Use Lumi to answer recurring website questions from your approved information, offer multilingual text or voice conversations, capture leads, share booking links and help connected SHOPLINE shoppers find products or manage a cart. Your team works from a private workspace containing conversations, leads, knowledge and first-party Customer Intelligence. The result is not generic automation: it is a consistent first response grounded in what your business has chosen to publish.
- Lumi answers recurring website questions from knowledge your business has reviewed and approved, rather than improvising from unrestricted information.
- Customer Lumi supports text chat, multilingual suggested questions, rich cards and plan- or setting-gated voice interaction.
- The private workspace brings together knowledge management, conversations, leads, first-party Customer Intelligence, team management and billing.
- Human attention remains essential for unusual requests, individual exceptions and decisions that depend on information Lumi does not have.
- One script tag adds Customer Lumi to a website; connected SHOPLINE stores can instead use the SHOPLINE App Embed.
Scale the first response, not the distance between your business and its customers
Lumi is an AI Concierge for small and mid-sized businesses and online stores. Customer Lumi sits on the business’s website as a chat and voice assistant, answering from information the business has reviewed for public use. This approach belongs within the wider question of matching AI communication to the right business and audience: the strongest use is not replacing every conversation, but removing the repetitive questions that consume the team’s day.
That distinction protects the personal touch. Routine questions receive a prompt, consistent answer, while the team has more capacity for a visitor whose circumstances require interpretation or an individual response. Lumi can also capture a lead or hand the visitor a booking link, creating a practical bridge from an automated first conversation to direct contact. See Lumi, the AI Concierge for your website for the product that delivers this experience.
Lumi turns repeated questions into consistent, grounded conversations
Customer Lumi’s text chat is current and working. It reads the business’s approved public information and tenant configuration, then records concierge conversations and messages. Text chat has been verified in 12 languages. Automatic multilingual starter questions help visitors begin, reducing the effort required to work out what to type.
Responses can include rich product or offering cards selected through references in the approved knowledge or catalog. Public cards use business-owned images from the website, merchant uploads or SHOPLINE; they do not use AI-generated images. This keeps the visual experience connected to the real products and services the visitor is considering. For a service business, that could mean explaining published offerings before directing the visitor toward a booking link. Teams considering that journey can compare it with exploring consultancy services from a business problem.
What Customer Lumi handles
Text, voice and multilingual access give visitors more natural ways to ask
Customer Lumi supports a mini-card launcher and both floating and embedded presentation. Its voice capability provides speech-to-text input and text-to-speech output when enabled by the applicable plan and settings. Voice input and output have been verified in English, Traditional Chinese, Japanese and Spanish, while text chat has been verified across 12 languages.
Voice input accepts WAV audio at 16 kHz with a 20 MB cap. Occasional mishearing has been observed in Traditional Chinese, so visitors should be able to see the resulting conversation and restate a question when necessary. Voice output uses one voice per business, with language managed through instruction. These boundaries make voice useful for accessibility and convenience without pretending that speech recognition is flawless.
The same principle applies in regulated or operationally sensitive contexts: present what the business has actually documented, then leave unsupported conclusions to the appropriate person. That discipline is illustrated by evaluating published shredding safeguards without assuming certifications.
A private workspace gives the team control behind the conversation
Visitors interact with Customer Lumi, while merchants work in a private workspace. The workspace contains knowledge management, conversations, leads, first-party analytics called Customer Intelligence, a private AI assistant called Business Lumi, team management and billing. Lumi supports multiple businesses per workspace and multiple workspaces per account, subject to plan limits.
This separation matters. The public concierge handles immediate visitor interaction, but the team still needs a reliable place to review what people asked, manage the information used in answers and follow up on captured interest. After an ownership or operational change, reviewing the underlying answers is especially important; reviewing Lumi answers after buying a dental practice shows the kind of continuity check a new team should prioritize.
Customer Intelligence classifies traffic as paid or campaign, organic, referral, direct or unknown, using first-touch attribution by default. Treat this as first-party context for understanding sessions and conversations, not as proof of revenue. Current SHOPLINE checkout reference work is experimental: it writes a random, non-personal reference into cart attributes, but Lumi does not have SHOPLINE customer identity or order access and therefore cannot confirm that a reference appeared on a completed order.
Use each layer for the right job
For online retail, separate standard policy answers from requests that need discretion. The workflow in triaging e-commerce policy questions and individual exception requests demonstrates why that boundary preserves both speed and personal service.
Setup starts with the website knowledge your team already owns
A business adds one script tag to load Customer Lumi in an iframe, or enables the SHOPLINE App Embed for a connected store. The script installation has been verified on six live sites. The business’s domain must be registered and enabled, and wildcard domains are not supported. Choose intentionally whether the concierge appears as a floating or embedded experience based on where visitors need help.
- Identify the recurring questions that absorb the most team time and collect the public pages that answer them.
