Week 21: Advertising

Learning objectives

By the end of this week, you will be able to:

1. Explain how paid advertising differs from organic reach, the kind of attention a message earns for free through a platform's ordinary feed, a share, or a search.
2. Write a strong ad headline and short ad body copy with AI, built around one clear benefit and one clear action.
3. Explain what A/B testing is and use AI to generate several genuinely different ad variations to test, not just reworded synonyms.
4. Describe, in general and durable terms, the basic building blocks of how an ad platform works: an audience definition, a budget, a bid, and a performance report.
5. Read basic ad performance metrics, including impressions, clicks, conversions, and cost, in plain language without relying on a visual dashboard.
6. Explain why a specific ad platform's exact interface should be explored directly with your own screen reader rather than assumed from a fixed set of instructions.
7. Apply honesty and disclosure standards to an ad, and avoid manipulative or exploitative persuasion techniques.
8. Identify special ethical considerations that apply to advertising for disability-related products or services.

Lesson

A marketing plan built around one clear message can reach an audience in more than one way. One way is to earn attention for free by publishing something people choose to find, read, and share on their own. Another way is to pay directly for guaranteed placement in front of a chosen audience. This week focuses on that second way: advertising. Advertising is the practice of paying a media outlet or a platform for guaranteed placement of a message in front of a chosen audience, rather than earning attention for free. Advertising is not better or worse than organic marketing and social media work. It is a different tool with a different guarantee, and knowing when and how to use it well is a core skill for anyone working in marketing or communications.

From Organic Reach to Paid Placement: How Advertising Is Different

Building a marketing plan starts with one clear core message aimed at one specific audience. That same message can then travel through social media, where you write a post, add clear alt text, and hope the platform's audience, and the people who choose to share it, carry the post further on their own. That kind of reach is called organic reach: the audience found the post through the platform's ordinary feed, a follower's share, or a search, with no payment involved. Advertising works differently. It is paid placement, meaning an organization pays a platform directly, in money, to guarantee that a chosen message reaches a chosen audience a set number of times, on a schedule the organization controls.

This is the same basic distinction that separates advertising from public relations. Public relations earns coverage by giving a journalist something genuinely newsworthy, while advertising buys guaranteed space and full control over the exact wording. Advertising, this week's subject, works the same way everywhere it appears, whether that is a spot on the radio, an ad next to a search result, or a promoted post in a social media feed: a marketer pays for guaranteed reach, and gains full control over the message in exchange for some of the automatic credibility that comes from an audience choosing, entirely on its own, to pay attention.

It helps to name three ways a message reaches an audience. Owned reach travels through a channel an organization fully controls, such as its own website or its own email list. Earned reach is attention a message receives because an independent person, a journalist or a follower, chose to pass it along on their own. Paid reach is attention bought directly from a platform, guaranteed in exchange for money, which is this week's subject. A strong marketing plan usually blends all three, and advertising supplies the one guarantee the other two cannot: a specific number of people will see this exact message, on a specific schedule, because it was paid for.

The Building Block of Every Strong Ad: One Benefit, One Action

A marketing message works best when it centers on one clear idea instead of trying to say everything at once. This is the single-core-message principle. Advertising takes that principle and makes it even stricter, because an ad usually gets only a few seconds of a stranger's attention before that person scrolls, changes the channel, or clicks away. A strong ad has room for exactly one clear benefit, meaning the one specific way this product, service, or cause improves the reader's life, and exactly one clear action, meaning the one specific thing you want the reader to do next, such as visiting a website, calling a number, or signing up for something.

Trying to fit a second benefit or a second action into the same short ad usually weakens both. A reader who is asked to consider two different benefits at once often ends up unsure which one matters most, and a reader given two different actions to choose between often takes neither. Before writing a single word of ad copy, it helps to answer two short questions out loud: what is the one benefit this ad promises, and what is the one action I want the reader to take. Every sentence that follows should support one of those two answers, and nothing else.

Writing a Strong Ad Headline

A headline is the first, and sometimes only, part of an ad a reader ever sees, so it carries the heaviest weight. A strong ad headline states the one core benefit as plainly and specifically as possible, in as few words as the platform allows. Vague, generic language, such as "amazing" or "the best," rarely persuades anyone, because it could describe any product at all. A specific detail almost always outperforms a vague claim: "Cut your morning routine to ten minutes" says far more, and is far more believable, than "Save time in the morning."

