AI Lead Response Assistant

AI Lead Response Assistant is a self-hosted workspace for organizing new inquiries and preparing faster, more professional follow-up.

The application brings prospect details, inquiry text, source information, internal notes, AI-assisted summaries, reply drafts, tone selection, and lead status into one focused workflow.

It is designed to support human sales judgment rather than replace it. Your business remains responsible for qualification, pricing, promises, availability, legal compliance, and final approval before any response is sent.

1. Create a MySQL or MariaDB database and database user.

2. Upload the application files to a dedicated folder or subdomain.

3. Copy config.example.php to config.php when included in the package.

4. Enter the database credentials, application URL, and initial administrator settings.

5. Visit install.php and complete the installer.

6. Log in using the administrator account created during setup.

7. Delete or rename install.php after installation.

8. Enable HTTPS and confirm that protected pages cannot be accessed without authorization.

Recommended environment:

- PHP 8.1 or newer
- MySQL 5.7+, MySQL 8+, or compatible MariaDB
- PDO MySQL extension
- cURL for live AI requests
- JSON and mbstring support
- HTTPS for production use
- Writable permissions for any configured cache, log, or storage directories
- A modern browser with JavaScript enabled

Demo mode can be used before live AI credentials are configured.

Demo mode allows you to explore the lead workflow without transmitting real prospect information to an external AI provider.

For live mode, open config.php and locate the AI settings. Select the supported provider, enter your API key, and choose a compatible model. Protect config.php and never place live credentials in a public repository.

Begin with fictional lead information. Confirm that summary and reply generation work, errors are handled safely, and provider usage appears in the intended account before entering real customer data.

1. Choose Add Lead.

2. Enter the prospect name and available contact details.

3. Record the inquiry source.

4. Paste or type the complete customer request.

5. Add private internal notes when helpful.

6. Save the lead record.

7. Generate and review the lead summary.

8. Choose a response tone and generate a reply draft.

9. Personalize and verify the final wording.

10. Update the lead status after contact and future follow-up.

AI output improves when the source information is clear and complete. Include the service or product requested, location when relevant, desired timing, budget information voluntarily provided, project details, important questions, previous contact, and the next action you want the prospect to take.

Do not invent missing facts. Mark unknown details as questions to confirm. Keep internal assumptions separate from information actually supplied by the lead.

When copying a contact-form submission, remove technical spam fields or unrelated tracking data that do not help prepare the response.

A useful summary should identify:

- Who the prospect is
- What they are requesting
- Important dates or urgency
- Location or service area when relevant
- Questions that need answers
- Missing qualification information
- Likely next step

Compare the summary with the original inquiry. Restore any omitted detail that affects pricing, scheduling, eligibility, safety, or customer expectations. Never treat a generated summary as a substitute for the original message.

A strong first response normally confirms receipt, acknowledges the specific request, answers what can be answered accurately, asks only the necessary follow-up questions, explains the next step, and gives the prospect a clear way to continue.

Personalize the greeting and reference the inquiry so the response does not feel generic. Remove unsupported pricing, availability, guarantees, deadlines, or claims. Keep the reply easy to scan and avoid overwhelming a new prospect with unnecessary detail.

Professional is the safest default for most business inquiries.

Friendly works well for residential services, local businesses, and conversational brands.

Short is useful when the next step is simple, but it should not omit important questions or expectations.

Detailed is better for complex requests that require explanation or preparation.

Appreciative acknowledges referrals, repeat customers, or especially thoughtful inquiries.

Sales-focused can emphasize value and next steps, but should not use false urgency, pressure, or unsupported promises.

New means the inquiry has been received but not yet handled.

Contacted means an initial response has been sent or direct contact has occurred.

Followed Up means an additional response or reminder has been completed.

Won means the lead became a customer or accepted the proposed next step according to your process.

Lost means the opportunity did not proceed.

Archived means the record is retained but no longer needs active attention.

Use statuses consistently. When version 1 does not include automated reminders, schedule follow-up through your normal calendar or business process.

Before sending a generated reply, confirm:

- The prospect name and contact details are correct.
- The response matches the original inquiry.
- Questions are relevant and not repetitive.
- Pricing, service area, scheduling, and availability are accurate.
- No unsupported guarantee, discount, deadline, or promise was added.
- Internal notes are not included in customer-facing copy.
- The tone matches your brand and the situation.
- The next step is clear.
- Sensitive or unnecessary information has been removed.
- The final reply complies with your business policies and applicable requirements.

Lead records may contain names, email addresses, phone numbers, locations, project details, and private business notes. Protect the installation with HTTPS, strong administrator credentials, secure hosting, and limited staff access.

Keep config.php and AI credentials private. Send only the information needed to generate the summary or reply, and redact unnecessary sensitive data before using a live AI provider.

Review your hosting provider, AI provider, privacy notice, retention obligations, consent practices, and internal access rules according to your business and jurisdiction.

Back up the database and customized application files before upgrades, configuration changes, or server migrations. Keep a separate record of source-code modifications.

Periodically verify PHP compatibility, database health, SSL renewal, file permissions, AI provider settings, and API usage. Review old lead records according to your retention policy and remove information that no longer serves a legitimate business purpose.

Test the complete lead workflow after updates so summaries, reply drafts, statuses, and administrator access continue to work as expected.

If installation fails, verify the database credentials, PHP version, PDO MySQL extension, application URL, and file permissions.

If live generation fails, confirm cURL is enabled, the API key is valid, the provider and model are supported, billing or credits are available, and outbound HTTPS requests are permitted.

If summaries or replies are too generic, provide more specific inquiry details, select a better tone, state the intended next step, and regenerate only after improving the source information.

If a blank page appears, inspect the PHP error log through a safe server-side logging configuration rather than displaying sensitive errors publicly.

Best results come from complete lead information, prompt human follow-up, accurate expectations, consistent statuses, careful review, and respectful customer communication.