- Use website discovery to read and organize up to 25 pages into pending knowledge items. This requires an eligible trial, plan or company status.
- Review and approve the knowledge before expecting Lumi to answer from it; no answer is provided from unapproved material.
- Add the website script or turn on the SHOPLINE App Embed, then choose where Customer Lumi should appear.
- Test text responses, suggested questions, rich cards, lead capture and any enabled voice experience against real visitor wording.
- Review conversations and leads in the private workspace, improve unclear source material and define which requests go directly to a person.
When the business serves customers across regions, operational details deserve the same care as product and policy information. For example, clarifying overseas e-commerce callback time zones before responding helps prevent an apparently simple handoff from creating a poor personal experience.
For businesses in Orange County, Los Angeles and San Diego, CA
Across these busy Southern California markets, growing service businesses and online stores may need to respond while customers compare providers, products, availability and policies online. The personal touch can erode when teams repeatedly answer the same questions or miss inquiries during high-demand periods. Start by reviewing the questions visitors ask most often, then publish clear, approved answers about services, locations, hours, booking and delivery. Lumi can use that information for an immediate first response, while the team keeps responsibility for nuanced requests, exceptions and relationship-sensitive follow-up.
Human escalation is what keeps scaled communication personal
The goal is not to remove people from customer communication. It is to stop using their attention on answers that can be delivered consistently from reviewed information. A strong operating model gives Lumi ownership of the immediate, repeatable first response and gives the team ownership of judgment, exceptions and relationship-sensitive follow-up.
Define that boundary in advance. Published hours, service descriptions, product discovery and standard policies fit the concierge well when they are present in the approved knowledge. Requests for an exception, interpretation of a sensitive situation or a decision not covered by that material should move to the business. This is how automation increases responsiveness without making the customer experience feel evasive.
Lumi is part of the OceSha ecosystem and is powered by OceSha. It is a product rather than a separate legal entity; OceSha Inc. is the billing seller. Readers who need context about the organization behind it can review OceSha Ventures and its AI-first solutions, while billing-specific questions are addressed in why Lumi billing names OceSha Inc..
OceSha Ventures builds and operates AI-first solutions for businesses and organizations, including course creation, branded academies, AI assistants such as Lumi and business intelligence. If you want to map your recurring questions, website knowledge and escalation needs to the product, contact the Lumi team about your communication workflow.
Lumi gives each website visitor an immediate, business-specific conversation while the team retains control of the knowledge, leads, conversations and follow-up behind it.
Talk to the Lumi teamFrequently asked questions
Does Lumi answer questions before the business approves its knowledge?
No. Customer Lumi does not answer from knowledge before it has been approved. Website discovery can organize material from up to 25 public pages into pending knowledge items for review.
Can visitors speak to Lumi instead of typing?
Yes, when voice is enabled by the applicable plan and settings. Speech input and output have been verified in English, Traditional Chinese, Japanese and Spanish. Voice input uses WAV audio at 16 kHz with a 20 MB cap.
Can Lumi support more than one business?
Yes. Multiple businesses per workspace and multiple workspaces per account are current and working, subject to plan limits. Proposed plan allocations are not final.
Does Lumi replace the customer service team?
No. It handles recurring questions grounded in reviewed public information and creates handoffs through lead capture or booking links. People remain responsible for exceptions, sensitive requests and decisions that require additional context.
Can Lumi prove that a SHOPLINE conversation produced an order?
No. Current checkout reference work is experimental. Lumi lacks SHOPLINE customer identity and order access, so it cannot confirm that a recorded cart reference appeared on a completed order or claim verified revenue attribution.
How are Lumi subscriptions structured?
Lumi supports subscriptions and trials, with monthly AI credits and optional credit top-ups. A 100-credit top-up for $12 has been tested. Other proposed prices, credit accounting and plan structures are not final.
The bottom line
Growing businesses should automate repetition, not relationships. Lumi gives website visitors an immediate text or voice conversation grounded in information the business has reviewed, then supports lead capture, booking handoff, product discovery and connected SHOPLINE cart assistance. The private workspace keeps knowledge, conversations, leads and first-party Customer Intelligence accessible to the team. The best rollout starts with a focused set of accurate, high-frequency questions and an explicit human path for exceptions. That combination scales responsiveness while protecting the judgment and care that make communication feel personal.
Rohan Hall
Founder of OceSha Ventures · AI architect and author
Rohan Hall is a technology entrepreneur, AI architect and author with years of technology experience, now focused on practical AI across business, education, government and global impact. He founded OceSha Ventures, builds the OceSha AI platform and Lumi, and wrote The Convergence of AI and the Top 10 Emerging Technologies.
OceSha Ventures — OceSha Ventures builds and operates AI-first solutions — course creation, branded academies, AI assistants such as Lumi, and business intelligence — for businesses and organizations.
Sources
About this page
It is based on Lumi's current approved product information. Product details come from Lumi's current product capabilities audit (Oct 2026).