5 Testing Tactics to Help You Troubleshoot Your Network

A few years ago, an Accenture report uncovered a statistic that may be particularly troubling for mobile carriers: 77% of all consumers would retract their loyalty from a particular business more easily than they would have just three years prior.

Though this study wasn’t focused on telecommunications in particular, we’re willing to bet that this insight rings true for any number of telco businesses. Telco businesses are looking at increased subscriber churn, new competition from OTT (“over the top”) services, and decreasing per-packet margins in the emerging era of the IoT (internet of things).

There’s no magic bullet for improving customer retention in an increasingly complex telco environment, especially as the introduction of 5G will force operators to run even more networks in parallel. There is one thing that probably impacts customer satisfaction more than most other factors: network quality.

And how do you maintain a high level of network quality? By troubleshooting any issues that crop up in your network quickly and efficiently. Easier said than done, we know. But here are a few testing tactics you can implement to boost network quality and resolve issues faster.

1. Automate as Much as Possible
One of the biggest pitfalls in modern telecom testing is time. Manual tests for most use cases can only be conducted at a rate of 6-8 per engineer per day. This means that 100 test cases can take more than two working weeks to complete.

This simply isn’t tenable in an era where pressure is building steadily to speed up time-to-market. As we alluded to above, the introduction of 5G means that operators will have to maintain 2G/3G, 4G, and 5G networks in parallel and ensure interworking among them for the foreseeable future, which increases complexity and slows down test flows even further.

Testing and operating multiple networks is an issue in its own right, but it’s especially problematic with regard to troubleshooting. Why? Because the slower your test flows are, the less time and people-power you have for actually addressing the test results. For this reason, automated testing quality assurance and verification – are quickly becoming a necessity for communications service providers.

2. Consider Keyword-based Tests
Automated tests can help you increase test throughput by leaps and bounds, empowering you to find issues and begin your troubleshooting process faster. That said, not all tests are created equal when it comes to troubleshooting.

If your tests are difficult to understand, or they’re stuck behind a silo somewhere, it’s unlikely that they’ll help you find and resolve network issues more quickly. On the other hand, tests that are easy to read can accelerate the troubleshooting process by making it possible for less technical staff to get involved.

From our perspective, the best way to make this happen is to utilize keyword-based tests. These are tests that can be executed via a library of pre-defined, device independent keywords corresponding to distinct actions and network elements. These typically generate keyword-based reporting, which can help you further leverage less technical staff for agile, inclusive troubleshooting workflows.

3. Define Roles and Responsibilities
We noted above that keyword-based tests are a useful way to include less technical personnel in your testing flows, but what responsibilities should these non-technical folks actually have? If your new network update seems to be resulting in an increase in packet loss for VoLTE subscribers, shouldn’t an engineer be given the reins?

Absolutely! This is why it’s so important to set out clear guidelines for roles and responsibilities when it comes to testing and test aftercare. While some of your employees might be tasked with combing through protocol-level test data to uncover exactly where your new functionality is going wrong, others on the team will have to deal with setting up retests, changing regression test cases as needed, and making any adjustments to expected time-to-market projections.

By clearly establishing these roles up front, you decrease the odds that team members will be working redundantly, inadvertently against each other, or not at all when it comes to addressing network issues.

4. Integrate OSS/BSS Systems
Your subscribers obviously care a lot about network quality, and they’re constantly seeking the best bargain they can get in terms of quality and coverage relative to price. At the same time, there are other factors that come into play when customers are deciding whether to renew their contract.

For instance, billing errors can cause a lot of anxiety when they occur, especially in an industry that has, shall we say, a less than stellar historical reputation regarding customer service. After all, who wants to spend potentially hours on the phone trying to clear up a billing error?

Some telco operators fall into the pitfall of imagining that test automation is limited to actual network activity, it’s actually fairly straightforward to incorporate backend OSS/BSS systems into a robust test automation framework. If the framework can access and control these systems the same way it would be a mobile app, it’s possible to orchestrate tests involving, say, your CRM and end-user phones.

For instance, the automation framework can place calls and use data, ensure that the call and data usage is reflected accurately in your backend systems, and then verify that you’re generating invoices for the right amount. Thus, your ability to detect and troubleshoot backend issues increases, along with customer satisfaction.

5. Go Beyond End-to-End
We have already talked a little bit about how keyword-based tests can help you troubleshoot more efficiently, but that’s not the only way you can get additional value out of your tests. Specifically, you can go beyond end-to-end. Don’t test with your eye on just endpoint behaviour, but also protocol-level activity that’s happening while a call is being placed or a text is being sent.

While a failed test can tell you a lot, it often leaves some unanswered questions from a troubleshooting perspective. By capturing signalling traces and other granular information about what exactly is happening on your network, you can get even more insight into what might be going wrong when a test fails.

By adopting these testing tactics, you will enhance multiple aspects of your organization. You can speed up testing and troubleshooting, which will positively impact your time-to-market and network quality. Ultimately, your world-class network quality will improve customer satisfaction, enabling you to retain and attract new customers.