AI is a strong partner for headline writing, because it can generate many options quickly once you give it the real facts: the product or service, the one benefit you have chosen, and the audience you are writing for. A useful habit is asking an AI tool for eight or ten headline options in one request, then reading them aloud using Say All, or your screen reader's continuous reading command (check your own screen reader's documentation for the exact keystroke), and picking, or combining, the two or three strongest, rather than accepting the very first suggestion the AI tool offers. As with any other AI-assisted writing task, the AI tool is fastest at generating options; a human still makes the final judgment about which one is honest, clear, and actually persuasive.

Writing Short Ad Body Copy

Ad body copy is the short block of text below or beside the headline, and it exists to do one job: support the headline's one benefit and lead the reader to the one action, in as few words as possible. Most ad platforms give body copy very little room, sometimes only a sentence or two, so every word needs to earn its place. A useful structure for short ad copy has three parts: a first sentence that restates or extends the headline's benefit in slightly more detail, an optional short sentence that adds one piece of supporting proof, such as a number or a specific detail, and a final short sentence that states the one action clearly, often called a call to action.

When asking AI to draft body copy, give it the same three things you gave it for the headline: the one benefit, the one audience, and the one action, and ask specifically for a short length, such as two sentences. AI tools left without a length limit tend to write longer than an ad actually has room for, so stating the limit directly in the request, and asking the AI tool to count its own words before giving you the final answer, produces a far more usable first draft.

A Worked Example: A Headline and Two A/B Variations

Here is a worked example for a fictional product: a talking kitchen scale designed for cooks who are blind or have low vision. The one benefit chosen for this ad is independence in the kitchen. The one action chosen is visiting a website to learn more.

Notice that Variations A and B are not the same sentence reworded. They lead with two different angles on the same underlying product: independence in one, and precise results in the other, while both still support the single overall benefit and point to the same one action. That difference between reworded synonyms and a genuinely different angle is exactly what makes a set of variations worth testing, which the next section explains.

What A/B Testing Means and Why It Matters

A/B testing is the practice of creating two or more versions of an ad, showing each version to a similar audience at the same time, and comparing which version performs better before committing your full budget to one of them. The name comes from labeling the versions Ad A and Ad B, though a test can compare more than two versions at once. A/B testing exists because even an experienced marketer cannot always predict, just by reading two headlines, which one a real audience will actually respond to better. Running a small, controlled test replaces that guess with real evidence.

A useful A/B test changes only one meaningful thing between versions at a time, or, in the case of two genuinely different angles like the worked example above, tests two different underlying ideas against each other on purpose. Testing the headline alone, while keeping the body copy, the image, and the audience identical, tells you specifically whether one headline outperforms another. Changing several things at once, a new headline and a new image and a new audience all in the same test, makes it far harder to know afterward which single change actually caused the difference in results. Because of this, it helps to decide, before a test begins, exactly what you are testing and why, the same disciplined habit that also applies to verifying any single AI-generated claim on its own.

Using AI to Generate Genuinely Different Variations

AI is well suited to A/B testing because it can produce several ad variations quickly. The habit worth building, though, is asking specifically for different angles, not different wording for the same angle. A request like "give me five different versions of this headline" often produces five sentences that all say the same thing with small synonym swaps, which is not a useful test, since a reader will likely respond to all five nearly the same way. A far more useful request names the angles directly: "write one headline focused on saving time, one focused on saving money, and one focused on peace of mind," or asks the AI tool to first list several different benefits or emotional angles the product could reasonably support, and only then write one short ad for each of those angles.

Once the AI tool returns a set of variations, read each one aloud using Say All, or your screen reader's continuous reading command, and ask a simple check question of each pair: if I covered up the product name, would a listener be able to tell these two ads apart just from the angle each one takes. If the honest answer is no, the variations are not genuinely different enough to test, and it is worth going back to the AI tool and asking for a wider range of angles before running any test with real money behind it.

