When potential customers ask ChatGPT, Gemini or Google AI Overviews about mobile providers, tariffs or network quality, the answer is shaped by a mix of sources rather than by one provider’s own content. Which sources form those answers, and how does this differ by industry? Peak Ace’s LLM Industry Benchmark analysed 1,000 standardised, commercially relevant prompts across ten industries in August 2026. This article looks at the telecommunications sector through its source mix, and at what this may mean for GEO and Digital PR.
Note: This article is an excerpt from the LLM Industry Benchmark. The complete analysis, including all ten industries, is available in the benchmark itself.
In short
- In the benchmark, comparison formats lead the telecommunications source mix for recommendation, comparison and pricing questions.
- For criteria and guidance questions, articles lead in telecommunications, while how-to guides lead across all industries.
- Telecommunications shows the most concentrated mention pattern of the ten industries in the benchmark.
- All findings refer to this prompt set, this time period (August 2026) and this setup.
Methodology and data corpus
The benchmark rests on a clearly defined data corpus. The following sections describe its structure and the key figures derived from it.
Prompt set and structure
The prompt set comprises 1,000 standardised bottom-of-funnel prompts, with 100 commercially relevant prompts per industry, distributed across ten topic clusters. All prompts are brand-neutral. They are geared towards purchasing decisions, that is, questions people ask shortly before signing a contract or booking a service.
Search intent
Each industry covers four search intents, with 25 prompts per intent:
- Best/Top/Test winner: recommendation questions, such as “Which mobile tariff is best for…?”
- Comparisons: contrasts, such as “Which is better, A or B?”
- Prices/Where to buy: questions about costs and sources, such as “Where can I order…?”
- Criteria/Guidance: questions about selection criteria, such as “What should I pay attention to when…?”
Market, language and time period
The prompts are in German and geared towards the German market. Responses were analysed from 1–31 August 2026.
Systems and repetitions
Gemini, ChatGPT and Google AI Overviews were observed. Each prompt was executed 15 times, five times per system. Individual responses are treated as samples, which allows fluctuations to be classified.
Database and key figures
The data comes from Peec AI exports. Fifty brands per industry were monitored, and reporting focuses on the top 20 brands by visibility.
- Visibility: how frequently an entity appears in responses to the prompt set.
- Share of Voice (SOV): the proportion of an entity among all mentions within the top 20.
- Sentiment: the tone of responses mentioning an entity, rated on a scale of 0 to 100.
These are indicators within the benchmark, not market share or absolute quality ratings. Source analysis measures retrieval, meaning that a source has been accessed. This does not automatically imply that it will be cited or that the brand will appear in the final answer.
Telecommunications sources mix
Source types differ by search intent and by industry. In telecommunications, the mix reflects what users ask for: a choice between providers, a tariff comparison, or an explanation of network and contract terms.
The table compares the leading source type in telecommunications with the leading source type across all ten industries.
| Intent | Telecommunications: leading source type | All ten industries: leading source type and retrieval share |
|---|---|---|
| Best/Top/Test winner | Comparison | Comparison, 28.31 per cent |
| Comparisons | Comparison | Comparison, 32.65 per cent |
| Criteria/Guidance | Article | How-to guide, 22.33 per cent |
| Prices/Where to buy | Comparison | Comparison, 22.64 per cent |
The cross-industry percentages show the retrieval mix across all industries, not telecommunications specifically. The telecommunications scorecard reports the order of leading source types, not their individual shares. “Other” is a residual category and not a specific format.
Three patterns stand out in the telecommunications data:
- Comparisons lead three of four intents. Comparison pages lead recommendation, comparison and pricing questions. Their role extends beyond explicit provider-versus-provider questions.
- Guidance changes the mix. For criteria and guidance questions, articles lead in telecommunications and how-to guides lead across all industries. Across industries, how-to guides account for 22.33 per cent of retrieval for this intent, compared with 5.81 per cent for recommendations.
- Category and product pages appear across intents. Category pages appear in every list and rank second for recommendations and pricing. Product pages feature in three lists.
The pattern is not universal, and the industry serves primarily as a lens. In Media & Publishing, comparisons top all four intent lists. In Health & Wellness, articles top three. The crucial factor is which source types are relevant for which search intent within a given product and service field.
Mention concentration in telecommunications
Average values do not show whether mentions are spread across many brands or concentrated on a few. The benchmark therefore considers two concentration metrics within the top 20 of each industry, both based on SOV shares normalised to 100 per cent:
- HHI: describes the concentration of SOV components. Higher values indicate a more uneven distribution.
- Top 3 share: the combined share of the three largest SOV contributions.
| Industry | HHI | Top 3 share in per cent |
|---|---|---|
| Telecommunications | 1,462 | 63 |
| Marketing & Advertising | 1,200 | 51 |
| Health & Wellness | 1,132 | 50 |
| SaaS & Technology | 548 | 24 |
In the telecommunications top 20, 63 per cent of normalised SOV falls to three entities. This may indicate a more uneven distribution of mentions in this sector, rather than different market opportunities. The prompt set likely explains part of this, as many well-known providers align directly with the tested purchasing decisions.
