EE One scored 60/100 for AI visibility
EE One was tested across 75 AI-led buyer prompts covering household connectivity, bundled mobile and broadband services, savings, flexibility, enrolment and competitor comparisons. The audit showed a stronger overall AI visibility profile than many products, but also revealed where prompt ambiguity and weak early-stage discovery can limit product visibility.
This is a public demonstration audit. EE Limited is not represented as an Odyssiant customer, partner or participant.
Why this audit matters
EE One shows a different kind of AI visibility pattern from the SAP Business Suite audit. The product achieved a stronger overall score, with particularly good performance in evaluation-stage competitor comparisons. But the audit still found important gaps.
The issue was not simply whether AI recognised EE. The more useful question was whether AI could surface EE One clearly when buyers were researching bundled household connectivity, savings, remote working reliability, enrolment and flexibility.
That is where product-level measurement becomes valuable. It shows where AI understands the product, where it needs the product to be named directly, and where prompt wording causes the answer to drift away from the intended category.
What was measured
- Company assessed
- EE Limited
- Product assessed
- EE One
- Audit type
- Product-level AI visibility audit
- Engine tested
- openai_gpt4o
- Run date
- 1 May 2026
- Total prompts
- 75
- Answers scored
- 75
- Coverage
- 100%
Themes audited
Buyer stages audited
The headline finding
EE One achieved an overall AI visibility score of 60/100. This is a stronger result than many products, but the stage-by-stage breakdown shows why one overall score is not enough.
Visibility was strongest at evaluation, where buyers asked direct comparison questions involving named alternatives such as Virgin Media Bundle, O2 Home and Vodafone Together.
The weaker areas appeared in awareness and decision-stage prompts, especially where the wording became broad, ambiguous or disconnected from UK mobile-broadband bundle context.
The strongest stage was Evaluation at 77/100. This suggests EE One performs well when AI is asked to compare named providers or specific bundle scenarios. The lower Awareness and Decision scores show that early discovery and evidence-led confidence still have room to improve.
EE One performed best in competitor comparisons
The highest-scoring answers came when buyers asked specific, comparison-led questions. These prompts gave AI enough context to connect EE One to the buyer's need and to named alternatives.
This matters because many buyers use AI to compare options before visiting provider websites or speaking to sales teams. Strong evaluation performance means the product has a better chance of being considered once it reaches the shortlist conversation.
AI produced a strong comparison answer using reliability and customer sentiment signals. This was the highest-scoring prompt in the audit.
AI connected EE One to remote-work connectivity needs and compared it against a named alternative.
AI was able to explain EE One's bundle benefits and compare savings against a named competitor.
AI identified EE One's enrolment process and compared the experience against another bundled-service proposition.
AI provided a useful explanation of EE One's combined broadband and mobile benefits.
The weak spots came from ambiguous or broad prompts
The lowest-scoring examples show why prompt wording matters. Some prompts were too generic, too broad, or drifted into the wrong context. In those moments, AI produced an answer, but not necessarily one that helped EE One appear as a relevant option.
This is important because real buyers do not always phrase questions perfectly. They may ask about "household plans", "bundled services", "consumer protections" or "security standards" without naming a category precisely. If AI interprets the question in the wrong context, product visibility suffers.
The prompt was too generic and unresolved. AI answered as if discussing broad SaaS or cloud compliance rather than mobile-broadband bundles.
AI produced a consumer-protection answer, but it drifted into a US regulatory context rather than a UK mobile-broadband bundle context.
AI interpreted "household plan" as architecture, budgeting or scheduling rather than telecoms, broadband or mobile bundles.
AI answered broadly about bundled services and overpayment risk, but did not strongly connect the answer to EE One.
AI interpreted the prompt as energy, home improvement or household budgeting rather than mobile-broadband bundle comparison.
Theme performance was consistent, but not complete
Unlike some audits where one theme collapses sharply, EE One showed relatively consistent theme performance. All six themes scored between 57/100 and 65/100.
That consistency is a positive signal. It suggests AI has some understanding of the product across savings, connectivity, flexibility, enrolment, market competitiveness and consumer insight.
But no theme scored above 65/100, which means the evidence base and product signals could still be strengthened.
Market Competitiveness was the strongest theme at 65/100, while Consumer Insights was the weakest at 57/100. The narrow spread shows that EE One is not invisible, but it also shows that stronger proof, clearer category language and more UK-specific evidence could improve AI confidence.
What the audit proves
The EE One audit proves that product-level AI visibility measurement is not only useful when scores are low. It is equally valuable when the product is already performing reasonably well.
A 60/100 score tells a more nuanced story than "visible" or "not visible". It shows where the product is strong, where it is vulnerable, and which types of buyer questions produce weaker answers.
From one audit, Odyssiant can show:
- where the product appears clearly
- where AI needs a named comparison to perform well
- which prompts drift into the wrong category or country context
- which buyer stages are strongest and weakest
- which product themes are consistently understood
- what content, proof and category signals should be strengthened next
What Odyssiant would fix first
Tighten awareness-stage category language
Create content that connects EE One to broader mobile-broadband bundle, household connectivity, remote work and family-plan research questions.
Strengthen UK-specific consumer evidence
Improve visibility of UK consumer-protection, reliability, satisfaction and switching evidence so AI does not drift into US or generic regulatory contexts.
Build clearer comparison content
Support prompts comparing EE One with Virgin Media, O2, Vodafone and other bundle alternatives using product-specific evidence.
Retest and track movement
Run the same prompt set again to see whether better content, proof and third-party signals change the answers.
The takeaway
EE One's audit shows why AI visibility is not a simple brand mention metric. A product can be reasonably visible overall and still have meaningful gaps by stage, theme and prompt wording.
The strongest results came from named competitor comparisons. The weakest results came from broad or ambiguous prompts that pulled AI into the wrong context.
That is exactly why product-level audits matter. They show not only whether AI mentions a product, but when, why and how confidently it appears.
More from this series
Run your first AI visibility audit
Sign up for a free Odyssiant account and see where your product appears, where it is misunderstood, and what to fix first.