How Ad Platforms Work, in General Terms

Every major ad platform, whether it places ads in a search engine, a social media feed, or somewhere else online, is built from the same small set of general building blocks. This lesson describes those building blocks in general, durable terms rather than any one platform's exact current screen, because ad platform interfaces and their specific rules change often, sometimes several times a year, and a screenshot-based instruction written today could easily be wrong by the time you read it.

An audience definition, sometimes called targeting, is the set of choices that tells the platform who should see your ad, such as a general location, an age range, or an interest area. A budget is the total amount of money you are willing to spend on an ad, usually set as a daily amount or a total campaign amount. A bid or spend setting controls how much the platform can use of that budget to compete for each individual placement, since many advertisers are often trying to reach overlapping audiences at the same time. A performance report is the platform's summary of what actually happened once the ad ran: how many people saw it, how many responded, and how much was spent.

These four building blocks, an audience, a budget, a bid or spend, and a performance report, exist in some form on every major platform, even though the exact screen layout, the exact wording, and the exact minimum spend required to start advertising differ from platform to platform and change over time. Rather than memorizing one platform's current interface from a written guide, the more durable skill is knowing these four concepts well enough to recognize them wherever they appear, and exploring the actual current screen yourself with your screen reader's heading navigation and a list of links or interactive elements (sometimes called an elements list) — check your own screen reader's documentation for the exact commands — the same exploration habit that serves you well with any new AI tool's interface.

Reading Ad Performance Metrics in Plain Language

Once an ad has run, a platform reports back a small set of numbers, called metrics, describing what happened. These numbers matter because they answer the one question every advertiser ultimately needs answered: did this ad actually work. Four metrics come up on nearly every platform, and each one has a plain-language meaning worth knowing by heart, independent of any single dashboard's visual layout.

Impressions count the number of times the ad was shown to someone, whether or not that person did anything in response. Clicks count the number of times someone actually interacted with the ad, most often by clicking or tapping it to go somewhere, such as a website. Conversions count the number of times someone who saw or clicked the ad went on to complete the actual goal you cared about, such as making a purchase, filling out a form, or signing up for something; a conversion is the metric closest to real-world success, since an impression or a click alone does not guarantee anyone took the action you actually wanted. Cost is simply the total amount spent to produce those impressions, clicks, and conversions, often reported as a total and also as an average cost per click or per conversion.

Reading these four numbers together, rather than any one in isolation, tells the real story. A high number of impressions with very few clicks often suggests the audience saw the ad but the headline or image did not interest them enough to act. A reasonable number of clicks with very few conversions often suggests the ad itself worked, but something after the click, such as a confusing website, lost the person before they completed the goal. Because none of these four numbers depends on seeing a bar chart or a colored graph, a performance report can be read perfectly well by listening to the numbers themselves in a plain table, once you know what each one means.

A Worked Example: Reading a Small Metrics Report

Continuing the talking kitchen scale example, imagine a small ad ran for one week and the platform reported these numbers: 4,000 impressions, 120 clicks, and 6 conversions, at a total cost of 60 dollars.

None of that reading required looking at a single chart. It required three simple comparisons: clicks against impressions, conversions against clicks, and cost against conversions, the same three comparisons that explain almost any ad performance report you will ever encounter, on any platform, at any point in the future.

A Reminder About Any Specific Platform's Accessibility

This lesson has deliberately avoided naming one specific ad platform's exact current screens, buttons, or menu paths, for the same honesty reason this course gives whenever it discusses any AI tool's or platform's interface: an interface described today can be different, sometimes significantly different, within months, and a fixed set of memorized steps can quietly become wrong and mislead you with confidence. If you decide to explore a specific ad platform, budget real time to explore its actual current screen reader accessibility directly, using your own screen reader's list of links or interactive elements and its heading navigation, the same exploration method described earlier in this lesson, rather than assuming any platform behaves a particular way because someone described it that way in the past, including in this lesson.

Ethical Advertising: Honesty and Avoiding Manipulation

The same ethical principles that apply to public relations and to marketing more broadly apply to advertising with extra weight, because an ad is a message you paid to guarantee people will see. The most basic rule is honesty: every factual claim in an ad, a price, a result, a comparison, a statistic, must be true and verifiable, the same fact-checking discipline that applies to any AI-generated claim or any press release. An ad that exaggerates or invents a claim to sound more persuasive is not just risky; in many places it is against consumer protection law, and it damages the very trust a marketing plan is built to earn.

