Why BigCommerce stores need an AI chatbot
BigCommerce sits in an awkward middle space between Shopify's DTC volume game and Adobe Commerce's enterprise-only posture. Most BigCommerce merchants are mid-market retailers, growing B2B operators, or brands running a hybrid hosted-and-headless setup who picked BigCommerce specifically because they wanted SKU depth, multi-storefront, and complex catalog logic without paying for Magento engineers.
The shared trait across all of those buyers: average order value is meaningfully higher than the average Shopify store. The 2025 mid-year BigCommerce merchant report had average BigCommerce AOV at roughly 2x the average hosted Shopify AOV, with B2B Edition stores tracking 5 to 10x. Higher AOV means each abandoned session, each unanswered product question, and each lost lead costs more. A chatbot that recovers even a small percentage of pre-cart drop-off has a much larger absolute return than the same chatbot on a $35-AOV Shopify boutique.
At the same time, BigCommerce stores tend to have deeper, more technical catalogs (product variants, modifiers, bulk pricing tiers, customer-group pricing, B2B price lists) that a buyer cannot navigate in 30 seconds the way they can a flat DTC catalog. The questions visitors ask are denser: "what is the bulk discount for 50 units of SKU X to a Net-30 wholesale account?" That is exactly the kind of question a well-trained AI chatbot can answer instantly from your published pricing pages, where a human agent would need three minutes and two tab-switches.
What "AI chatbot" means for BigCommerce specifically
"AI chatbot" gets used to mean three different products in the BigCommerce ecosystem, and they are not interchangeable.
The first is a rule-based / decision-tree chat widget. These match keywords to canned answers from a flowchart you build by hand. They are the cheapest and the least useful past the second turn of a conversation. Most legacy "BigCommerce chatbot apps" on the Marketplace are this category.
The second is an LLM-grounded answer engine that indexes your storefront content and answers in natural language using only that content. This is what ChatRaj is. The bot reads your storefront pages once, builds a private knowledge base, and uses retrieval-augmented generation so every answer is grounded in your actual catalog and policies. No flowchart. No hallucinated product names. New product pages are picked up on the next scheduled re-index.
The third is a full conversational commerce agent that takes actions: applies discount codes, updates the cart, creates orders, looks up live order status through the BigCommerce Admin API. These are powerful but require the merchant to grant an OAuth app broad Admin API scopes, which is an actual security decision and not a 60-second install.
If you want a same-day install that handles 60 to 80 percent of pre-sales questions and email capture, the LLM-grounded answer engine is the right choice. That is what the rest of this page is about.
ChatRaj on BigCommerce: the 60-second install
The install path looks like this. Open your BigCommerce control panel. Navigate to Settings, then Scripts. Click Create a Script in the upper right. Pick "Manually" as the script type. Set Location on page to Footer. Set Select pages where script will be added to All pages. Paste this single line into the Scripts contents box:
<script async src="https://chatraj.com/widget.js" data-bot-id="YOUR_BOT_ID"></script>
Save. That is the entire integration. The chatbot loads asynchronously after your storefront renders, never blocks Largest Contentful Paint, and shows up as a floating bubble in the bottom-right corner of every page. Scripts Manager is the BigCommerce-supported way to do exactly this kind of injection without editing Stencil theme files yourself, and it survives theme upgrades, theme switches, and Page Builder edits.
The only field that needs your actual value is the data-bot-id attribute, which you copy from the Embed tab inside your ChatRaj dashboard after you sign up and create your first bot. Everything else stays exactly as written.
Compared to a BigCommerce App Marketplace install, this skips the OAuth consent screen, skips the Admin API scope grant, and skips the monthly Marketplace billing arrangement. You manage the bot from chatraj.com instead of from inside the BigCommerce control panel, which is usually what mid-market operators actually want anyway: the marketing team can manage the chatbot without getting access to your store admin.
What ChatRaj ingests from your BigCommerce store
When you point ChatRaj at your storefront on the Sources tab, you have three sensible options.
The first is your domain root, e.g. https://yourstore.com. ChatRaj crawls up to 20 pages by default on the free tier and more on paid plans, starting from your home page and following internal links. Good for a quick proof-of-concept on a small catalog.
The second is your sitemap.xml URL. Every BigCommerce store automatically publishes a sitemap at yourstore.com/sitemap.xml. Submit that and ChatRaj walks every product page, category page, blog post, and information page that you publish, in one shot. This is the right answer for any catalog larger than 50 SKUs.
The third is a list of specific URLs you care about. Useful for B2B stores that want to scope the bot to a public catalog but exclude gated wholesale pages. You can also exclude specific URL patterns from indexing.
ChatRaj extracts the main content from each page, strips navigation, headers, footers, and Page Builder boilerplate, and indexes the remaining text into a private knowledge base scoped to your chatbot_id. The bot answers using only that knowledge. If something is not on your published storefront, the bot will say so rather than invent an answer. New product pages are picked up on the next scheduled re-index (daily on paid plans).
Lead capture for high-AOV BigCommerce baskets
The economics of lead capture are different on BigCommerce than on a typical DTC Shopify store. A captured email on a store with a $180 average order value and a 2 percent recovery rate from a downstream email automation is worth roughly $3.60 of expected revenue per email. That math makes ChatRaj's Pro plan ($29 per month) pay for itself at roughly 9 recovered emails in a month, which most stores hit in the first week.
