful in identifying demand within your CRM.

Humans can’t identify all the signals that may indicate buying intent nor establish how significant they are in predicting buying intent and that’s why implicit signals are so important.

An implicit signal might be that a past purchase of a product or category is actually an indicator of a purchase of a different type of product in the future. Email addresses, zip codes, or interactions with your website or marketing campaigns can all tell you more than you know. And some of these signals will be more significant in predicting intent for a specific offer than others. 

With the right tools that can find the signals that most significantly impact intent, you can bypass the pitfalls of relying only on explicit signals and find customers who want to and are able to buy specific offers. Even if there hasn’t been a clear indicator of intent in the past. 

Deep learning is the only way to find the right signals and identify demand

Finding untapped demand is incredibly complicated. It takes all customer attributes multiplied by the number of offers times customer events shifting over time. A sophisticated data science team could build prediction models for a few offers a couple times a week but what about doing that for every offer updated every day? Only with that information can marketers make real-time decisions about where to unlock demand and only Deep Learning powered AI can deliver that.

Deep Learning is machine learning that works similar to how the human brain learns. It can take a look at all your first party data and identify patterns that indicate exactly what buyers of a certain offer look like and then find those buyers within your customer database. 

The result is the ability to identify unidentified intent for any offer, category, brand or combination of these so you can market offers to the right people. And, in many cases this results in new campaign opportunities to deliver revenue that simply never existed before. 

The marketing applications of finding untapped demand are powerful

With this information, your targeting strategy becomes a game changer for your business. You can target based on demand or intent to purchase delivering offers to people who want to buy and are able to buy. In some cases these customers may not even know they are in the market for an offer themselves until they see it! That’s how accurate Deep Learning is when using implicit signals to identify intent.

Marketers can use Deep Learning to find new campaigns with offers that might be underpromoted or not promoted at all.

They can also understand the overlap of buying audiences between offers within multi-offer campaigns so they can be targeted to those who want to buy all of the included offers within the campaign. The impacted metrics are revenue per campaign and overall sales.

Additionally, you can avoid sending campaigns with offers people don’t want to or are unable to buy. Irrelevant messaging leads to campaign fatigue, unsubscribes, and reductions in customer lifetime value. 

Acquisition is definitely an essential part of a marketing strategy to maintain above normal ROI into the future. But, many marketers overlook the demand within their current customer database that’s just sitting on the table. Wouldn’t it be great to be able to tell your boss you increased revenue substantially without having to spend a dollar on advertising? 

You might also like