Our AI feedback analysis works exactly the same way an experienced data analyst would approach analyzing customer feedback. First, it breaks down everything in the feedback into problems and praise, as one piece of feedback might mention multiple problems. Afterwards, it categorizes the feedback into themes based on meaning and not keywords, as many people can complain about the same thing using different keywords. Once the feedback is categorized, you can see all the biggest pain points and praise you’re receiving. Each category comes with an executive summary, quotes, recommendations, a timeline, and the ability to filter and drill down into a problem area to learn more.

You can expect human-level accuracy from our AI analysis. We built and trained our own AI models, and our AI engineers will work closely with you if we need to apply any fine-tuning. We have full control and flexibility to tailor the AI to your specific needs. Rest assured that the quality will be great.

Yes. Today, we have customers with extremely complicated and diverse products. From technical API’s to business with a large number of diverse services.

The reason why we can handle the complexity is because we are an AI company with our own AI models and AI engineers that can fine-tune and adjust our AI to your company context and products. During our onboarding process our AI engineers would evaluate the results together with you to make sure that you are happy with the results.

However, you should not expect any problems with complexity. It’s rare that fine-tuning is required, but if it’s needed, then rest assured that we will fine-tune and solve it quickly.

Yes, you probably already have a list of categories or taxonomy you use to categorize your customer feedback. With our AI analysis, you can use your own categories and, furthermore, allow the AI to create new categories if some feedback falls outside the categories you have. You can also give our AI autonomy to create all the categories based on the feedback. But don’t worry—no matter if you prefer to use your own, our AI’s, or a combination of both, we will make sure our AI analysis is aligned with the way you prefer to categorize your feedback.

Yes, you can always see the original data that supports what the AI presents to you. The only exception is if you have chosen to delete original data from our platform, then you will have a link to where the original data lives in your internal system.

Today, the original data is always text. However, we are going to add the possibility to have videos, audio and more in case you want the original data displayed in other formats than text.

Yes, it’s nice to see all your biggest problems on a high level, but it’s even nicer to be able to drill down into the underlying issues behind a big problem. Our AI analysis allows you to dive deeper into any topic and see the feedback that supports the problem. You can see how many times the feedback was mentioned, by whom and when, get a list of customer quotes to support internal business decisions, and filter by dates, customer segments, products, services, and much more. Lastly, you are always one click away from reading the original data yourself. We know that context matters, and that’s why we built full transparency into how the AI came to the conclusions it did.

Yes, we allow you to create your own categories, products, services, and much more. This allows you to easily filter the AI analysis based on how you work internally. This is done in collaboration with you during our onboarding process.

We support more than 100 languages and have no issues analyzing feedback in different languages and unifying the insights into one common language of your choice. Today, we are conducting analyses across the world in many languages, from Finnish to Arabic. You should not expect any issues when working with feedback across languages.

Yes, we allow you to integrate with your CRM and enrich all your feedback with CRM data such as company, revenue, geography and much more. We can tie CRM data to customer feedback with a unique identifier such as an email address or similar information. That means that when we analyze a support ticket, we identify the user in the ticket in your CRM data and enrich the ticket with relevant information such as customer, revenue, geography and more. Once we enrich the analysis with CRM data you can filter and segment the feedback by the CRM data. For example, you can filter the biggest problems from support by company size or/and geography.

You can create reporting in two different ways inside Magic Feedback.

First, you can create reports the traditional way, where you put together a report with visualizations, tables, and any data that might be interesting for you to include.

Second, you can enable AI reporting that proactively creates reports based on what happens in your feedback. You only need to specify what you are interested in keeping an eye on and how you want the report to be put together. The AI will then put together a report and send it to the right people the moment it identifies something of interest.

You can schedule reports to be sent out on a recurring basis, ad hoc, or with any frequency or timing that makes sense for you. Some reports might be monthly, while more urgent reports can be sent immediately if something critical happens, such as login bugs or services not working.