⬅️ Back to posts
Filter By Categories

How to Increase Recurring Revenue from Existing Customers: Part 2 of 4

Parry Bedi
September 27, 2022

In 1926, Nikola Tesla predicted people would someday walk around with phones in their pockets. 


Impressive, right? 

On the other hand, in 1935 Tesla predicted that by the 21st “coffee will be no longer in vogue” because it would be viewed as a “harmful stimulant”.  Though he thought alcohol would stand the test of time because it was the elixir of life…… garfield
How is it that some predictions can be so spot on while others can be so wildly off? That’s the question I’m going to explore in this second part of my four-part series about increasing recurring revenue. 

Building a data-driven system is critical to success

In Part 1 of this series, I discussed the importance of building a data-driven system for customer engagement.  I mentioned that a true data-driven system has three distinct but reinforcing components: 

  1. A 360° Customer View 
  2. Predictions 
  3. Automation & Orchestration 

This post will focus on Predictions and how we move from solid, reasonable guesses to data-driven predictions with astounding accuracy. 

As a quick refresher, a true 360° Customer View should include:  

  • A customer profile  
  • Customer relationship details 
  • Customer Signals from Sales, Marketing, Product, and CS interactions 
  • User-level data  

This can quickly add up to be a lot of data.  And in a data-driven world, lots of data can be a good thing
lots of data

But how do we make sense of all this data? 

Of course, there are a number of ways to use data to make decisions. But I've generally observed that most companies choose from two commonly used approaches... 

  1. Rules-based systems
  2. Data-driven systems

Rules-based systems have serious limitations 

Rules-based systems are implemented by creating simple rules based on your knowledge of the customer.  

For instance, you would start by defining things important to you. Things like the number of employees a customer has. Or the number of monthly active users (MAUs). Or, perhaps, the number of “core features” being utilized.  

major key
If these attributes were important to you, then you might have a rule that says a customer with 1,000+ employees, 50 MAUs, and an average weekly use of 5 core features is a great customer! Based on this, you’d have your enterprise AE contact them.  
A rules-based system is straightforward and might be appropriate if you are just getting started on your journey to be a data-driven organization.  That said, a rules-based system has three major systemic limitations

Limitation 1: Gut instincts are not enough

A rules-based system is based on a flawed premise. It assumes the person establishing the rules has complete and total knowledge of all signals that drive the business. 
By asking probing questions about these rules, the limitations of this approach become readily apparent. Like, in the example above, why are 5 core features chosen and not 3?  trust
The truth is that systems built on gut instincts are inadequate in a fast-paced, dynamic business.  
Time and again, I’ve seen these systems devolve into background noise that employees simply ignore. It’s a bit like all the beeps you tune out in an ER room!  

Limitation 2: Lagging indicators are... lagging

Rules-based systems often rely on lagging indicators of customer intent and actions. This is especially true when it comes to rules related to customer churn and account expansion.  
For example, it’s often too late to prevent churn by the time a customer has stopped using a product “X number of times per day or week” (or whatever your rule is).   

Limitation 3: Obsolescence 

Even the best designed rules-based systems can become obsolete rather quickly.  
The problem is business is incredibly dynamic. Either you change, your competitors change, the market changes, or all of the above.  

In business, change is a constant. Therefore rules-based systems just can’t keep up.  
For example, what's the impact on already established rules if you decide to launch a new feature, or an existing feature is deprecated? 
When these natural inflection points happen, you’re forced to constantly update the rules (again, based on your gut). Or you end up ignoring the “out of touch” system. 

The natural end-state of a rules-based system 

Given that it’s incredibly hard to keep a rules-based system up-to-date, most companies tend to fall back on high-level metrics. Unfortunately, this further exacerbates the problem, especially since high-level metrics are not very actionable.  
Consider a rule that says that if the customer’s NPS score is below 70 and their daily active users (DAU) are less than 100, then the customer is a churn risk.  

This rule, while understandable, leads companies down a slippery slope. How can you truly determine if this customer is a churn risk without talking to them? So it begs the question if this approach is scalable.  

Yet, we see rules like this all the time from our SaaS customers.  
Even if it’s scalable for you to frequently speak with your customers, there is still the question of whether your customers will make time for you.


Not to mention whether they will provide the type of feedback or insights you need to make substantive changes.  

To add insult to injury, there is no straightforward way for you to increase the DAU. 
If there was, then your product team hasn’t been doing its job all along! 


Sure, build a better product, invest in onboarding, etc. But again, these are general guidelines. They might not be applicable to this customer or your business at a particular point in time.  

Now you’re stuck in this downward spiral of obsolete rules producing lagging indicators that force you to rely instead on gut instincts… 

Data-driven systems avoid the limitations of rules-based systems 

👆🏾This is why a data-driven approach is so much better. And it is unequivocally more effective.  

👆🏾👆🏾 This is why Tingono is building a data-driven system to increase recurring revenue from existing customers. 
Our platform identifies leading indicators of both churn risks and expansion opportunities.  

We do this by using advanced auto-ML techniques that work extraordinarily well with enterprise data. It even works well on sparse datasets.   
After we identify the leading indicators, we then turn these predictions into action that immediately impacts your business.  

So, how does this work? 
First, the system connects to your existing enterprise systems (e.g., CRM, product telemetry, support systems, etc.).

Then it gathers all your relevant customer journey data and applies ML to decipher what signals predict who your best customers are.

Finally, it intelligently guides you down the path of turning all your customers into your best customers.   

A predictive system like this does not rely on gut instincts or lagging indicators. It also naturally updates with each new data point, so it doesn’t quickly become obsolete.  In short, it helps you avoid the inherent limitations of a rules-based system.   
A data-driven system will enable you to progress beyond the uncertainty of sometimes right, sometimes wrong, erratic predictions. It will empower you to create powerfully accurate predictions you can rely on. 

Moving on… 

The next question becomes, “how can I act on these predictions?”. This is the heart of Part 3 of this series. Please tune in to read about the Automation & Orchestration element of a data-driven system.  
In the meantime, if you have any questions or want to learn more about how Tingono’s data-driven system is helping companies increase revenue from their existing customers, hit me up! 


Check out the other 3 parts of the series: