Glossary

Comparing two variants (A and B) to determine which performs better.

Account Executive. The closer role in B2B sales.

An LLM in a loop with tools. The LLM reasons, picks actions, executes via tools, observes results, and continues until a...

Attention-Interest-Desire-Action. The oldest copywriting formula (1898).

Annual Recurring Revenue. Contracted subscription revenue on an annualized basis.

The mechanism by which transformers weight the importance of different parts of input when producing output.

Determining which marketing touchpoints get credit for a conversion.

A link from another website to yours. A key ranking signal.

Classical keyword-based information retrieval algorithm. Scores documents by term frequency and document rarity.

Monthly cash loss (gross burn) or net cash decrease (net burn).

A fictionalized representation of an ICP individual: their day, worries, decisions, language.

Customer Acquisition Cost. Total spend to acquire one customer. Includes ads, salaries, tools, commissions.

Splitting documents into smaller pieces for retrieval. Affects RAG quality significantly.

Customers or revenue lost in a period. Monthly churn = cancellations / total customers.

A pillar page + supporting pages on a topic, all interlinked to build topical authority.

Maximum number of tokens an LLM can process in a single request (input + output combined).

A desired action completed by a visitor: signup, purchase, demo request, form fill.

Percentage of visitors who complete a conversion.

The craft of writing words designed to persuade a reader to take an action.

Cost Per Acquisition. Cost for one conversion event (lead, sale, signup).

Cost Per Click. Average cost for each click on your ad.

Cost Per Mille. What you pay per 1,000 ad impressions.

Conversion Rate Optimization. The practice of improving the % of visitors who convert.

Call-To-Action. The specific next step you want the reader to take.

Click-Through Rate. Percentage of impressions that resulted in a click.

Advertising designed to elicit a measurable immediate response (buy, click, call, reply).

The first substantive sales conversation. Goal: qualify the prospect, understand their situation.

Scores (Ahrefs DR, Moz DA) estimating a site's authority based on backlink profile.

A vector representation of text where semantically similar text produces similar vectors.

A special SERP result that answers a query directly from your content, placed above normal rank-1.

Giving the LLM 2-10 examples of desired input-output pairs before the actual task.

Training an existing model on additional data to adapt it to a specific task or style.

Prediction of what will close in a period. Usually Commit, Best Case, Pipeline, Omitted categories.

A large pre-trained model that can be adapted to many tasks via prompting or fine-tuning.

The API pattern where LLMs emit structured requests to call developer-defined functions.

The conceptual journey from first touch to paying customer, stages narrowing like a funnel.

Free business listing on Google, shows in Maps and local search.

Ensuring LLM responses are anchored in specific factual context rather than the model's internal beliefs.

A promise to the buyer that reduces their risk (money-back, satisfaction, performance).

An LLM producing false information that sounds plausible. The model invents facts.

The first line of a piece of copy. The gate that decides if the reader continues.

The opening line or moment that stops scroll / earns attention.

Ideal Customer Profile. Detailed definition of the customer you most want to acquire.

Leads that come to you (via content, ads, referrals).

Testing whether a marketing channel or tactic actually causes conversions or just captures them.

A word or phrase users type into search engines.

Estimate of how hard it is to rank for a keyword. Based on competing sites' authority.

Key Performance Indicator. A metric linked to a specific business goal.

The process of attracting and capturing potential customer contact info for follow-up.

A free resource offered in exchange for contact information. The opt-in incentive.

Large Language Model. A neural network trained on massive text data that predicts the next token in a sequence, producin...

Optimization for location-based searches ('dentist near me'). Focuses on Google Business Profile, local citations, revie...

An audience the ad platform builds to resemble your existing customers based on a 'seed' list.

Lifetime Value. Total gross profit a customer generates over their relationship with you.

Ratio of Lifetime Value to Customer Acquisition Cost. Below 3:1 is fragile, 5:1+ is healthy.

Statistical modeling of total marketing spend vs total revenue to estimate each channel's contribution.

Model Context Protocol. Anthropic's open protocol for LLMs to connect to external tools and data sources.

Qualification framework. Metrics, Economic Buyer, Decision Criteria, Decision Process, Identify Pain, Champion.

Media Efficiency Ratio. Total revenue divided by total ad spend. Blended across all channels.

The text snippet shown under your page title in search results.

Monthly Recurring Revenue. ARR / 12. Useful for month-over-month growth tracking.

Net Revenue Retention. Revenue from existing customers this period ÷ their revenue last period.

What you're asking the prospect to do and what they get in return.

Objectives and Key Results. Goal-setting framework: one qualitative objective, 2-4 measurable key results.

Optimizations on the page itself: title, meta description, headings, content, internal links.

On-Target Earnings. Total compensation (base + variable) if quota is met.

Sales activity where you initiate contact with prospects (cold email, calls, LinkedIn).

Profit and Loss statement. Revenue - COGS - OpEx = Net Income.

Problem-Agitate-Solution. Classical copywriting formula. Name the problem, make it vivid, present the solution.

Problem-Amplify-Story-Transformation-Offer-Response. Expanded formula for long-form.

How long it takes to recover CAC in gross profit from a customer.

Active sales opportunities at various stages, tracked in a CRM.

The stage where a product reliably converts prospects to paying customers in a real market.

The craft of designing LLM prompts to reliably produce desired outputs.

Paid advertising aimed at new audiences who haven't engaged with the brand.

Revenue target assigned to a sales rep for a time period.

Retrieval-Augmented Generation. Fetch relevant context from a knowledge base, paste into the LLM prompt, generate an ans...

Position on the SERP. Rank 1 gets the most clicks; rank 10 gets very few.

Reasoning + Acting. The canonical agent pattern: think, act (call tool), observe result, repeat.

A second-stage model that reorders retrieved documents by relevance. Typically uses a cross-encoder.

Showing ads to people who've previously interacted with your brand (visited site, watched video, etc.).

The percentage of customers still active after a time period.

Return On Ad Spend. Revenue divided by ad spend, expressed as a multiple.

How many months a company can operate at current burn rate before running out of cash.

A long-form written piece that walks the reader from problem to solution to offer.

Quantity-based reason to act now. Limited seats, limited stock, limited capacity.

Structured data that tells search engines what your content is about (article, product, event, FAQ).

Sales Development Representative. The prospecting role in B2B sales teams.

Estimated number of monthly searches for a keyword.

Search based on meaning rather than exact keywords. Uses embeddings to match intent.

Search Engine Optimization. The practice of improving a site's visibility in organic search results.

Search Engine Results Page. The page Google shows for a query.

The instructions given to an LLM defining its role, behavior, and constraints. Appears before user messages.

Site-level optimizations: crawlability, speed, mobile-friendliness, structured data, redirects.

Unit of text processed by LLMs. Roughly 3-4 characters or 0.75 words in English.

An LLM's ability to request function/API calls and incorporate results into its reasoning.

The neural network architecture underlying modern LLMs. Uses self-attention to process sequences.

Time-based reason to act now. Cohort closing, price rising, bonus expiring.

URL parameters tagged onto links to track the source, medium, campaign in analytics.

Perceived value = (dream outcome × likelihood of achievement) / (time delay × effort).

A database optimized to store and query high-dimensional vectors by similarity, not by exact match.

Video Sales Letter. Long-form video that pitches a single offer, structured like a classic sales letter.

Asking an LLM to do a task with no examples, relying entirely on task description.