Beyond plain honesty, ethical advertising also means avoiding manipulative or exploitative persuasion techniques: tactics designed to pressure someone into a decision rather than genuinely inform them. Common examples include a false sense of urgency, such as a countdown timer claiming a deal ends soon when it does not, or deliberately preying on a reader's fear, insecurity, or loneliness to push a sale rather than offering a real, honest benefit. A useful test before publishing any ad is asking honestly: would I still feel good about this ad if the reader fully understood every persuasion technique I used inside it. If the honest answer is no, the ad needs to be reworked before it goes live.

Disclosure: When Content Is Sponsored or Paid

Disclosure means clearly telling an audience when content is paid advertising rather than an independent recommendation. On social media, a paid partnership or a sponsored post must be labeled as such, honestly and visibly, rather than blended in to look like an ordinary organic post. Most platforms provide a built-in way to label sponsored content, and beyond any platform requirement, disclosure is simply an honesty practice: an audience deserves to know when they are looking at a paid message so they can weigh it accordingly, the same trust-first standard that applies to honest and ethical use of AI throughout this course.

Advertising and Disability Products or Services: A Special Responsibility

The blind and low-vision community deserves particular care whenever it is the subject of a message, and advertising is no exception. An ad for a disability-related product or service, such as an accessible screen reader tool, an assistive device, or a service aimed at blind or low-vision customers, carries extra ethical weight, because it is speaking directly to, and about, a community many learners of this course belong to.

Advertising claims about a disability product deserve particular scrutiny for three reasons. First, an exaggerated claim about what a product can do, such as promising it will fully solve a real accessibility barrier when it only partially helps, can lead someone to spend money on false hope rather than a fair, accurate expectation. Second, imagery or language in the ad should avoid pity-based or inspiration-based framing, meaning language that portrays a blind or low-vision person as an object of pity or as automatically inspiring simply for living an ordinary life, since that framing disrespects the very audience the ad is meant to serve. Third, wherever possible, an ad about a disability product or service should be reviewed by an actual member of that community before it runs, the same standard that applies to any nonprofit or advocacy communication about the disability community. As a learner in this course, you bring a genuinely informed, first-hand perspective to this exact kind of review, one that many marketing teams lack entirely, and it is a real, professional skill worth naming directly on a resume or in an interview.

Where This Fits in the Bigger Picture

Advertising completes the picture: a marketing plan defines the one message and the one audience, social media and other organic channels earn attention for that message for free, and advertising guarantees paid reach for that same message when free reach alone is not enough or not fast enough. All three depend on the same underlying discipline: write clearly, verify every claim, and respect the reader on the other end of the message.

Key terms from this week

Use this list to review the vocabulary introduced in this lesson before you start the exercises and the test.

Advertising

Paid placement of a message, guaranteeing it reaches a chosen audience a set number of times, in exchange for payment to a platform or outlet.

Organic reach

Attention a message earns for free because an audience found it through a platform's ordinary feed, a share, or a search, with no payment involved.

Call to action

The one specific action an ad asks the reader to take next, such as visiting a website or signing up for something.

A/B testing

Creating two or more versions of an ad, showing each to a similar audience, and comparing which one performs better before committing a full budget to it.

Audience definition (targeting)

The set of choices that tells an ad platform who should see a given ad.

Budget

The total amount of money an advertiser is willing to spend on an ad, often set as a daily or total campaign amount.

Impressions

The number of times an ad was shown to someone, whether or not that person responded to it.

Conversions

The number of times someone who saw or clicked an ad went on to complete the actual goal the advertiser cared about.

Disclosure

Clearly telling an audience when content is paid advertising rather than an independent recommendation.

Keyboard-only exercises

These exercises use only your keyboard. Instructions describe what each action accomplishes rather than any one screen reader's exact keystroke; check your own screen reader's documentation for the exact key combination it uses.

