Dprime AI Smart Woo Chatbot v1.0.8 Pro WooCommerce

An AI shopping assistant for WooCommerce. Answers product questions, searches your catalogue, and guides customers to the right product — no lead forms, no scripts, no friction.

Installation

  1. Install and activate WooCommerce and Dprime Core.
  2. Download Dprime AI Smart Woo Chatbot from your account at my.dprime.au.
  3. Install and activate it. It appears under Dprime > AI Smart Woo Chatbot.
  4. Add your OpenAI API key on the Settings tab, click Test connection to confirm it works, then sync your products.
Requires an active Dprime subscription and WooCommerce. Dprime Core handles licence verification. The chatbot will not initialise if WooCommerce is not active.

OpenAI setup

The chatbot uses the OpenAI API to generate replies. You need an API key from platform.openai.com.

  1. Go to the Settings tab in the plugin admin.
  2. Paste your API key into the API key field and tick Enable AI.
  3. Click Test connection. A green message confirms the key works and shows the active model.
  4. Choose a model. gpt-4o-mini is the default — fast and inexpensive. Use gpt-4o for more nuanced product matching.
  5. Set Max AI replies per session to control cost. After this number of replies the chat shows a message asking the customer to contact you directly.
ModelBest forSupports Live mode?
gpt-4o-miniMost stores — fast, low cost, accurate enough for product matchingYes
gpt-4oComplex catalogues, nuanced questions, higher accuracy neededYes
gpt-4.1-miniAlternative low-cost option with strong instruction followingYes
gpt-4.1Maximum accuracy, higher costYes

Product access

The AI can access your WooCommerce catalogue in two ways. Choose on the Settings tab under Access mode.

ModeHow it worksBest for
SyncProducts are cached when you click Sync now. The cache is sent to the AI on every chat turn.Catalogues up to ~500 products. Fastest replies.
LiveThe AI uses a tool call to search WooCommerce in real time when it needs product information.Large catalogues or when stock status must always be current.

Sync mode

After enabling Sync mode, click Sync now on the Settings tab. The plugin fetches your published products and stores them. Re-sync whenever you add new products or make significant changes.

Use the Product fields checkboxes to choose which data the AI sees: name (always included), price, short description, categories, and stock status. Including more fields gives the AI more to work with but increases token usage per turn.

Set Product sync limit to cap how many products are included in the cache. For large catalogues, increase this or switch to Live mode.

Live mode

The AI issues a search tool call when a customer asks about a product. This means replies are always based on current stock and pricing. Requires a tool-capable model (all currently supported models qualify).

Set Live search result limit to control how many products the AI gets per search. Lower values reduce token cost; higher values give the AI more context to work with.

Knowledge Base

Add sections for information the AI should know beyond the product catalogue: shipping policy, return conditions, size guides, warranties, FAQ answers. The AI uses this alongside product data when generating replies.

Writing good content

  • Keep each section focused on one topic. A section called "Shipping and returns" is harder for the AI to use than two separate sections.
  • Write factual statements. "We offer free shipping on orders over $100" is better than "We have a great shipping policy."
  • For FAQs, write the question on one line and the answer on the next. The AI will use both.
  • Avoid marketing language. The AI will echo what you write, so neutral facts produce more trustworthy answers.

Appearance

Configured on the Appearance tab. Theme changes save immediately on click — no save button needed.

SettingEffect
ThemeDefault (blue header) or Dark Mode (navy panel). Saves instantly on click.
Text sizeSmall, Medium, Large, or Extra Large. Scales all text inside the chat panel.
Bubble colourBackground of the launcher button and chat header.
Accent colourUsed on action buttons and link highlights inside the panel.
Bubble positionSeven positions: bottom-right, bottom-left, bottom-centre, middle-right, middle-left, top-right, top-left.
OffsetsHorizontal and vertical distance in px from the chosen corner. Separate mobile offsets let you clear other floating elements like cart buttons.
Launcher greetingA small text bubble shown next to the launcher before the visitor opens the chat. Toggle on or off; set custom text up to 200 characters.
Show header barDisplays the bot name at the top of the open chat panel.

Usage tracking

The Usage tab shows AI turn counts, token totals (prompt and completion separately), and estimated OpenAI cost for the last 7, 30, or 90 days. Data is broken down by model so you can compare costs if you switch.

Usage records are kept for 90 days and then automatically deleted. For authoritative billing figures, check your OpenAI dashboard — the plugin's estimates are based on published rates and may differ slightly from your invoice.

FAQ

The chat bubble does not appear.

Confirm Enable chatbot is ticked on the Settings tab and your Dprime subscription is active. If you use a caching plugin, purge the page cache after saving settings.

The AI says it is not configured.

Enable AI and save a valid API key on the Settings tab. Click Test connection to confirm the key is accepted. The message "The AI assistant is not configured yet" appears when the AI is disabled or the key is missing.

Products are not appearing in chat replies.

In Sync mode: click Sync now after adding new products. In Live mode: confirm the selected model supports tool calls. If the product sync limit is set too low, increase it on the Settings tab.

Can I use this on a site without WooCommerce?

No. This plugin is built specifically for WooCommerce stores. For a general-purpose AI chatbot without product search, use the Dprime AI Lead Chatbot instead.

Does this plugin store any visitor data?

No. Conversation history is held in the visitor's browser session only. No chats, no names, and no contact details are stored by the plugin.

How do I control the cost?

Set Max AI replies per session to a low number (6–10 is typical). Use gpt-4o-mini rather than gpt-4o. In Sync mode, reduce the product fields sent to the AI — excluding description and categories cuts token usage significantly. Monitor spend on the Usage tab.