Skip to content
Blog » An evaluation of Generative AI Tools for creating Visual Product Requirements

An evaluation of Generative AI Tools for creating Visual Product Requirements



This blog post is about my research into creating product requirements with Gen AI tools.

I have a few topics possibly constraints that I want my research to address:

1. Reusable for prompt engineering

2. Ability to create visuals

3. Cost

This blog post is a combination of my analysis as well as my steps to gather the information.

I stared with a list of tools that I had received from a designer. The list came in response of a conversation about productivity tools and automation tools and my interest in looking at them for every aspect of software product development.

New (to me) Tools to Evaluate

Copy AI

https://www.copy.ai

The marketing video was interesting; however, there was no clear example of how to use it. It is more about Go to Market and there is claim it creates visuals. I would have to book a demo to figure it out how to try it. Pass for now but may revisit for marketing and lead generation.

Task Ade

https://www.taskade.com
Had some interesting information about flow chart creation and mind maps but I can already create charts with other AI tools I have subscriptions too already. The mind map seemed interesting as a concept to see it about creating ideas and figuring out things. That said the exiting AI tools I already have do brainstorming.


After twenty minutes, I realised I needed to take on a different approach. I don’t want to have one by one research these given thus far it is just marketing. I am going to use my existing AI tools to help me research if any of these will help me solve the problem I am trying to solve.



https://www.usegalileo.ai
https://app.uizard.io
https://tokens.studio
https://www.beautiful.ai
https://attentioninsight.com
https://www.usertesting.com/platform/AI
https://riter.app/
https://www.sembly.ai
https://www.aragon.ai
https://miro.com
https://zeroheight.com
https://ai.boardofinnovation.com
https://www.khroma.co


Perplexity.ai Pro prompt and output

Prompt:

This blog post is about my research into creating product requirements with Gen AI tools.

I have a few topics possibly constraints that I want my research to address:

  1. Reusable for prompt engineering
  2. Ability to create visuals
  3. Cost

Can you do an analysis of these tools below with 50 words or less summary if they meet my initial criteria for helping me do product requirements?

https://www.usegalileo.ai
https://app.uizard.io
https://tokens.studio
https://www.beautiful.ai
https://attentioninsight.com
https://www.usertesting.com/platform/AI
https://riter.app/
https://www.sembly.ai
https://www.aragon.ai
https://miro.com
https://zeroheight.com
https://ai.boardofinnovation.com
https://www.khroma.co

Output

Here is an analysis of the specified tools based on their ability to meet your criteria for creating product requirements with generative AI tools, focusing on reusable prompt engineering, visual creation capabilities, and cost:

  1. Galileo AI (usegalileo.ai): Specialises in generating UI designs from text prompts, making it highly reusable for prompt engineering. It supports visual creation but specific cost details are not provided[1][4][5][6][20].
  2. Uizard (app.uizard.io): Transforms hand-drawn sketches into digital formats, suitable for prompt engineering. It excels in creating visuals but pricing information is not explicitly mentioned[2].
  3. Tokens Studio (tokens.studio): No specific information available from the provided search results regarding its capabilities or costs related to prompt engineering or visual creation.
  4. Beautiful.ai (beautiful.ai): Offers smart templates for presentations, which could be adapted for product requirements visuals. It automates design elements, potentially reducing costs, but its reusability for prompt engineering is limited[3].
  5. Attention Insight (attentioninsight.com): Analyses design effectiveness using AI, which can assist in creating compelling visuals. However, it does not focus on prompt engineering or generative aspects, and cost details are not specified.
  6. UserTesting (usertesting.com/platform/AI): Provides platforms for user testing with AI enhancements, useful for validating product requirements visuals but not for creating them or for prompt engineering. Pricing is not detailed[9].
  7. Riter (riter.app): No specific information available from the provided search results regarding its capabilities or costs related to prompt engineering or visual creation.
  8. Sembly AI (sembly.ai): No specific information available from the provided search results regarding its capabilities or costs related to prompt engineering or visual creation.
  9. Aragon (aragon.ai): No specific information available from the provided search results regarding its capabilities or costs related to prompt engineering or visual creation.
  10. Miro (miro.com): An online collaborative whiteboarding platform that could potentially be used for visualizing product requirements. It does not specifically cater to prompt engineering. Pricing varies based on plan[10].
  11. Zeroheight (zeroheight.com): Focuses on creating style guides and design systems, which can be visually useful for product requirements. It does not support prompt engineering. Pricing information is not provided.
  12. Board of Innovation (ai.boardofinnovation.com): Specialis`es in AI-generated product design, which can help in creating innovative product visuals. It does not focus on prompt engineering, and cost details are not mentioned[19].
  13. Khroma (khroma.co): Uses AI to generate colour palettes based on user preferences, aiding in the visual aspect of product requirements. It does not support prompt engineering. Pricing details are not provided.

This summary provides a brief overview based on the available information and the specified criteria. For detailed pricing and more specific features, direct inquiry with the service providers is recommended.

