Build complete chatbot experiences with Maxbot Core.

Maxbot is a visual chatbot builder for WordPress. It combines guided quick-reply and card-based conversations, structured data capture, reusable templates, rich content, testing, and website deployment.

The Flow Editor is the center of the product. Each block represents one conversational moment: what the user can say, what the bot sends, what data is stored, and what happens next.

• A WordPress website where you have administrator access.
• A currently supported WordPress and PHP environment.
• HTTPS for production use and for any channel integration that requires a public callback.
• The installable Maxbot plugin ZIP from the product download.
• Your CodeCanyon purchase code for authorization.
• A staging site and current database/files backup before installation or updates.

1. In WordPress, open Plugins → Add New Plugin.
2. Select Upload Plugin, choose the installable Maxbot ZIP, and select Install Now.
3. After installation finishes, select Activate Plugin.
4. Open the new Maxbot menu.
5. Enter the purchase code and complete license authorization.
6. Confirm that Agents, Topics, Entities, Templates, Projects, Users Data, and Training are available.

The WhatsApp Integration is a separate add-on. Its buyers must also install and authorize Maxbot Core.

Envato licensing note: Maxbot’s unlimited-project allowance does not mean unlimited website installations. Under Envato’s standard Regular and Extended Licenses, one purchase covers one end product; an Extended License does not provide unlimited-site use.

1. Define the chatbot’s outcome and the information it needs.
2. Create the agent identity.
3. Create one focused topic.
4. Create or review the required data entities.
5. Build the conversation in the Flow Editor.
6. Test every button, card, saved value, shared continuation, and ending.
7. Create a project, assign the agent and flow, configure its trigger and locations, then publish.
8. Review Users Data, improve the guided journey, and test again.

The agent is the identity users see in the chat. A strong identity explains its purpose without pretending to do more than the assigned flows support.

1. Open Maxbot → Agents and select Add New Agent.
2. Enter a display name and a clear occupation such as Support Assistant, Sales Assistant, Booking Assistant, or Product Guide.
3. Add a recognizable avatar and a short first-person description.
4. Select Add Agent and confirm it appears in the agents list.

A topic contains one conversation structure. Keep its scope narrow enough that a user can understand the promised outcome.

1. Open Topics and select Add New.
2. Enter a recognizable topic name.
3. Add a short description that states the conversation’s purpose.
4. Save the topic, then open its Flow Editor.

Good topic scopes include Pricing Questions, Technical Support, Product Recommendation, Contact Us, Demo Booking, and FAQ Assistant.

Answer these questions before creating blocks:

1. What should the chatbot achieve?
2. What should the user be able to do?
3. What data must be collected, validated, stored, or reused?
4. Which answers should use a quick reply button, a card, or a collected data field?
5. Is there any step where an optional typed reply is genuinely more useful than a guided choice?
6. Which branches share the same continuation and should use a join?
7. What are the successful endings and recovery paths?

Templates are prebuilt chatbot structures for common goals. They reduce the blank-page problem, demonstrate good flow patterns, and give you editable starter content.

1. Open Templates.
2. Search by use case or browse categories such as Booking, Commerce, Support, or All.
3. Use filters such as location, specialty, or author when they are available.
4. Preview a result before acquiring it.

1. Open the template preview and select Acquire or Use Template.
2. Open the acquired template and review its conversation from the first block to each ending.
3. Adapt the visible bot messages, quick reply buttons, cards, and collected fields to the experience you want to publish.
4. Keep the useful branches and remove options that do not belong in your conversation.
5. Save the flow, then use Test Flow to select every retained button and card before publishing.

Templates are useful for lead capture, contact requests, support intake, bookings, FAQs, service recommendations, commerce, real estate, restaurant, and healthcare journeys.

The Flow Editor is the visual workspace where blocks become a complete conversation tree.

A block can define how it is reached, send one or more messages, wait for user input, require a fixed choice, store an answer, show rich content, redirect, end, or record unmatched replies for training.

