The AI Playbook is the central brain of your Enginy account. It is where you configure your company context, personal identities, AI Variables, AI Campaigns, and AI Conversation Tags — everything the artificial intelligence needs to generate accurate, personalized outreach at scale.

The AI Playbook is located in the left-hand menu of the Enginy sections panel. It contains five tabs, each serving a distinct purpose:

To begin teaching the AI about your business, navigate to Enginy > AI Playbook from the left-hand menu. You will work primarily in the first two tabs — Company Information and Personal Information — to give the AI the context it needs.

Step 1 — Enter Company Information:

Step 2 — Enter Personal Information:

Once these two tabs are complete, you have taught the AI what you sell, who you sell to, and who you are. The other tabs — AI Variables, AI Campaigns (and AI Tags) — do not provide context but instead guide how the AI should take action.

This tab provides the AI with all the relevant context about your company. It is used by both AI Variables and AI Campaigns to generate personalized, accurate output. It also sets default filters (saved profiles based on ICPs) for your searches.

Fill in the following core fields:

Click the Fill with AI button to activate a workflow that searches your company's Website and LinkedIn for all the information needed to populate the fields above. The system fills in whatever it finds automatically. You can then review and adjust any details.

Below the general details, fill in the strategic modules that train the AI. The Products & Services section includes:

Define your ICP so the AI can tailor outreach to the right audience:

Select companies so the AI can use them as examples to find similar ideal customers.

These FAQs give the AI ready-made answers it can reference during conversations and message generation.

To save or update any changes you have made, click the Update button at the bottom right of the page.

This tab holds information related to the Identities connected to your account. It gives the AI context about who is sending each message, enabling more authentic personalization.

First, select an identity from the dropdown menu. Then fill in the following identity fields:

Click the Fill with AI button to automatically populate these fields using data from the selected identity and their LinkedIn profile. Review and edit as necessary. Repeat this step for each identity you want to use.

An AI Variable is a reusable prompt-and-settings bundle that tells Enginy how to research and return a specific piece of information for a Contact or Company. Think of it as a smart column: instead of manually researching, the AI reads the attributes you provide (name, domain, job title, LinkedIn, etc.), optionally searches external sources, and returns the answer in the format you specify.

AI Variables are used to generate additional information columns in your lists. You can use ready-made templates or build your own, then run them in bulk and store results as columns on your table.

Example use cases:

You can start from default Templates to be able to configure the new AI variable even faster:

Select the Output (response type) to define how the AI returns results:

Example: An AI Variable with the instruction "Tell me if the company is either a B2B, B2C, or B2C with B2B products" and the output type set to One of will only return one of those three values. This ensures consistency, makes results easier to filter and analyze, and prevents unexpected or vague answers.

You can activate the Provide explanation option when creating or editing a variable. When enabled, the AI returns both the selected result and a short reasoning behind it, giving your team more context about why the AI chose that answer.

The Prompt field is the main instruction. This is where you tell the AI exactly what to return and the format. Use the following recommended structure:

You can personalize your prompts using variables and even AI variables inside of it. Start typing { to see the list of available (AI) variables.

Examples of effective prompts:

Avoid vague prompts like: "Tell me everything about this company." (Too broad, no structure.)

Below the prompt editor you will find the Enhance button. Clicking it lets the AI automatically polish and refine your prompt — improving its structure, wording, and references to Enginy attributes. This is especially useful when you have written a first draft and want to make sure it follows best practices before saving. You can review the enhanced version and accept or adjust it as needed.

Use the Add attributes option to select which row fields the variable can use as context. Inside your prompt, reference these fields with the {field_name} syntax. When writing in the prompt editor, you can also type { to see and select available fields.

Common placeholders include:

Contact attributes:

{firstname}, {lastname}, {job_title}, {geo_region},{last_linkedin_post}, {job_change}, {previous_positions}, {mentioned_in_news}, {latest_news}, ...

Company attributes:

{company_name}, {description}, {industry}, {iq_company_news}, {yearly_headcount_growth}, {website} ...

Use these inside prompts to make the AI answer relevant to each row.

When creating or editing an AI Variable, you can enable the Deep Research toggle to let the AI Agent browse the web and find any information you need. This is a single, comprehensive search option that covers web pages, news, company websites, and other public sources in one unified search.

When Deep Research is enabled, the AI goes beyond the data already available in your lists — it actively searches the internet, visits relevant web pages, and extracts the information your prompt requires.

The technology behind Deep Research uses Firecrawl, a specialized web-scraping tool designed to extract content from websites and convert it into a format the AI can understand.

Deep Research is optimized for batch data enrichment — answering the same question across many leads/companies consistently.

Deep Research uses more credits than regular AI Variables because:

When creating or editing an AI Variable, you can choose the AI model to use (e.g., Grok 4, GPT 5.2, Gemini 3 Pro). Each model has different strengths — see Available AI Models for a full comparison.

Each AI model has an associated cost. Beyond the model, cost can also increase based on:

Once created, AI Variables appear as columns (but you might make them visible) in your Contact or Company lists. There are several ways to run them:

All AI Variables are managed from the AI Playbook > AI Variables tab. The left panel displays all variables organized in folders, with two tabs: Contacts and Companies.

You can use AI Variables to write variations of a message template you already have. This is useful when you want parts of your message to change from contact to contact — either personalized based on data or randomized for variety.