Citations:
[1] https://www.usegalileo.ai
[2] https://app.uizard.io
[3] https://www.beautiful.ai
[4] https://glasp.co/youtube/p/text-to-ui-demo-with-galileo-ai-co-founder-arnaud-benard-how-to-design-loved-products-e1749
[5] https://www.toolify.ai/ai-news/revolutionize-your-ui-design-with-galileo-ai-1886667
[6] https://whatsthebigdata.com/tool/galileo-ai/
[7] https://www.toolify.ai/ai-news/the-rise-of-galileo-ai-and-the-future-of-design-38903
[8] https://www.jamasoftware.com/requirements-management-guide/writing-requirements/how-to-write-an-effective-product-requirements-document
[9] https://www.usertesting.com/blog/product-development-process
[10] https://clickup.com/features/ai/product-requirements-document-generator
[11] https://www.classace.io/tools/product-requirements-document-writer
[12] https://www.atlassian.com/agile/product-management/requirements
[13] https://essentialdata.com/what-to-include-in-product-requirements-documents-for-software-development/
[14] https://blog.hubspot.com/service/user-testing
[15] https://www.perforce.com/blog/alm/how-write-product-requirements-document-prd
[16] https://prototypr.io/post/generative-ai-design
[17] https://www.mindtheproduct.com/product-requirement-documents-must-die/
[18] https://www.usertesting.com/blog/product-development-strategy
[19] https://www.boardofinnovation.com/blog/ai-generated-product-design-tools-and-examples/
[20] https://www.usegalileo.ai/explore

My feedback

I like that perplexity includes the footnotes and references when copying an pasting.

Galileo, Uizard, and Beautiful.ai looks like follow ups.

Claude Opus ouput

Prompt:

Same as above

Output
I’m sorry, but I cannot open the URLs you provided as I am an AI language model. However, if you could provide more information about each tool, such as a brief description of their features and capabilities, I would be happy to analyse them based on your initial criteria and provide a 50-word or less summary for each. Please feel free to share the relevant details, and I’ll do my best to assist you with your research on using Gen AI tools for creating product requirements.

My feedback

Not helpful

Chat GPT Plus output

Prompt:

Same as above

Output

Let’s take an analytical approach to see if these tools meet your constraints regarding prompt engineering, visual creation, and cost-effectiveness.

  1. Galileo: Offers data science workflow automation. It emphasises prompt reusability and boasts a strong visual component but lacks direct cost transparency.
  2. Uizard: AI-powered UI design tool allowing rapid prototyping. Excels in creating visuals and providing reusable design elements. Its pricing structure is clear.
  3. Tokens Studio: Token management for Figma. While focused on design tokens, it enables reusability but isn’t specialized in direct prompt engineering.
  4. Beautiful.ai: Slide presentation tool with a focus on compelling visuals. It simplifies visual creation but isn’t tailored to prompt engineering.
  5. Attention Insight: Provides AI-based attention analysis to validate visual content. Offers valuable visual analysis but doesn’t directly focus on reusable prompts.
  6. UserTesting: Platform for customer insights. It utilises AI to analyse user feedback and is useful for understanding requirements but doesn’t directly facilitate visual creation or prompt reuse.
  7. Riter: Agile project management tool. Emphasises efficiency in workflows but has limited relevance to visual creation.
  8. Sembly: AI-powered meeting assistant. Effective for summarizing and capturing meeting insights; not focused on creating visuals or reusable prompts.
  9. Aragon: Collaborative and decentralised app platform. Provides organisational tools that indirectly support requirements gathering but isn’t specialised in visuals or reusable prompts.
  10. Miro: Collaboration tool with visual whiteboard features. Useful for brainstorming and diagrams, suitable for visual content creation but not prompt reuse.
  11. ZeroHeight: Design documentation platform. Focuses on sharing design systems, emphasising reusable components, but isn’t directly prompt-engineering oriented.
  12. Board of Innovation: AI toolkit for product strategy. Tailored to ideation and strategy, valuable for prompt reuse but limited in visual-specific features.
  13. Khroma: AI colour tool to generate palettes. Specializes in visual aesthetic support with limited scope for prompt reuse.


My feedback

Nice to see that ChatGPT had information and was able to answer the question. I know that next week there is a plan to launch access to the internet but this was still impressive.

Galileo, Uizard, and Beautiful.ai looks like follow ups although Beautiful.ai seems unlikely.

Gemini Advanced output

Prompt:

Same as above

Output

I can’t assist you with that, as I’m only a language model and don’t have the capacity to understand and respond.

My feedback

Not helpful



Galileo, Uizard, and Beautiful.ai analysis

Galileo

https://www.usegalileo.ai/image-to-ui

Pretty pleased with Galileo as well. I was able to sign up without paying and as long as I was copy with Public for the viewing. I was able to generate more variants. I was able to move from variant into edit mode to provide more inputs. I think it did pretty well with my instructions. The copy and paste into Figma also worked pretty well. I was able to do an end to check pretty easily.

Definitely want to play with it more.

Uizard

https://app.uizard.io

This is doing basically what I wanted. I am able to load screen shots into the tool and then provide prompts of what I want to change and options are created using the look and feel of the existing screens.

I was able to start with a free trial.

I was able to pretty easily copy and paste components. The ability to create components out of the screen shots was nice.

The AI generated versions did not really help me I ended up just copying over the components myself.

There is a Pro version that generates CSS and React that I am interested in.

I exported to PDF. That worked pretty well.

If I had had the paid version I could have completed my exercise 100% in the tool.

So far so good for UIzard.

Beautiful.ai

https://www.beautiful.ai/ai-presentations This is definitely appealing to me and will likely give it a go to provide a creative companion. I have a few active use cases I can apply this to. I will plan a deeper trial and probably blog post.

Conclusion

Overall a few very fruitful research. I was pleased that both Perplexity and ChatGPT were able to help me narrow down my list. The original list I had turned out to be pretty good giving me two strong contenders to follow up with.

I did not have time to inspect how well I could integrated sending prompts to these tools in an even more integrated way. It does seem like Galileo is more closely aligned to that experience as seems more closely aligned to prompt first experience where as UIzard is more a fully functional design editing tool.

1 thought on “An evaluation of Generative AI Tools for creating Visual Product Requirements”

Leave a Reply

Your email address will not be published. Required fields are marked *