Use the canvas to understand the whole journey. Open a block only when you need to edit its settings.

One block should represent one conversational moment. The main block areas are:

• User Input: the quick reply button or card that opens this child block, plus optional matching rules for typed replies.
• Bot Responses: one or more messages sent when the block is reached.
• Next Step: how the conversation waits, stores, routes, joins, or ends.
• Rich Content: cards, images, GIFs, YouTube videos, and links.
• Training: unmatched-reply review when optional typed routing is used.

Block action buttons let you zoom, collapse or expand, add a child block, and delete a block. Collapse finished branches in a large flow so the active area remains readable.

Most flows use a parent-child structure. A parent asks a question; each child represents one recognized answer or selectable option.

Example: a parent asks “How can I help?” and its children represent Pricing, Support, and Book a Demo. Child blocks may continue to their own children or redirect into a shared continuation.

The User Input tab defines the choice that leads from a parent block into the current child block. For most Maxbot conversations, start with one of the two guided formats:

• Quick reply button: enter the short text displayed as a fixed choice.
• Quick reply card: add the image, title, description, and button text that help the user choose.
• Typed reply: optionally add trigger keywords and phrases when the user needs to type instead of selecting a guided choice.

Do not confuse User Input with Next Step. User Input describes how the current block is reached; Next Step describes what happens after its responses are sent.

Quick reply text becomes a fixed button displayed to the user. Selecting that button takes the conversation directly into its child block, so the route is predictable and does not depend on typed wording.

Use short, distinct labels such as Yes, No, Contact Sales, Learn More, or Book a Demo. Every visible button should lead to a useful response or next action.

Cards give the user more context before choosing. Each card can contain an image, title, short description, and button text.

Use cards for products, services, plans, categories, or recommendations where a plain text button would not give enough information. The card button leads directly to its child block.

Use trigger keywords only when a step should accept a typed reply. Add the clear words and phrases that should route the user into this child block.

Matching weight: if typed phrases for two sibling blocks overlap, the higher weight is preferred. Keep this value simple and use distinct phrases whenever possible.

The Bot Responses tab defines what Maxbot sends when the block is reached. A block can contain one or several messages.

Split a heavy paragraph into a short sequence when that improves pacing:

• “Welcome to Maxbot.”
• “I can help with pricing, support, and bookings.”
• “What would you like to do?”

Keep one idea per bubble, state the next expected action clearly, and do not repeat information already visible in the quick replies.

Reuse values captured earlier with their variable names, for example:

Thanks @name. We will contact you at @email about @preferred_service.

Variables can come from data entities, saved quick-reply selections, or raw free-text replies. Use stable names and verify the output in Test Flow when a value may be empty.

The Next Step tab controls what happens after the current block’s messages are sent. Maxbot can:

• Require the user to choose one of the available quick replies.
• Save the selected quick reply into the database.
• Wait for a reply and validate it with a data entity.
• Let child-block keywords decide the next route.
• Save the raw free-text reply.
• Join or redirect to another block.
• End the conversation.

Choose one behavior based on what the conversation needs next. Do not leave a published block without a deliberate continuation or ending.

This mode prevents free typing and requires one of the child options. Use it for menus, surveys, support categories, product pickers, qualification, and other steps where reliable selection matters more than natural-language freedom.

1. Add one child block for every supported option.
2. Set each child’s Quick Reply Text or card content.
3. On the parent block, select the option that requires a quick reply.
4. Save and test every button once.

When quick reply mode is active, Maxbot can store the chosen label or value. This is useful for analytics, qualification, personalization, exports, and later workflow decisions.

Choose a stable variable name such as selected_plan, issue_type, or preferred_service. Verify the value in Users Data and inside any later message that references it.

Use this mode to collect a structured value such as name, email, phone, company, city, country, or a custom field.

