Glossary

Growth Glossary

Terms we use in Growth Marketing and Digital Marketing in General together with important Frequently asked questions from our community

hallucinations

Hearing voices or something presented as very real without existing, that's a typical hallicination

Questions

What are hallucinations in real life?

In general, “hallucinations” refer to experiences where someone perceives something that isn’t actually present—like seeing, hearing, or sensing something that isn’t real. For example, a person might hear a voice speaking when no one is around or see shapes or patterns that don’t exist.

What are hallucinations in AI?

In the context of artificial intelligence, “hallucinations” are a term often used when an AI system generates incorrect or nonsensical outputs. For instance, if you ask an AI about a historical event, and it fabricates a fact or cites a source that doesn’t exist, that’s considered a “hallucination.” It’s not that the AI is seeing things, but rather that it’s producing information that doesn’t correspond to reality.

Why does AI fabricate hallucinations?

AI systems like GPT or other large language models are trained on vast amounts of text data. During training, they learn patterns, correlations, and relationships between words and phrases. However, they don’t have direct access to factual information or a reliable way to verify their responses. Instead, they generate text based on the patterns they’ve observed in their training data.

Doesn't AI model always include a fact-checking mechanism to avoid fabricating news?

AI models, especially large language models like GPT, generally do not have a built-in fact-checking mechanism by default. They don’t inherently cross-reference a trusted database or verify their outputs against authoritative sources. Instead, they rely on the patterns and associations learned from their training data to generate responses, and they don’t have a way to confirm whether the information they produce is correct independently.

This means that while an AI might produce text that sounds plausible, it can still fabricate details, mix up facts, or present outdated or incorrect information. In some cases, developers can integrate additional fact-checking systems—like linking the AI to verified data sets or using external APIs for confirmation—but this isn’t always part of the core model.

What's the difference between reasoning and factual verification

Reasoning and factual verification are related but distinct concepts. Let's analyse them

Reasoning:

Reasoning is the process of drawing conclusions based on available information, logical relationships, or underlying principles. It involves interpreting data, identifying patterns, and applying logical steps to arrive at a conclusion. Reasoning doesn’t necessarily involve checking whether the information used is correct—it’s more about understanding relationships and making inferences.

For example, if you know that “all humans are mortal” and “Socrates is a human,” reasoning helps you conclude that “Socrates is mortal.” The logic is sound even if you never verify the fact that Socrates is human.

Factual Verification:

Factual verification, on the other hand, is about determining whether a statement is true or false by comparing it against trusted sources or evidence. This process involves cross-referencing information with reliable data, records, or authorities to confirm accuracy. It’s not just about making logical connections—it’s about ensuring that each piece of information is grounded in truth.

For instance, if someone claims that “Socrates was born in 470 BCE,” factual verification would require you to check historical records or credible sources to confirm whether that birthdate is correct.

Key Difference:

Reasoning involves logical inference and deriving conclusions from premises.

Factual verification involves checking the accuracy of the premises themselves against evidence or trusted references.

Headings

Headings structure the entire page. For your readers, for google for everyone.
They’re important for readability, but for SEO as well. Headings help Google to grasp the main topics of a long post and therefore can help in the ranking. The H1 tag is the most important heading because it’s the highest level tag that shows what a specific page is about. Search engines give this tag more weight over other headings.

Questions

What are the headings?

Headings structure the entire page. For your readers, for google for everyone.

They’re important for readability, but for SEO as well. Headings help Google to grasp the main topics of a long post and therefore can help in the ranking. The H1 tag is the most important heading because it’s the highest level tag that shows what a specific page is about. Search engines give this tag more weight over other headings.

Health Checks

Health Check will assist you in defining “Unmeasured Marketing,” meaning actions you miss that would complete your picture of your funnel and, thus, your company. For example, what marketing assets do you have that may not be included in your data? How effectively can you measure the internet activity of someone searching for a product when he gets information from a third-party site?

