User reviews and comments that are freely available on review platforms offer a rich source of data for analysis.
Prompt 1 (example):
I am working on defining target groups for an innovation project for public libraries in Berlin. To do this, I first need an overview of as many libraries as possible. This includes city libraries as well as university and state libraries. You can also include publicly accessible private libraries in the collection.
In addition, create an overview of review platforms where users have reviewed and commented on libraries in Berlin. Assign one or two comments and reviews from users to each library in the list. Provide me with the link to the source on the review platform.
Since AI applications such as ChatGPT can only collect limited amounts of data such as reviews, some manual work is now necessary*. Clicking on the sources for the comments should lead to different review platforms such as Reddit, Tripadvisor, GoLocal, and Yelp. It is worth supplementing the result of the first prompt with manually copied user comments by copying and pasting them into a Word document.
Now the next prompt can be placed as analysis support for the AI.
Prompt 2 (example):
I am working on defining target groups for an innovation project for public libraries in Berlin. To this end, I have compiled an overview of libraries in Berlin and quotes from users on online review platforms, which I will make available to you.
Your task is to analyze these reviews and comments to describe the needs of the users. Analyze these reviews and comments to describe the needs of the users. In the second step, you will analyze the target groups that have expressed these needs. You can also refer to studies with user analyses of public libraries. Please indicate in your results which studies you are referring to.
Objectives:
- Identify needs: Based on the comments and reviews, summarize four to five needs of library visitors. This should answer the following question: “What do visitors expect when they visit a public library?” Only refer to the data from the comments in this analysis.
- Describe target groups: Assign the respective target groups to these needs based on the comments or ratings. You can also refer to the results of user analyses of publicly accessible libraries. Please indicate the source if you have used studies.
Results:
- A detailed summary of the four to five most important needs of public library users and a description of the target groups for whom the respective need is particularly relevant. Each target group is illustrated by a representative selection of quotes from the comments.
- A detailed description of the five most important target groups for public libraries based on the needs analysis. Please prioritize them: The most important target group comes first, followed by the second most important target group, and so on. In addition, the target group description should be compared with findings from user analyses of public libraries and supplemented in terms of content.
Here is the collection of ratings and comments on libraries in Berlin on rating platforms:
(Insert the collected list from the first prompt plus manually collected ratings).
The AI application, such as Claude or ChatGPT, should now have created a structured analysis that you can use as a basis for a group discussion.
If necessary, you can have ChatGPT create a persona for each of the five prioritized target groups. This is an example of a person from the target group with a description of their preferences.
Prompt 3 (example):
Please create a representative persona for each of the five target groups mentioned.
You should now have sufficient material for your discussion on the question “Which target group(s) should we focus on when working on our design challenge?”
*There are also AI tools such as Simplescraper for this purpose. However, these tools require some time to learn. For a quick preliminary analysis, a manual collection of user comments is usually sufficient.
