Is your OSS a thermometer or thermostat? 10 Critical Questions to find out

Most OSS platforms monitor, but only the best ones take action. Is yours a thermometer or a thermostat?  Does your OSS only report the temperature, or can it actually help change your environment? Do you think your OSS is doing enough? It could be missing the crucial capabilities that power future-ready networks.

Below we provide you with 10 questions to help you uncover the truth about whether your OSS fixes problems or just reports them.

Thermometer vs. Thermostat: How’s Your OSS Really Helping?

Operational Support Systems (OSS) play a crucial role in managing network operations, but not all OSS solutions are created equal. Some merely collect data and generate alerts (like thermometers), while others take action and drive change (like thermostats).

A thermometer-like OSS reports issues, requiring manual intervention to resolve them. It’s a dashboard full of graphs and metrics, but nothing to highlight whether any of the facts and figures are expected / unexpected, nor suggestive of what needs to change if outside expected tolerance levels.

In contrast, a thermostat-like OSS can not only detect issues but also trigger automated (or semi-automated) responses, optimise network performance and support business decision-making at both operational and executive levels.

With the increasing complexity of modern networks, having a thermostat-like OSS is no longer optional – it’s essential. They can no longer just service the O in OSS. Their relevance must extend beyond your Operations Team in terms of value provided.

Think for a moment about the advice you receive in your daily life (not just in your OSS world). Which do you find more valuable?

• Someone who tells you about a problem you have; OR
• Someone who provides a step-by-step solution to a problem you have

The following questions will help you assess whether your OSS is truly proactive or just an observer / reporter.

10 Critical Questions to Assess Your Current OSS Capabilities

1. Does your OSS only provide insight to operational staff, or does it inform executive decision-makers too?
   A proactive OSS should deliver insights beyond network engineers. It should provide business-relevant analytics to executives. Importantly, it’s not just about having a dashboard of metrics, but having a clear set of response levers that enable leaders to informed strategic responses
2. Does your OSS only alert you to problems, or can it automatically (or semi-automatically) resolve them?
   An effective OSS should go beyond notifications and facilitate automated or semi-automated issue resolution, reducing the need for human intervention (though noting that human-in-the-loop involvement is also highly recommended in many situations)
3. How much manual intervention is required to act on insights presented by your OSS?
   If your OSS constantly requires engineers to interpret data before being able to take action (think dozens and dozens of graphs on a dashboard, but where the engineer needs to identify the anomalies manually), it’s functioning more like a thermometer. By comparison, a thermostat-like OSS minimises manual workload by automating responses, or at least reducing human interpretation time significantly
4. Can your OSS predict and prevent issues before they impact operations?
   Modern OSS solutions leverage predictive analytics and AI to foresee potential failures and take preventive measures before they cause disruptions. It’s one of the reasons why many network operators are moving from traditional alarm lists and incident management tools to AIOps solutions (although at the time of writing, AIOps solutions are still nascent and have a long way to go before providing reliable closed-loop predictive responses)
5. Is your OSS capable of making decisions in near-real-time, or does it just provide historical data (and how historical)?
   While historical data is valuable, an advanced OSS should also support near-real-time decision-making, adapting dynamically to live network conditions. And when we say historical, in terms of MTTx metrics (Mean Time to Detect / Respond / Know / Repair / Resolve / etc), anything longer than 5-10 minutes is often too long
6. How well has your OSS been integrated and optimised with automation and AI-driven workflows?
   OSS effectiveness isn’t measured in individual transactions, it’s measured in the ability to complete end-to-end workflows (design, assurance, fulfilment, billing, etc). An OSS that can seamlessly integrate with AI and automation whilst continuously learning, optimising and evolving provides most operators with a significant competitive advantage
7. Can your OSS adapt to network changes dynamically with closed-loop mechanisms, or does it require constant manual or rules-based reconfiguration?
   Thermostat-like OSS solutions use closed-loop automation to self-adjust based on real-time data, whereas static OSS requires ongoing manual updates. If the closed-loop mechanisms are reliable, they can significantly reduce response times and human effort / involvement. Speed is of the essence in most workflows in a modern telco as it can also be a significant competitive advantage
8. Does your OSS provide actionable insights, or does it just overwhelm you with data?
   More data isn’t always better. A smart OSS filters noise and presents only the most relevant, actionable insights to ensure quick and informed decision-making. It then also provides the levers for decision-makers to pull to ensure actions/responses are performed
9. Is your OSS enabling business growth, or is it slowing down operational efficiency?
   A well-implemented OSS should streamline operations, reduce costs and enable network scalability, ultimately driving business growth. A less well-implemented OSS can do the exact opposite. It can actually be a significant constrainer of growth
10. How quickly can your OSS respond to incidents, and does it help reduce downtime?
    Reducing MTTx is a key indicator of an effective OSS. The faster your OSS can detect, diagnose and respond to issues, the less downtime your business will experience. The less downtime or more reliable your network, the better the customer experience (not to mention all the other beneficial metrics downstream of that!)