Step 1 — Write your base message:

Step 2 — Identify the parts to personalize or randomize:

Step 3 — Define the rules for each tag:

Step 4 — Create the AI Variable:

"No Data Found" results:

When AI Variables return "No data found" or produce no results, this is often because the account’s API usage credits are exhausted. Another common reason is that the selected model isn’t strong enough to reliably complete the prompt’s objective (especially for complex extraction or reasoning tasks).

AI Campaigns manage the prompts used by the AI Agent (chatbot). When you enable the option in a campaign's settings for the AI to continue the conversation once a lead replies, the AI uses these campaign prompts to craft automatic responses.

All AI Campaign agents are listed in the left panel. Select any campaign to edit its prompt, change the AI model, and click Update to save changes.

Enginy includes pre-built AI Campaign templates such as:

You can adjust these templates or create your own from scratch.

Once your AI Campaign is configured and enabled in a campaign's settings (3rd step: AI Agent; Edit prompt to chose an AI Campaign template; Enable to activate AI auto-reply), Enginy AI takes over the conversation flow:

Monitoring conversations:

All incoming and outgoing messages from AI-powered campaigns are stored in the Inbox tab. AI replies appear automatically when a lead responds, allowing you to monitor and intervene if necessary.

AI Conversation Tags let you label conversation replies to organize and act on your sales conversations. The system includes Auto tagging, which analyzes the last message sent by a lead to automatically assign the appropriate tag. AI Tags are dynamic — they are updated as the conversation evolves.

Enginy includes essential default tags. You can choose to enable them as AI Auto Tags, so the system automatically detects and applies the most common conversation outcomes.

All tags are grouped into three categories:

On the Analytics Page, you can track a single metric per category. For example, all Positive tags can be aggregated into one main metric for easier reporting. You can also view individual tag counts for more granular insights.

Auto tagging is controlled by the toggle switch located below the Instructions text box (and also on the tag card in the main view).

Auto tagging is dynamic. Each time a new message is sent in a conversation, the AI re-evaluates which tags apply.

The AI uses a compatibility matrix that controls how tags interact:

This ensures your analytics remain accurate while reflecting the full status of each conversation.

In the center of the AI Conversation Tags module, all available tags are displayed as cards. Each card has:

The module provides two views:

A search bar is also available to filter tags by name.

The ICP Quick Filter uses the Ideal Customer Profile defined in your AI Playbook to automatically apply filters when you Search for Contacts or Companies on LinkedIn (for prospecting and list building).

When you do a Search for Contacts or a Search for Companies through LinkedIn, the system:

You can still adjust any filters manually. Use the ICP filters as a starting point, then fine-tune based on campaign goals or testing variations.

No ICP set yet?

Enginy offers a curated selection of AI models from leading providers. Each model has different strengths, and choosing the right one can make a significant difference in output quality and cost. Models marked as Recommended in the platform are the best starting point for most use cases.

Generative AI models are extremely powerful, but they only produce good results when they receive good instructions. The quality of the output depends directly on the quality of the prompt.

A good prompt is much more than a question. It is a structured instruction that:

Think of the prompt as a briefing for a professional: the better the briefing, the better the work you will get.

1. Objective first, always

Before writing your prompt, ask: What exactly do I want to achieve? What is the final output I need?

The more concrete the objective, the more actionable the response.

2. Provide relevant context

The AI does not know your organization, product, or business reality unless you explain it referencing variables ({). Include:

3. Define the AI's role

Asking "Explain this to me" is much less powerful than asking "Act as...". Examples:

This sets expectations and elevates the quality of the response.

4. Specify the output format

Explicitly indicate how you want the result:

This makes the content directly usable without reformatting.

5. Include quality criteria and constraints

Do not assume the AI understands your standards. State them clearly:

6. Provide examples (and counterexamples)

Examples significantly reduce iteration and rework:

7. Think in iterative mode

The first prompt will rarely be perfect. Work in cycles:

The AI does not get offended if you ask it to redo, correct, or restructure.

Use this base template and adapt it for each use case:

Writing:

Analysis:

Training:

Before sending your prompt, verify:

If most answers are "yes," your prompt is in the best-practice zone.

No. Default tags cannot be edited or deleted to ensure consistency across teams. However, you can duplicate them and deactivate any you do not need.

This is often caused by exhausted API credits, not a platform bug. Check your usage dashboard or notifications to see if you have hit any limits. Once credits are replenished (by purchasing more or when your billing cycle resets), the AI Variables will resume populating normally. If you have remaining credits and still see empty data, hover the '?' symbol in the output to see the provided explanation. Contact support if you need further investigation.

When enabled, Deep Research lets the AI Agent browse the web to find information relevant to your prompt. It searches public websites, reads page content, follows links, and extracts structured data. It uses more credits than a standard AI Variable, so enable it only when the information you need is not already available in your list data.

No. An AI Variable's name cannot be changed once created, as renaming would cause issues in campaigns that reference it. If you need a different name, duplicate the variable, give the copy the desired name, and delete the original.

When Auto tagging is enabled, the AI analyzes the last message in a conversation against the Instructions defined for each active tag. If the message matches the criteria of a tag, the AI applies it automatically. Tags are re-evaluated dynamically as new messages arrive, so they are added or removed as the conversation context changes.