For context:
- Telecommunications: 12.60 per cent visibility, 4.15 per cent SOV, sentiment of 53.8
- All ten industries: 6.94 per cent visibility, 3.72 per cent SOV, sentiment of 54.58
Sentiment reflects the tone of answers, not network quality, service quality or customer satisfaction.
Which providers are most visible in this prompt set?
In this prompt set, Telekom (54 per cent visibility), Vodafone (50 per cent) and o2 (40 per cent) are the most visible during the observation period. This is a snapshot only, not market share or a provider ranking.
These operators sit at the centre of the answers because the prompts ask about their networks, tariffs and services. Their prominence reflects what is being evaluated, not superior content performance.
What this means for GEO and Digital PR
The data does not prove impact, but it points to where SEO and GEO teams may focus their baseline work.
Comparison formats
Comparisons lead three of four intents. Practical implication: check that tariffs are described correctly in comparison sources.
Key details should be consistent across sources: data allowance, contract length, recurring prices, introductory periods, one-off fees and conditions.
The question is not only whether a provider is mentioned, but whether there is enough clear information to explain when its offer fits.
Category and product pages
Practical implication: make tariff groups useful for selection, distinguishing offers by data allowance, contract length and requirements rather than promotional prices alone.
Keep tariff details aligned across category, product and comparison content.
Criteria and guidance
Articles and how-to guides dominate consultation. Practical implication: explain criteria such as network choice, tariff requirements and contract terms before asking for a choice, and link them to relevant offers.
Digital PR
Practical implication: identify the key third-party sources per intent and check what they say about your brand.
Build outreach around verifiable information about services and tariffs, as not every telecom publication is equally relevant.
Gordon Herenz, Team Lead Content Strategy & Digital PR at Peak Ace, recommends combining comparison and list formats with placements in relevant rankings and consistent Digital PR on independent sources.
Consistency and monitoring
Outdated or conflicting information raises the risk of inaccurate statements in generated answers. Keep product names, prices and terms consistent across your website and external sources.
Monitor brand mentions and accessed sources as an ongoing task, ideally with a consistent prompt set, analysed by category, intent and system.
Limits of the findings
The results are valid only for this prompt set, this time period and this setup. In particular, note the following:
- The prompt set shapes the outcome. Different questions or wording can lead to different answers.
- Individual responses are samples. Small differences should not be overinterpreted.
- Three systems are not one system. Gemini, ChatGPT and Google AI Overviews cannot simply be equated, and patterns may differ by engine.
- Retrieval is not citation. Accessed pages do not necessarily appear in the final answer, and the data does not establish causal mechanisms.
- Dialogue and personalisation are missing. Typical purchasing conversations cannot be reliably reconstructed from this dataset.
- Sentiment is not a quality assessment. It does not measure network quality or actual customer satisfaction.
Read the full benchmark
This article provides an excerpt. The complete LLM Industry Benchmark includes:
- The source mix for each of the ten industries and all four intent types
- Visibility, SOV and sentiment across the benchmark
- HHI and top 3 share within the top 20 of each industry
- The industry scorecards and the complete methodology
Had a look at the benchmark and want to strengthen your telecoms brand’s presence in AI answers? Get in touch with Peak Ace’s LLMO and GEO team.
FAQ: The telecommunications industry in LLMs
How visible are telecommunications brands in AI systems?
A cross-system visibility figure cannot be quantified reliably, because no unified measurement methodology covers all models, markets, prompts and points in time. The benchmark offers a standardised view for one prompt set, one period and three systems, aggregated across Gemini, ChatGPT and Google AI Overviews.
Which telecommunications brands are most visible?
This cannot be answered in general terms. Mentions in AI answers can change with the model, time, phrasing and context. The benchmark reports which brands were mentioned most often in this dataset during the observation period.
What do the visibility figures show?
They show how often brands appear in AI answers within the defined prompt set and setup. Top 3 share and concentration metrics describe how strongly mentions cluster within the top 20. They are not market shares and say nothing about real demand or revenue distribution.
Which content types do LLMs use for telecommunications questions?
In the benchmark, comparison formats often lead recommendation, comparison and pricing questions, while articles lead criteria and guidance questions. This applies to the examined prompt set and time period only.
Is benchmarking relevant for telecommunications brands?
As a benchmark, it adds a perspective alongside traffic, leads and conversions. It is not standalone proof of success, but a way to observe touchpoints, topic coverage and consistency in AI answers under constant conditions.
How can telecommunications providers improve their presence in AI answers?
Start with one tariff category and review its comparison, category, product and guidance content against typical customer questions: data allowance, contract length, prices, network and differences between offers. Fix outdated or unclear information, then re-check with the same test questions under the same conditions so that results remain comparable.
