Same Prompt, Different Tools: Genially

I'm back with round three of my Same Prompt, Different Tools experiment!

If you missed the first two, I'm giving different AI-powered tools the same learning problem, the same source material, and the same basic requirements to see what they create. I'm intentionally not telling them what kind of learning experience to build or what instructional strategies to use. Then I look at the instructional decisions they made, what worked (and what didn't), and how accessible the result is.

My first test was Mindsmith, an AI-powered eLearning authoring tool. Next was Claude Design, which isn't specifically designed for eLearning but produced some surprisingly good results.

This time, I'm testing Genially's AI capabilities. Genially is already an established tool for creating interactive content, so I was curious to see what its AI features could do with the same prompt and OSHA source document.

And... this one was rough.

Please also keep in mind this is just a fun little experiment to satisfy my curiosity, and to help raise awareness about the importance of accessibility testing with ANY tool, even when you prompt AI to make accessible content.

The Prompt

I gave each tool the same general prompt and source material. Keep in mind, this is skipping most of the instructional design process. This is all hypothetical so I could keep the scope fairly narrow. I found this OSHA document on Heat Illness Prevention, and decided to narrow it down to landscaping-related work environments.

Here is the exact prompt that I pasted into each tool:

"Create a 10–15 minute digital learning experience for newly hired seasonal employees at a landscaping company who will be working outdoors in hot weather.

Employees receive information about heat safety, but supervisors have found that new employees do not consistently recognize signs of heat illness or know how to respond appropriately when they or a coworker begins experiencing symptoms.

Use the provided OSHA document, Protecting Workers from the Effects of Heat, as the source material. The learning experience should prepare employees to apply this information and make appropriate decisions when they encounter potential heat illness on the job.

The learning experience must conform to WCAG 2.2 Level AA.

Use a clean, professional visual style appropriate for the audience, with a color palette inspired by the provided OSHA document."

I didn't give any more instruction unless I was specifically asked questions that I needed to answer in order for the tool to move forward.

About Genially

Genially is an online authoring tool for creating slide-based interactive presentations, eLearning materials, quizzes, games, and other digital content. Unlike Claude Design, Genially is specifically designed to support interactive learning experiences, although it isn't exclusively an eLearning tool.

Genially offers a free plan and several paid subscription options. I have the Pro plan, which I purchased during a Black Friday sale last year for $72 for the full year. You can find the current plans and pricing on Genially's website.

You can create projects using Genially's templates or start from scratch, adding interactive elements, animations, multimedia, and assessments. Genially also includes AI features that can help generate content and create projects from a prompt.

I've experimented with Genially a little before, including creating an interactive slide that I embedded in a Rise course, but I haven't completed and published a full project with it.

How Genially Handled the Prompt

Oh, boy. This was rough.

It started out okay, updating me with cute little lists of imaginary "people" and the tasks they were completing.

It then recommended the format of a presentation, which I accepted. The point of this little "experiment" is to see what the output is with little to no additional input from me after the prompt.

Then it made me choose a template. It didn't actually tell me I was choosing the template. It just said "I've chosen these styles for you. Which one do you prefer? You can edit and customize later." I wasn't sure exactly what "styles" it would be taking from the selection of the 6 options it presented. I chose one that looked like it used an accessible font and a professional layout.

It took a long... and I mean a LONG time. There was no progress bar. I couldn't tell if it had frozen. After 15 minutes with no update, I refreshed and assumed it was all gone. But later that evening, the project showed up on my dashboard. This was after I had already watched it working through the course for quite a while.

The result... was sort of a mess. I would consider this not salvageable. If we were talking about a car, it would be totaled, because the time it would take me to make this into a decent learning experience would be longer than starting from scratch on my own.

The Course

I evaluated the course using 10 different criteria. Remember, this isn't a true experiment with strict controls. I skipped most of the instructional design process, and the prompt was intentionally pretty generic. I didn't even provide learning objectives.

I'll highlight a few of my observations below, but overall, the “out-of-the-box” course scored 8/30 using my self-created criteria.

Content Selection & Performance Alignment: Genially included relevant information about heat illness, but the content was poorly organized. It seemed more concerned with filling the template than presenting information in a way that supported the performance goal. Some headings didn't even match the content underneath them, and the course consisted primarily of informational slides followed by knowledge checks rather than opportunities to practice applying what was learned.

Content Accuracy: Much of the information appeared to come from the OSHA source material, but the way it was organized sometimes made otherwise accurate information misleading. For example, one slide placed emergency response instructions under a heading about behavioral changes and neurological symptoms under a heading about skin indicators. (See slide image below).

