Choose a website answer tool that uses reviewed knowledge, fits the visitor journey and shows you what customers need

Start with controlled answers, not unrestricted generation

The first selection criterion is knowledge control. Lumi, the AI Concierge, answers website visitors only from approved public knowledge belonging to that business. No answer is available before the relevant material has been approved. That constraint is valuable: it gives the business a defined point at which discovered information becomes suitable for customer-facing use.

Lumi’s website discovery process reads up to 25 public website pages and organizes the content into pending knowledge items. A business then reviews what was found before it becomes available to Customer Lumi. Start with the broader principles for helping customers get reliable business answers, then inspect the operational path from discovery to review and publication.

Review must continue after launch. Website copy, policies and offerings change, and new source material can conflict with information that was previously accepted. A sound process surfaces version conflicts for review rather than silently replacing earlier facts. If your website is reimported, establish who will recheck answers after website reimports and how discrepancies will be resolved.

Assess how visitors can ask questions and understand the response

A useful concierge should support the interaction styles your visitors actually use. Customer Lumi currently provides text chat through a mini-card launcher and can appear in floating or embedded form. Text chat has been verified in 12 languages. It also generates multilingual suggested questions automatically, helping visitors begin without having to formulate a perfect first prompt.

Voice input and output are current capabilities, but they are controlled by plan and settings. Voice input converts speech to text and has been verified in English, Traditional Chinese, Japanese and Spanish. Input uses 16 kHz WAV audio with a 20 MB cap, and occasional Traditional Chinese mishearing has been observed. Voice output has also been verified in those four languages; each business uses one voice, with language directed through instruction.

Rich cards should also preserve the integrity of the business’s presentation. Lumi’s public cards use images from the business website, merchant uploads or a connected SHOPLINE catalog; they do not use AI-generated images. When consistency is a priority, compare how structured answers differ from general chat in approved website answers versus chat for boutique policies.

Look beyond answers to the next customer action

Answer quality is only part of the customer journey. A website assistant should make the appropriate next step clear without pretending that an early action is a completed result. Lumi can capture a lead through an optional in-chat form containing name, email, company, phone and challenge fields. It can also hand the visitor a booking link.

Keep outcome language precise. Opening or starting a booking flow is not the same as completing or confirming an appointment. This distinction affects reporting and follow-up: the business still needs to know whether the visitor finished the external process. For questions where timing matters, compare immediate website answers with next-day inbox replies rather than treating every response channel as equivalent.

Email automation serves a different role. An acknowledgment can confirm that a message arrived, while an on-site answer addresses the visitor’s question during the website session. Review email acknowledgments versus approved website answers when deciding which interaction belongs at each stage.

For commerce, require a real catalog connection and honest attribution

Commerce answers should be based on current store information rather than generic product descriptions. When a business connects a SHOPLINE store, Lumi uses synchronized catalog information for product answers. Conversational commerce tools support product search, variant selection and item presentation, and Customer Lumi can help shoppers manage their real store cart.

This capability requires the SHOPLINE connection. Product search has been verified with a development store, and there is no dedicated product-comparison tool. That makes the distinction between discovery and comparison important: Lumi can help a shopper find and select relevant catalog items, but a merchant should not assume a separate structured comparison function.

Attribution deserves equal scrutiny. SHOPLINE cart checkout-reference tagging is experimental. It does not confirm an order and should not be presented as revenue attribution. Current Customer Intelligence is based on Lumi interactions and first-party website journeys where Lumi is installed; it is not unified with Google Analytics, CRM or order data.

The same discipline applies in service businesses. A website tool should not imply access to live operational details it does not hold. The question of whether plumbing answers know a technician’s location illustrates why buyers should separate published business information from live workforce data.

Examine installation, administration and evidence of customer demand

Installation should be understandable at the website level. Lumi loads Customer Lumi through one script tag in an iframe, and the business registers and enables the domain where it will appear. The installation has been verified on six live sites. Wildcard domains are not supported, so each relevant domain needs to be handled explicitly.

The private workspace brings together knowledge management, conversations, leads, Customer Intelligence, a private assistant called Business Lumi, team management, installation management and billing. This is the operational layer that turns customer questions into a maintenance cycle rather than a one-time deployment.

Before a redesign, review the source material that will carry forward. The workflow for reviewing rewritten answers before a redesign launch helps prevent polished new copy from introducing inconsistencies into customer-facing responses.

Lumi supports subscriptions and trials, with monthly AI credits and optional top-ups. Usage therefore belongs in the selection process: determine which interactions consume credits and how the business will monitor demand, rather than evaluating only the launcher’s appearance.

For businesses in Orange County, Los Angeles and San Diego, CA

In these busy Southern California markets, service businesses, retailers and other local operators often face visitors comparing several options online before making contact. A website question-answering tool should provide reliable, business-approved details about services, availability, locations and next steps while making it easy to continue toward an inquiry or booking. Start by reviewing the questions your team answers most often by phone, email or at the counter. Turn those responses into approved source material, then test the tool with real customer wording—including relevant English and Spanish queries—before expanding its role.

Verify the company behind the tool and every capability you depend on

Lumi is an AI Concierge in the OceSha ecosystem and is also the AI Concierge for OceSha.ai. It is powered by OceSha and sold by OceSha Inc.; Lumi is a product, not a separate legal entity or seller. OceSha Ventures builds and operates AI-first solutions spanning course creation, branded academies, AI assistants such as Lumi and business intelligence for businesses and organizations.

Evaluate current, connected, limited and planned functions separately. A target audience or desired outcome is not proof of a deployed feature. The same applies to provider names: the durable question is whether the capability works for your business, not which underlying model happens to be involved.

Have accountable people review the information before enabling public answers. A useful starting point is how a business should review and approve AI knowledge. If you need to confirm installation, plan gating, SHOPLINE connectivity or another requirement for your configuration, talk with the Lumi team about your website.

For a direct view of the customer-facing product, explore Lumi as an AI Concierge for your website. Judge it against the requirements that matter here: controlled knowledge, appropriate interaction formats, honest handoffs, explicit connection requirements and an ongoing review workflow.

The bottom line