How To: Revenue Assurance Testing

Oct 17, 2019, 9:00:00 AM / by Petri Karppinen

If there’s one thing your customers are certainly paying attention to, it’s their bills at the end of the month. We’ve all read about customers who have accidentally been charged enormous sums above and beyond the correct amount, and we’ve all no doubt recoiled in horror at the PR snafus that inevitably result for the telco operator in question. Needless to say, any steps that a business can take to avoid this kind of error are worth their weight in image consultations and damage control.

Business man watering a growing green dollar sign tree concept

Though things like network migrations and IoT devices get a lot of the attention when people talk about test automation, the example above is a pretty clear example of the importance of testing OSS/BSS infrastructure. Though on the surface it might seem as though things like latency, jitter, packet loss, roaming coverage, etc. are the biggest factors determining quality of service (and thus customer retention and profitability), lower key activities like invoicing can have an outsized impact on how your customers perceive you. As such, these back-office processes are actually ideal areas for test automation.

 

The Challenge

When it comes to revenue assurance, most telco operators are well aware of how sensitive customers can be on the topic of invoicing. They have to be certain that they’re correctly documenting and managing all of the intricacies involved in each subscriber’s selected plan, actual usage, and preferred payment options. If you inadvertently send out a bill that’s too large, you risk exactly the kind of negative PR that we discussed above; if you send one that’s too small (or fail to send one at all), you lose out on revenue. You need to be absolutely sure that you’re sending the right bills to the right people at the right time, or the business will begin to hemorrhage money.

 

To add an additional layer of complexity, these back-office services have to be able to gather data from the devices associated with users’ accounts in order to arrive at the correct number. Even with “unlimited” plans there are still some services that need to be invoiced separately. This means that the billing systems have to integrate reliably with CDRs (call data records) to connect data about subscriber identity, used services (in bytes), and/or usage length in seconds to billing systems. As such, they have to verify that CDRs are being consulted and validated before invoices are sent to the consumer.  

 

The Solution

Testing functionality and compliance for this area is obviously mission critical, so anything that can be done to ensure that these tests are conducted reliably and in a standardized manner should be a top priority. This makes automation a particularly attractive choice for revenue assurance. Unlike many of the varieties of test that we refer to on this blog, automation here requires access to the CRM (customer relationship management) system being used by the business in question—in addition to a robust test framework, 2 PCS operator SIM cards, and two standard (i.e. iOS or Android) mobile phones. (It’s important that these phones by out-of-the-box devices—as rooted or jailbreaked phones have the potential to deviate just enough from end-user devices to skew test results.) Once these elements are in place, automated tests can be set up within a day or two and tailored to the specifications of the particular telco operator. Once all of the set-up has been completed, users can automatically perform calls and use data with the phones, verify that the voice and data usage is being recorded properly, and then ensure that the data is being transmitted to the right place in the CRM for future invoice generation. If there’s anything that’s not working as it should, you’ll find out before your customer invoices are being generated.

 

Here, it’s important to note that success in revenue assurance testing is in large part about going deeper than end-to-end tests. After all, you’re not verifying the completion of a simple transaction (a call placed on your network, e.g.) or even a set of transactions (a call placed, then redirected to voicemail, and that voicemail checked by the second party later)—rather, you’re digging into call logs in a way that not all end-to-end systems are automatically poised to do. If you’re not able to verify that your CDR information actually reflects customer network usage, then it doesn’t matter how effectively the rest of the process is working. Why? Because you’ll still wind up billing the wrong amount.   

 

The Benefits of Revenue Assurance Automation

Okay, it should be pretty clear from the way we’ve spoken about it above, but we’re not afraid to state the obvious. By reducing errors and fixing any bugs that could arise in your billing processes, you can pave the way for a number of benefits:

  • Reliably producing the correct invoice
  • Avoiding customer complaints, reclamation, and sometimes even law suites
  • Maintaining a positive reputation
  • Creating a financially healthy operation with a foundation for growth.
  • Speeding up implementation time for billing systems.

While it’s hypothetically possible to gain most of these benefits while testing these systems by hand, some—like speeding up implementation time—simply aren’t feasible without automation. Why? Because manual testing not only takes longer, it also less reliably produces the kind of thorough, standardized documentation that automated tests offer up. In the case of something like your billing system, it’s up to your tests to point engineers to the exact spot where an issue is occurring, and the more consistent your test results are in displaying that information, the more efficiently engineers can assess and resolve issues. By the same token, once these tests are set up it’s easy enough to run them again, which means that ongoing regression tests, rather than competing for valuable human test resources, can be run regularly and efficiently to ensure continuing functionality.

 

In this way, you do more with your service verification than simply avert catastrophe—you actually add value for future tests and product rollouts.

 

learn more about test automation

Topics: Time to Market, End-to-End Testing

Petri Karppinen

Written by Petri Karppinen

Petri has over 20 years of startup business and management experience in IT services, software and consulting. As Chief Financial Officer he leads the global operation to prepare the financial management process for growth and business control.