1. Write a bot message that asks for one value.
2. Open Next Step and choose to save the reply into an entity.
3. Select the correct entity.
4. Confirm its validation rule and validation message.
5. Connect the block to the next question or ending.
6. Test valid, invalid, corrected, and empty input.

Quick reply buttons and cards are the recommended choice for most branches. Use typed-reply routing only where visitors genuinely need to enter their own wording.

1. Create a parent block whose message asks an open question, such as “What can I help you with?”
2. Add one child block per supported intent,for example Pricing, Support, and Book a Demo.
3. Open each child’s User Input tab and enter realistic trigger keywords and phrases.
4. Keep sibling phrase lists distinct so each typed reply has a clear destination.
5. Return to the parent’s Next Step tab and choose the option that lets keywords decide which child block comes next.
6. Write a helpful fallback message for inputs that match no child.
7. Set a retry limit and a quick-replies prompt so repeated unmatched input converts into guided choices.
8. Optionally save the raw reply for analysis or support context.
9. Save the flow and test a direct keyword, a phrase, a spelling variation, and an unmatched message for every intent.
10. After publishing, review Training and add genuine missing expressions to the correct child.

Fallback message

The fallback is shown when no child keyword matches. It should explain what the user can do next: “I didn’t catch that. You can ask about pricing, support, or a demo.”

Retry limit

The retry limit prevents an endless loop. One or two free-text retries are usually enough before Maxbot switches to guided options.

Quick replies prompt

This prompt appears above the rescue choices after the retry limit, for example: “Please choose one of these options to continue.”

Maxbot can store raw text even when the message is used for keyword routing. This supports lead qualification, support intake, later review, personalization, and analytics.

Use a stable variable such as user_question, company_need, or request_details. Collect only text the business genuinely needs and protect it according to the site’s privacy policy.

In Next Step, choose End the conversation or jump to another block. The What happens next setting then gives you two simple choices:

• Close the discussion when this branch is complete.
• Continue from another block when the conversation should reuse an existing part of the flow.

1. Select the option to continue from another block.
2. Choose a topic from the list. Maxbot then shows the available blocks in that topic.
3. Use Expand to reveal a block’s children. Select it again to collapse them when you no longer need that part of the tree.
4. Find the destination using the First bot message column, which is only a preview that helps you distinguish similar blocks.
5. Choose the destination with the radio button in the Select column, then save the flow.
6. Test the conversation from the branch that performs the jump and confirm it continues through the intended destination without creating a loop.

End the conversation when the goal is complete, the bot has handed off the next action, or no additional input is required.

The last bot message should confirm what happened and explain any next expectation: a form was saved, a link was provided, the team will follow up, or the user can reopen the launcher later.

Entities are structured fields used to validate, store, and reuse user information. Maxbot includes protected system entities for common values and supports custom entities for business-specific fields.

1. Open Entities and review the system fields.
2. Select Add New Entity for a custom value.
3. Enter a readable label and a stable variable name without spaces.
4. Enable the appropriate validation rule and write an instructional validation message.
5. Save the entity and select it in a block’s Next Step settings.

Use built-in validation for email, phone, digits, alphabetic text, or alphanumeric values. Use a custom regular expression only when the built-in rules cannot describe the expected format.

Write the error as an instruction: “Enter a valid email such as name@example.com” is more useful than “Invalid value.”

After a value is stored, reuse it through its variable in later bot messages, joins, confirmations, and project outcomes.

Users Data contains values captured from entities, saved quick replies, and raw free-text answers. Use it to review submissions, verify mappings, follow up with leads, and inspect whether the conversation stored the intended fields.

1. Edit the Email system entity and add a useful validation message.
2. Create a custom Country entity with variable country, Alphabet validation, and a clear retry message.
3. Open a topic and ask for the user’s email.
4. Save that reply into the Email entity.
5. Add the next bot message, ask for a phone number, and save it into Phone.
6. Add another message, ask for location, and save it into Country.
7. Add a final thank-you message and reuse variables where helpful.
8. End or redirect the conversation, then save the flow.
9. Test valid and invalid values, publish the project, complete it on the website, and verify the record in Users Data.

