CASE STUDY / US HOME & RENOVATION PRODUCTS

From ~2% to 47% AI visibility.

AI visibility rose from about 2% to 47% across more than 200 commercial prompt clusters during a 12-month GEO engagement for a US home and renovation products seller with approximately RMB 30 million in annual revenue.

US market · 12-month GEO engagement

B2C · US home & renovation products

AI visibility: ~2% → 47%

A business with approximately RMB 30M in annual revenue expanded its AI presence across 200+ core commercial prompt clusters.

AI visibility 0–100% scale
Starting point~2%
After 12 months47%
Core commercial prompt clusters
200+
Appearance in high-purchase-intent prompts
53%
AI recommendation rate
32%
AI referrals as a share of total website traffic
~3.1%
AI-influenced US order journeys~8–12%
Monthly new or assisted revenue at project maturity~US$30k–52k

Project-reported figures. AI-influenced orders and assisted revenue are not equivalent to fully incremental, AI-attributed sales.

USD figures are rounded equivalents using CNY 6.7132 per US dollar. ECB reference rates, September 25, 2026.

A US consumer market engagement.

The company sells home and renovation products directly to consumers in the United States, with annual revenue of approximately RMB 30 million. The GEO engagement focused on its US market over 12 months.

The project tracked more than 200 core commercial prompt clusters. A prompt cluster groups related questions around a commercial topic or buying need. The supplied results describe the position after 12 months of service.

The starting point.

The site had few backlinks. Many product descriptions were embedded in images that AI systems could not read, and product information relied on promotional language rather than specifications and data.

Shipping and return policies were not stated clearly. The work focused on these gaps: readable product text with concrete details, explicit shipping and return policies, and a stronger backlink profile.

Visibility across commercial buying questions.

Reported AI visibility increased from approximately 2% to 47%. Appearance in high-purchase-intent prompts reached 53%, while the AI recommendation rate reached 32%.

The three project metrics cover different aspects of commercial discovery: visibility across core question clusters, presence in high-purchase-intent prompts and AI recommendations. Each is presented separately so the result remains clear.

Traffic and purchase journeys.

AI referral traffic represented approximately 3.1% of total website traffic. Separately, an estimated 8–12% of US order purchase journeys were influenced by AI search or AI recommendations.

The referral figure captures identifiable website traffic from AI sources. The purchase-journey figure captures a broader influence from AI search or recommendations during buying. Together, they describe how discovery can connect with customer decisions.

Approximately US$30k–52k per month at project maturity.

At project maturity, GEO reportedly contributed the equivalent of approximately US$30,000–52,000 per month in new or assisted revenue. The range describes the mature stage of the engagement and is rounded from the original RMB figures.

The revenue measure includes both new business and sales assisted along the purchase journey. It is presented with that attribution scope, alongside the traffic and visibility measures.

Read the result as a project account.

The client is anonymized and all figures are project-reported. Visibility percentages follow the project’s measurement definitions; order influence and assisted revenue describe contribution across the buying journey.

The practical lesson is to evaluate a GEO program at several levels: visibility in commercially relevant questions, identifiable referral traffic, influence on purchase journeys and business outcomes with their attribution limits stated.

CLIENT FEEDBACK

A client’s perspective.

What we appreciated most was that they did not start by asking us to publish large volumes of articles. They first worked through the website’s underlying issues.

They looked closely at whether product pages had complete information, whether AI could crawl and understand them, whether facts conflicted across pages, where content lacked evidence, and which keywords reflected real buying intent.

The recommendations were easy to follow, with almost every item laid out as Problem, Why, Action and Priority. This made execution straightforward for our internal team.

The team was also a pleasure to work with. They proactively followed up on details instead of stopping at report delivery. For an independent online store, it is not easy to find a GEO team that understands technology, content and conversion so well.

B2C · US home & renovation online storeFeedback translated from Chinese. Client identity withheld.

How to read these results.

Source
Project team reports supplied to SemShift. Client identity is withheld.
Period
12 months serving the US market; revenue range describes the mature stage.
Comparison
AI visibility rose from approximately 2% to 47% across more than 200 core commercial prompt clusters.
Measurement
Prompt presence, recommendation rate, referral traffic, AI-influenced order journeys and new or assisted revenue are distinct project-reported measures.
Limits
Prompt sampling details and the full attribution model were not supplied. US$30k–52k/month is a rounded conversion of RMB 200k–350k using ECB rates on September 25, 2026; it includes assisted revenue.
FROM READING TO A REAL CONVERSATION

Build a GEO plan for your priority market.

Talk to SemShift