New AI applications come onto the market every day. Most of us find it difficult to keep track of everything. And if we’re not careful, we’ll have countless test accounts and paid subscriptions for AI applications in no time at all.
While AI assistants such as ChatGPT, SwissGPT or Perplexity are used almost daily by many people and are therefore not forgotten, other AI applications are only used for specific issues and therefore run the risk of being forgotten. At the same time, we often don’t even ask ourselves the honest question of whether we really need the tool or AI application and whether it will bring us the desired benefits compared to the license costs.
The following article provides a basis for solving precisely these problems and facing up to these challenges from the outset.
The AI application map for measuring the success of AI applications
The AI application map solves 2 central challenges:
1. it provides an overview of all AI applications in a company
In the AI application map, all human tasks that are to be supplemented or replaced by an AI application are listed first. These tasks are then mapped with the appropriate AI applications. Further detailed descriptions follow, which are particularly useful for any applications that are not in daily use.
2. the AI application map measures the benefits and efficiency of AI applications
In the further detailed view of the AI application map, the users of the respective AI applications enter how often they use the respective applications, the degree of automation, how high the increase in efficiency is and also how satisfied they are with the quality of the AI output. The AI application map then visualizes these results in the form of charts and graphs.
Working with the AI application map
Preparation
Before the AI application map can help to measure the increase in efficiency and quality of the individual applications so that further decisions can be made on the use of AI applications, the AI application map template must be completed carefully.
Step 1: Describe AI tasks
The first spreadsheet lists all tasks that are initially still performed by humans but are soon to be completed by an AI. These tasks are then mapped to suitable AI applications or the selected AI tools. The corresponding spreadsheet is called “Task overview”. It also records how important the respective task is for the company or the team and how often this task is carried out.
Step 2: AI applications in detail
The second table sheet “AI applications in detail” describes the individual selected AI applications and their functions in detail. This makes it possible to determine whether the company is already utilizing the full range of functions of the AI tools or whether there is further potential in the use of the AI application.
Step 3: Efficiency measurement
For many, the third table sheet is the most interesting part of the entire AI application map. This is where efficiency is measured. First, how often the application is used is entered, followed by information on quality, productivity and the degree of automation. Charts and graphics are created to visualize the results.
Step 4: Further tool information
The last sheet on “Further tool information” takes a little more time to fill in, but is very important, especially with regard to the long term. Any information on use, functions, data protection requirements, costs and other questions and answers about AI applications and AI tools are added here. This sheet is particularly important for any applications that are not used on a daily basis and it also helps employees who are new to a company to find their way around the existing tool landscape.
Utilization
It is important to understand that the AI application map is not a static list, but a tool that is regularly in motion and must be constantly supplemented and updated.
1. continuous tasks
It must be ensured that the AI application map is always up to date. If new AI applications are added, they must be listed immediately.
If AI applications are replaced or terminated, this must also be recorded in the AI application map.
It must also be ensured that the existing information on the AI applications is always up to date. In most cases, the functional scope of the individual applications changes and this must also be recorded.
2. monthly tasks
A key aspect of the AI application map is measuring the efficiency and determining the benefits of the respective AI applications. First and foremost, you need to keep a regular eye on Table 3 for measuring efficiency. As soon as it becomes apparent that individual applications are no longer used at all or only very little, or that the efficiency or quality of the applications has declined, this must be questioned. Falling figures can initially be an indication that AI applications are being used incorrectly or have simply been forgotten. If it becomes apparent at this point that individual applications are really no longer needed, they should be canceled.
Start with the AI application map
To get started with the AI application map, you can download the Excel template of the AI application map via the following paid download link. The Excel file is fully editable and not write-protected.
Support desired
If you’re not quite sure how to really use the AI application map or you’re having some difficulties getting started, feel free to contact me at any time. I am happy to help you fill out the AI application map. Sometimes a little start-up help is enough, sometimes a continuous workshop format makes sense. We can find out exactly what you need in a get-to-know-you call.
Further information on AI applications
How do I find new AI applications?
I generally recommend not actively looking for new applications, but first determining what needs, processes or tasks you actually have. The next step is to find out which of these needs can be supported by AI. And then I usually recommend asking existing software contracts or providers whether they have already integrated these applications themselves. The majority of traditional software providers now also offer functions with AI, so it is not always necessary to purchase completely new software.
I would be happy to support you in your search for the right AI application.
What license models are there for AI applications?
Most AI providers offer monthly and annual licenses. The annual licenses are usually cheaper. Nevertheless, I usually recommend taking out monthly subscriptions first and renewing them if necessary. This reduces the commitment and you don’t run the risk of having to pay for too many unused AI applications.
Do we all need new AI applications?
No, by no means. Generally speaking, you can get very far with generic AI applications such as ChatGPT, SwissGPT or Perplexity. The key to the success of these tools is usually just the right prompts. I have already saved many companies from new AI tools by working with them to develop the right prompts for ChatGPT and co.
What will happen to existing tools and software contracts?
As mentioned above, existing software providers often also offer AI applications. These functions can often be used instead of simply buying new ones.
Did you like the article?
If you liked this post or found it inspiring, I’d be delighted to receive a review on Google. You can simply click on the following button.