Cognitive Load: The templates included decorative photos, icons, and other visual elements that weren't at all related to the content and were a part of the original template. (Side note: many of the photos and icons don't actually match the content in the original template either.) Combined with the mismatched headings and layouts, these added unnecessary complexity rather than helping learners understand the information.

Feedback: There wasn't feedback for the only two knowledge check questions, and the knowledge check slides automatically advanced to the next question after a set delay, without giving learners control over when to continue. Although the timing could be adjusted, I couldn't find a way to disable automatic advancement entirely.

Accessibility

Oh boy. Where do I even start?

The first problem is that the templates themselves weren't accessible before AI even got involved. The reading order was incorrect, and several of the interactions presented accessibility problems right out of the box.

Genially does allow you to adjust the reading order, but I didn't find the process particularly intuitive. The reading order panel identifies many elements simply as 'SVG,' rather than giving them meaningful names, making it difficult to tell which item you're working with. (You can change the titles, but this is how they all start). The reading order controls are located in a dropdown labeled "Accessibility," but that's the only accessibility feature available there.

The bigger problem was that I couldn't fix many of the issues I identified. I didn't have enough control over how the interactions worked to make them fully accessible to screen reader users.

For example, the knowledge checks used radio button behavior, but the answer choices were visually presented as regular buttons. For sighted keyboard users, this could be confusing because the controls look like buttons you'd navigate between using Tab, but they actually require radio button keyboard navigation using the arrow keys. I couldn't change that behavior. The questions also automatically advanced after a set delay, and although I could adjust the timing, I couldn't give learners a button to move forward when they were ready.

Other button interactions presented accessibility problems that I couldn't resolve through the editor either.

And then there was the alt text. One image of a perfectly ordinary business meeting received a description that made it sound like a scene in a horror movie. (I'll let you see that one for yourself in my video posted below). The alt text can easily be changed, but I had a good laugh over the automatic alt text created.

Unlike my experience with Claude Design, where I could describe an accessibility problem and prompt Claude to fix it, Genially left me with issues I could identify but couldn't adequately remediate.

That's a pretty significant limitation for an authoring tool, especially when some of those issues were already present in the templates.

I think the most frustrating part was that I couldn't find a way to get in touch with Genially about the issues. They have what is essentially a message board to post problems you encounter, and other users upvote problems to get attention.

Using Genially's AI Generation in Practice

I really wanted to find something positive to say about this experience, but I'm struggling.

The AI-generated course wasn't something I could use without significant revisions. The content was poorly organized, the layouts didn't fit the information, and the knowledge checks offered little opportunity for meaningful practice. I would have needed to rethink and rebuild most of the course before it would be useful to learners. It would absolutely be more efficient for me to start from scratch or a generic template.

And then there's accessibility. Even if I were willing to spend the time reorganizing the content and replacing the irrelevant visuals, I still wouldn't have enough control over some of the interactions to make them accessible. That's a much bigger problem than simply needing to edit an AI-generated first draft.

The whole point of using AI to help create a course is presumably to save some time or make the development process easier. In this case, the generation process itself was frustrating, and the result created far more work than it saved.

I wouldn't use Genially's AI course-generation feature again based on this experience. That doesn't mean Genially as a whole has no value. I've used it to create a small interactive element before, and there may be other types of projects where it works better. But for generating an accessible, instructionally sound learning experience from a prompt? This experiment didn't give me much reason to return to it.

The Bigger Accessibility Takeaway

If there's one thing this experiment reinforced, it's that even when you know how to make something accessible, an authoring tool can prevent you from doing it.

Genially's AI didn't just create accessibility problems of its own. It also used existing templates and interactions that already had accessibility issues. And when I tried to fix those problems, I discovered that the editor didn't always give me the control I needed.

That's particularly frustrating because identifying an accessibility issue is only half the battle. If an authoring tool doesn't allow you to make the necessary changes, knowing exactly what's wrong doesn't get you very far.

And here's what really bothers me about this experience: Why are we still being given inaccessible templates and interactions in tools designed to create learning experiences? Why should accessibility be something we have to fight to achieve, rather than something that's built into the tools we're using?

I'm not expecting AI to create a perfectly accessible course from a generic prompt. That's why I'm doing this experiment in the first place! But I do expect an authoring tool to give me the ability to fix the problems I find. And frankly, I don't think that's asking too much.

This isn't just a Genially problem, either. Other authoring tools provide inaccessible templates, and I think we should be asking more of all of them.

Make sure you check out the full video where you can follow along while I work through accessibility testing this course.

Watch full video of Genially testing on YouTube