Rich Content adds visual or actionable material to a block. Use it only when it improves understanding or gives the user a useful next action.

Confirm every media URL is accessible, optimized for mobile, and appropriate for the site’s privacy and copyright requirements.

• Use descriptive titles and button labels.
• Do not hide required instructions only inside an image or video.
• Test links and media on mobile and desktop.
• Provide a text alternative when important information is visual.

The visual canvas shows structure; Test Flow shows the real conversation experience.

• Verify bot-message order and pacing.
• Select every quick reply and card action.
• If the flow accepts typed replies, test its phrases, unmatched text, fallback, and rescue choices.
• Submit valid and invalid entity values.
• Confirm variables render and rich content loads.
• Test every join from its source and every explicit ending.
• Reset between independent scenarios.

Training surfaces replies that failed to match a child block. These records show how users actually describe their needs and where the current keyword coverage or prompt is weak.

1. Review unmatched replies regularly.
2. Decide whether the reply belongs to an existing intent, reveals a missing branch, or indicates an unclear prompt.
3. Add precise expressions to the correct child rather than making every keyword group broader.
4. Replace repeated difficult free-text steps with guided choices when appropriate.
5. Retest overlaps, fallbacks, and the updated branch before publishing again.

A project is the deployment layer that connects an agent and conversation flow to a website location or supported channel.

1. Open Projects and select Add New Project.
2. Enter a recognizable name.
3. Select the conversation flow and agent.
4. Configure the trigger and any applicable message or matching options.
5. Select the website pages or channel location.
6. Save, test in context, and enable the project.

Widget customization

• Write a short greeting that explains what the assistant can do.
• Choose text and secondary colors with sufficient contrast.
• Select a layout and placement that do not cover important page controls.
• Keep the main launcher prominent and additional channel buttons smaller.
• Show secondary labels as tooltips rather than permanently covering the page.

Triggers

Only relevant triggers should appear. If the site has only a WhatsApp project, the floating interface should show only WhatsApp. If it has the core web chat, the native Maxbot launcher should open the assigned flow.

• The intended agent, topic, flow, and project are selected.
• All quick replies, cards, validations, shared continuations, and endings have been tested.
• Any optional typed-reply routes, fallbacks, and retries have also been tested.
• Variables render correctly and Users Data receives the intended fields.
• Placeholder content, sample data, broken links, and inaccessible media have been removed.
• The project is enabled only on the intended pages or channel.
• The widget is readable and usable on desktop and mobile.
• Privacy messaging, consent, access, and retention are appropriate for the captured data.
• The site and database have a current backup.

• Keep each topic focused: one main objective is easier to build, test, and improve.
• Use guided choices when reliability matters: reserve free text for situations that need flexibility.
• Write like a chat: use short bubbles, clear prompts, and one idea at a time.
• Always provide a way forward: fallbacks should guide, retries should end, and every button should work.
• Store data intentionally: save business-relevant values instead of creating a noisy dataset.
• Review optional typed routing when used: unmatched replies reveal missing expressions or a step that would work better as guided choices.
• Test frequently: the Flow Editor is for structure; the test widget and published page reveal the real experience.

Integrations connect Maxbot flows to additional channels or external systems. Add-ons are modular products that extend the core builder without forcing every advanced feature into the base plugin.

Current add-ons may introduce another communication channel, while future directions can include broader automation, analytics, lead handling, premium content, or AI-assisted capabilities. Availability depends on the installed product and version.

• Back up important projects, flows, configuration, and captured user data before an update or removal.
• Test updates on staging and complete an existing flow before moving the update to production.
• Do not treat plugin deactivation as a data backup.
• Review the supported uninstall or cleanup setting before deleting Maxbot when data should be removed.
• Use the supported cleanup path or product support instead of deleting unknown database records manually.
• Remove or reconfigure dependent add-ons before permanently removing Maxbot Core.