This case at a glance
| Scope | Market research and analysis by our sales team for planning new services |
|---|---|
| Challenge | Researching competing services and organizing the results in a form that is easy to compare took a lot of time |
| Our approach | Developed Marketing AI, which searches the web based on keywords and generates comparison tables |
| Products | Amazon Bedrock (Claude, Bedrock Agents), AWS Lambda |
| Duration | About one month (about three people) |
| Results | Research and organizing work that took about two hours per plan was cut to about two minutes |
We developed Marketing AI to support market research and analysis when planning new services. It brings generative AI into the work, from gathering information to organizing it into comparison tables.
-
The challenge: time spent on research and organizing
Our sales team spent a lot of time researching competing services and summarizing the information. To move planning forward, they needed more than search results: the information had to be organized in a form that is easy to compare.
-
Connecting search and comparison tables
Using Claude models on Amazon Bedrock, Amazon Bedrock Agents, and AWS Lambda, we built a system that searches the web based on the keywords entered and generates a comparison table. During development, we adjusted how instructions are given, so that output that used to take several rounds of conversation can now be obtained with a single request.
-
Results we confirmed in our own work
The development introduced on the AWS official blog took about three people about one month. For the market research and organizing work it covered, a task that took about two hours per plan now takes about two minutes. The result of this development is that search and comparison table creation are connected, and information is output in a form we can use when considering plans.