Questions

What is a health check?

A health check is a 360-degree audit of all your digital channels, including all marketing configurations you may have done. Examples of the health checks to be performed are:

* Analytics Goals and funnels

* Any HeatMaps you have captured

* Any Paid ads you have configured and run in the past

* Any Content you have written

* Any other marketing activities you performed

Health Check will assist you in defining “Unmeasured Marketing,” meaning actions you miss that would complete your picture of your funnel and, thus, your company. For example, what marketing assets do you have that may not be included in your data? How effectively can you measure the internet activity of someone searching for a product when he gets information from a third-party site?

How often do I need to perform one?

Performing a health check every year or when transitioning from one agency to another—or from an agency to an internal team—will give you insights into understanding our optimization options and help you identify blind spots and low-hanging fruit.

Helpful Content Algorithm

Introduced in August 2022, prior to the surge in AI-generated content, Google's "helpful content algorithm" aims to improve search result quality by prioritizing content created for users over material designed primarily for search engine optimization (SEO). This approach addresses concerns about potential penalties for AI-generated content, emphasizing that the use of AI in content creation is acceptable as long as the final product is valuable and relevant to the intended audience.

Questions

What is Google's "helpful content algorithm"?

Google's "helpful content algorithm" is designed to prioritize content that is created primarily to assist and inform users, rather than content crafted solely to perform well in search engine rankings.

Introduced in August 2022, prior to the surge in AI-generated content, Google's "helpful content algorithm" aims to improve search result quality by prioritizing content created for users over material designed primarily for search engine optimization (SEO). This approach addresses concerns about potential penalties for AI-generated content, emphasizing that the use of AI in content creation is acceptable as long as the final product is valuable and relevant to the intended audience.

Does it affects our Rankings?

Google's "helpful content algorithm" is designed to prioritize content that is genuinely useful and informative to users, rather than material created solely to perform well in search engine rankings. This means that if your content is crafted with the primary goal of assisting and engaging your audience, it is likely to be favored by this algorithm, potentially improving your search rankings. Conversely, content produced primarily for search engine optimization (SEO) purposes, without genuine value to users, may be devalued, which could negatively impact rankings as per Google Developers

It's important to note that the algorithm does not penalize the use of AI in content creation per se. As long as the final deliverable is useful and relevant to the intended audience, the method of content production—whether AI-generated or not—should not adversely affect your rankings.

Which are the criteria or guidelines it sets?

The guidelines for helpful content are separated in two (2) categories:

Category 1: Content and quality

* Does the content provide original information, reporting, research, or analysis?

* Does the content provide a substantial, complete, or comprehensive topic description?

* Does the content provide insightful analysis or interesting information that is beyond the obvious?

* If the content draws on other sources, does it avoid copying or rewriting those sources and provide substantial additional value and originality instead?

* Does the main heading or page title provide a descriptive, helpful content summary?

* Does the main heading or page title avoid exaggerating or being shocking in nature?

* Is this the sort of page you'd want to bookmark, share with a friend, or recommend?

* Would you expect to see this content in or referenced by a printed magazine, encyclopedia, or book?

* Does the content provide substantial value when compared to other pages in search results?

* Does the content have any spelling or stylistic issues?

* Is the content produced well, or does it appear sloppy or hastily produced?

* Is the content mass-produced by or outsourced to a large number of creators or spread across a large network of sites so that individual pages or sites don't get as much attention or care?

Category 2: Expertise

* Does the content present information in a way that makes you want to trust it, such as clear sourcing, evidence of the expertise involved, background about the author or the site that publishes it, such as through links to an author page or a site's About page?

* If someone researched the site producing the content, would they come away with an impression that it is well-trusted or widely-recognized as an authority on its topic?

* Is this content written or reviewed by an expert or enthusiast who demonstrably knows the topic well?

* Does the content have any easily verified factual errors?