Introduction
Telecoms, like much of the business world, often revolves around measurements, metrics and KPIs. Whether these relate to coverage of networks, net-adds and churn rates of subscribers, or financial metrics such as ARPU, there is a plethora of numerical measures to track.
They are used to determine shifts in performance over time, or benchmark between different companies and countries. Regulators and investors scrutinise the historical data and may set quantitative targets as part of policy or investment criteria.
This report explores the nature of such metrics, how they are (mis)used and how the telecoms sector – and especially its government and regulatory agencies – can refocus on good (i.e., useful, accurate and meaningful) data rather than over-simplistic or just easy-to-collect statistics.
The discussion primarily focuses on those metrics that relate to overall industry trends or sector performance, rather than individual companies’ sales and infrastructure – although many datasets are built by collating multiple companies’ individual data submissions. It considers mechanisms to balance the common “data asymmetry” between internal telco management KPIs and metrics available to outsiders such as policymakers.
A poor metric often has huge inertia and high switching costs. The phenomenon of historical accidents leading to entrenched, long-lasting effects is known as “path dependence”. Telecoms reflects a similar situation – as do many other sub-sectors of the economy. There are many old-fashioned metrics that are no longer really not fit for purpose and even some new ones that are badly-conceived. They often lead to poor regulatory decisions, poor optimisation and investment approaches by service providers, flawed incentives and large tranches of self-congratulatory overhype.
An important question is why some less-than-perfect metrics such as ARPU still have utility – and how and where to continue using them, with awareness of their limitations – or modify them slightly to reflect market reality. Sometimes maintaining continuity and comparability of statistics over time is important. Conversely, other old metrics such as “minutes” of voice telephony actually do more harm than good and should be retired or replaced.
Enter your details below to download an extract of the report
Looking beyond operator KPIs
Throughout the report, we make a semantic distinction between industry-wide metrics and telco KPIs. KPIs are typically generated for specific individual companies, rather than aggregated across a sector. And while both KPIs and metrics can be retrospective or set as goals, metrics can also be forecast, especially where they link operational data to other underlying variables, such as population, geographic areas or demand (rather than supply).
STL Partners has previous published work on telcos’ external KPIs, including discussion of the focus on “defensive” statistics on core connectivity, “progressive” numbers on new revenue-generating opportunities, and socially-oriented datasets on environmental social and governance (ESG) and staffing. See the figure below.
Types of internal KPIs found in major telcos
Source: STL Partners
Policymakers need metrics
The telecoms policy realm spans everything from national broadband plans to spectrum allocations, decisions about mergers and competition, net neutrality, cybersecurity, citizen inclusion and climate/energy goals. All of them use metrics either during policy development and debate, or as goalposts for quantifying electoral pledges or making regional/international comparisons.
And it is here that an informational battleground lies.
There are usually multiple stakeholder groups in these situations, whether it is incumbents vs. new entrants, tech #1 vs. tech #2, consumers vs. companies, merger proponents vs. critics, or just between different political or ideological tribes and the numerous industry organisations and lobbying institutions that surround them. Everyone involved wants data points that make themselves look good and which allow them to argue for more favourable treatment or more funding.
The underlying driver here is policy rather than performance.
Data asymmetry
A major problem that emerges here is data asymmetry. There is a huge gulf between the operational internal KPIs used by telcos, and those that are typically publicised in corporate reports and presentations or made available in filings to regulators. Automation and analytics technologies generate ever more granular data from networks’ performance and customers’ usage of, and payment for, their services – but these do not get disseminated widely.
Thus, policymakers and regulators often lack the detailed and disaggregated primary information and data resources available to large companies’ internal reporting functions. They typically need to mandate specific (comparable) data releases via operators’ license terms or rely on third-party inputs from sources such as trade associations, vendor analysis, end-user surveys or consultants.
Table of content
- Executive Summary
- Key recommendations
- Next steps
- Introduction
- Key metrics overview
- KPIs vs. metrics: What’s in a name?
- Who uses telco metrics and why?
- Data used in policy-making and regulation
- Metrics and KPIs enshrined in standards
- Why some stakeholders love “old” metrics
- Granularity
- Coverage, deployment and adoption
- Mobile network coverage
- Fixed network deployment/coverage
- Usage, speed and traffic metrics
- Voice minutes and messages
- Data traffic volumes
- Network latency
- Financial metrics
- Revenue and ARPU
- Capex
- Future trends and innovation in metrics
- The impact of changing telecom industry structure
- Why applications matter: FWA, AR/VR, P5G, V2X, etc
- New sources of data and measurements
- Conclusion and recommendations
- Recommendations for regulators and policymakers
- Recommendations for fixed and cable operators
- Recommendations for mobile operators
- Recommendations for telecoms vendors
- Recommendations for content, cloud and application providers
- Recommendations for investors and consultants
- Appendix
- Key historical metrics: Overview
- How telecoms data is generated
- Index