The pattern works like this. A visitor lands on a product page from a paid ad. They have a specific question that they cannot find an answer to in the spec table. They open the chat. ChatRaj answers their question, then offers to send them a 10 percent off code if they want it. They drop their email. The email flows out of ChatRaj via webhook (or CSV export) into Klaviyo, Mailchimp, or your existing CRM. If the visitor still does not convert, your standard abandoned-browse automation takes over the next day.
For B2B Edition stores, the same lead-capture flow works for capturing wholesale-account requests: ChatRaj answers a tier-pricing question, then offers to route the visitor to a sales rep if they want a custom quote, capturing company name plus business email instead of just an email. Those leads flow into the same webhook destination as your DTC leads, tagged differently.
ChatRaj vs BigCommerce App Marketplace alternatives
The BigCommerce App Marketplace lists a meaningful number of AI chat apps, ranging from rules-based widgets like Tidio and HelpCrunch to LLM-backed apps like ELX and full conversational-commerce platforms like Gorgias and eesel AI. They are not bad products. They are just structurally a different deal than a script-tag install.
App Marketplace installs go through BigCommerce OAuth. You grant the app a set of Admin API scopes (typically Products read, Information and Settings read, Storefront API tokens write, sometimes Customers read or Orders read). The app then has API access to your store data going forward, until you uninstall it and revoke the token. For some chatbot use cases (live order status lookup, cart manipulation) this is genuinely required. For pre-sales Q&A and email capture, it is not.
ChatRaj's storefront widget runs entirely in the visitor's browser. The widget fetches your bot's knowledge base from chatraj.com when a visitor opens the chat, sends the visitor's messages to our inference API, and renders the answers. No part of that loop ever touches your BigCommerce Admin API. There is no token to revoke later. There is no scope to audit. The blast radius if our service has a security incident is the chat content visitors typed into the widget, not your product catalog or customer list.
The other structural difference is billing. Apps installed through the BigCommerce App Marketplace bill through BigCommerce's billing arrangement. The Marketplace economics ultimately get reflected in pricing. ChatRaj billed direct from chatraj.com does not have that layer.
Common BigCommerce install gotchas (Page Builder, custom themes, headless)
A handful of edge cases come up often enough to call out.
Page Builder pages. Page Builder is BigCommerce's drag-and-drop storefront builder for marketing pages. Scripts Manager injects into Page Builder pages exactly the same way it injects into Stencil-templated pages, because the Footer location renders on every page the storefront serves. You do not need a separate install for Page Builder pages.
Custom Stencil themes. If your store runs a custom or heavily-modified Stencil theme, the Scripts Manager Footer location still works as long as the theme correctly includes the {{footer.scripts}} handlebars helper, which is part of the Stencil theme contract and present in every Cornerstone-derived theme. If your developer ripped that helper out, the Scripts Manager footer scripts will not render. Fix: add the {{footer.scripts}} helper back into your theme's footer.html partial. This is a one-line fix and a good practice anyway, since BigCommerce's own platform features rely on it.
Headless BigCommerce. If you run a headless storefront (Catalyst, a custom Next.js front end, BigCommerce + Hydrogen, or Composable Commerce setups), Scripts Manager scripts do not render because the storefront HTML is served from your own front end, not BigCommerce. The fix is straightforward: paste the same script tag directly into your custom front end's HTML shell (the closest equivalent to a footer partial in your framework). For a Next.js Catalyst storefront, that is typically a Script component inside your root layout. The chatbot loads identically once embedded.
Multi-storefront. If you run multi-storefront on BigCommerce (Enterprise plan and B2B Edition), Scripts Manager scoping is per-channel. Each storefront has its own Scripts Manager and you add the chatbot script once per storefront. You can use the same bot across storefronts or create separate bots per storefront depending on whether the catalogs differ. Most multi-storefront operators use a separate bot per storefront so the knowledge base matches what visitors see.
B2B Edition. ChatRaj works inside B2B Edition the same way it works on a standard storefront. Public catalog pages are indexed; gated B2B pages behind login can be indexed if you provide ChatRaj a service account login, but most operators leave the gated pricing pages out of the index and answer those questions with a "talk to a sales rep" intent instead.
When ChatRaj is NOT the right call (you need agent-assist + ticketing)
A few honest signals that ChatRaj is not the right tool for your BigCommerce store.
You need real-time order status answers ("where is my BigCommerce order #1234?"). ChatRaj indexes your public storefront and does not connect to the BigCommerce Admin API. For order-status queries the right answer is BigCommerce's built-in order status page (already linked in every confirmation email), an integration like Gorgias, or eesel AI which connect via Admin API for that exact reason.
You need an integrated ticketing / agent-assist platform. ChatRaj answers visitor questions and captures leads. It does not route conversations to a human agent's inbox, does not have a built-in ticketing layer, and does not blend AI replies with human handoff in the way Gorgias or Re:amaze do. If your business model requires a unified queue for AI replies plus human escalations across email, chat, and social, you are buying a help-desk platform, not a chatbot. That is a different product category.
You need a B2B sales-enablement chat that proactively pushes account managers into conversations. ChatRaj is a self-serve answer engine. It does not (yet) support the proactive routing patterns that BigCommerce B2B sales-rep teams sometimes want. Those teams should look at Drift, Qualified, or a similar conversational-sales platform.
For everyone else, especially BigCommerce stores looking to deflect 60 to 80 percent of pre-sales product questions and capture more email leads on higher-AOV baskets, the 60-second Scripts Manager install path is real, and the free tier is enough to confirm whether the bot's answers are good enough for your specific catalog before you spend a dollar.