Exercise 1: Draft a headline and two A/B variations with AI

1. Open a plain text editor or word processor, using whatever file-management skills you already have, and create a new plain text file.
2. Choose one product, service, or cause. Type one sentence naming the one benefit you want the ad to promise and one sentence naming the one action you want the reader to take.
3. Open an AI chatbot of your choice, such as Claude, ChatGPT, or Gemini. Read or paste your benefit and action to the AI tool, and ask it to write one headline plus a two-sentence body, then ask for two more genuinely different variations built around two different angles on the same product.
4. Read all three versions in full using Say All, or your screen reader's continuous reading command (check your own screen reader's documentation for the exact keystroke).
5. For each pair of variations, ask yourself whether a listener could tell them apart by angle alone, not just by wording. If two variations sound too similar, ask the AI tool to widen the angle further, then save the three finished versions to your working files folder.

Exercise 2: Practice reading a small ad metrics report

1. Open a plain text file or a simple table in a word processor, using whatever tool you are comfortable working in.
2. Type four made-up numbers for a practice ad: impressions, clicks, conversions, and total cost, using round numbers such as 5,000 impressions, 150 clicks, 8 conversions, and 80 dollars.
3. By hand, using this week's lesson, calculate the click rate (clicks divided by impressions), the conversion rate (conversions divided by clicks), and the cost per conversion (cost divided by conversions). Type each result under its matching number.
4. Read your completed practice report back in full using Say All, or your screen reader's continuous reading command, confirming each number and its plain-language meaning out loud.
5. Write one sentence stating, in plain language, whether this practice ad looks like it performed reasonably well, based only on the numbers you calculated.

Exercise 3: Review an ad for honesty and disclosure

1. Open an AI chatbot and ask it to draft one short, deliberately flawed practice ad for a fictional product, containing one exaggerated or unverifiable claim and no urgency language, so you have realistic practice text to review.
2. Read the draft in full using Say All, or your screen reader's continuous reading command, and identify, out loud or in writing, the exaggerated or unverifiable claim in the draft.
3. Using your screen reader's or word processor's find or search command (check the relevant documentation for the exact keystroke), search the draft for urgency words such as "now" or "today" and decide whether any of them cross the line into manipulative pressure, based on this week's lesson.
4. Rewrite the flawed sentence yourself so the same benefit is stated honestly, and add one sentence clearly disclosing that the message is a paid advertisement.
5. Save both the original flawed draft and your corrected version, side by side, in your working files folder, so you can compare them later.

Portfolio project: My Ad Campaign Concept

This week's portfolio piece asks you to produce a complete, realistic ad concept ready for A/B testing, exactly the set of materials a beginning advertiser prepares before spending any real money.

1. Choose a product, service, or cause. If you have a marketing plan or core message already in mind, reusing it here is encouraged, since it lets you build directly on work you have already done.
2. Write down the one clear benefit and the one clear action this ad will focus on, before drafting anything else.
3. With AI's help, draft one strong headline plus short body copy for your main ad.
4. With AI's help, draft at least two genuinely different A/B variations of that ad, each built around a clearly different angle or benefit, not just reworded synonyms of the same sentence.
5. In plain language, define a simple target audience for this ad (who you want to see it) and a simple budget concept (roughly how much you would be willing to spend and over what period), without naming any specific platform's exact interface.
6. Write a short note, in your own words, explaining how you would measure whether this ad worked, naming the specific metrics from this week's lesson you would check, and what ethical considerations, including honesty, disclosure, and, if relevant, disability-related sensitivity, apply to this particular ad.
7. Save your headline, your A/B variations, your audience and budget note, and your measurement and ethics note together in your portfolio folder, named clearly, such as my-ad-campaign-concept.

Before you consider this project finished, check it against this completion checklist:

• I chose one product, service, or cause, and named the one clear benefit and one clear action my ad focuses on.
• I drafted one strong headline and short body copy with AI's help.
• I drafted at least two genuinely different A/B variations, each built around a different angle, not just different wording of the same idea.
• I defined a simple target audience and a simple budget concept in plain terms, without relying on any one platform's specific current interface.
• I wrote a short note naming the specific metrics I would use to measure whether the ad worked.
• I wrote a short note on the ethical considerations that apply to this specific ad, including honesty, disclosure, and disability-related sensitivity if relevant.
• I saved all of these materials together in my portfolio folder.

Weekly test

Answer key

Each answer below includes a one-sentence explanation drawn from this week's lesson.