Thermostat-like Proactive OSS are the Future: Key Industry Trends

As networks become more complex, passive OSS solutions are no longer sufficient. The shift toward intelligent, proactive OSS is being driven by several key trends:

• Thermostat-like OSS start with intuitive UIs / dashboards that have clear metrics and equally clear levers to pull in response. They can no longer be “Designed by Engineers for Engineers” but designed with much of the decision-making already built in for a new generation of network operations at speed and scale
• The role of AI, automation, and closed-loop systems in modern OSS – Modern techniques that include AI-driven analytics and automation are increasingly enabling networks to self-heal and self-optimise, reducing reliance on human intervention
• Organisations are shifting from passive to intelligent OSS – Companies that invest in proactive OSS gain a competitive advantage by improving network resilience and service quality metrics

Seven Steps to Make your Current OSS more Themostat-like

The journey from a thermometer-like OSS to a thermostat-like OSS requires careful planning. Here are 7 key steps to consider:

1. Evaluate your current-state OSS capabilities – Use the 10 questions above to identify gaps in your current-state solutions
2. Build more informed decision-making from current-state solutions – Identify whether there are any ways that you can make your current-state solutions more informed / informative. There are numerous ways to do this, including:
   1. Build dashboards that show threshold-crossing alerts (TCA) such as if a core router has an expected latency / RTT of 40-50ms and an alert is raised if RTT exceeds 100ms
   2. Don’t just show dozens of graphs in a dashboard, but prioritise attention on any graphs that have exceeded and/or are nearing acceptable thresholds
   3. Define procedures of response if certain metrics or situations appear, such as a customer activation goes into jeopardy status and needs attention to ensure an SLA breach doesn’t occur
   4. Deeply understand your processes, especially process variants, then optimise them before attempting to automate
3. Invest in process optimisation and automation – Automations and AI-driven decision-making (potientially even including closed-loop automation) are essential for modern OSS.
4. Focus on actionable insights – Understand your metrics that matter. Ensure your OSS provides meaningful, decision-ready intelligence rather than overwhelming data. If there’s nothing you can (or should) do in response to a certain metric, then it’s arguably just a vanity metric – there to look good, but with no real value added
5. Prioritise executive-level visibility – Understand who makes the decisions and under what conditions. A thermostat-like OSS should deliver insights at both operational and business levels – to the people (or systems) that can then take action
6. Use Re-Framing Techniques when conducting OSS selection and deployment – Don’t just look for an OSS that meets today’s needs; ensure it can scale and evolve with your business for the projected lifetime of your OSS
7. Prioritise what launches the rocket – A lot of effort and expense tends to be invested into OSS enhancements at the “easy edge” of functionality. It’s often a lot harder to re-invent and optimise the really important functionality. Thermostat-like functionality should be prioritised in your OSS development backlog

The days of passive, monitoring-only OSS are fading. As networks grow more complex, businesses need an OSS that doesn’t just observe but actively manages, predicts, adapts to and optimises network performance. If your OSS is merely a thermometer (designed by Engineers for Engineers), it’s time to upgrade to a thermostat-like system – one that not only reads the temperature but also adjusts to changing demands.