The goal was to demonstrate that AI could aid in email classification, extract relevant information from emails, and generate suggested responses, including details on additional hotel availability and event impact information.
Tour Partner Group asked for STX Next's assistance during the Proof-of-Concept stage of their project. The main challenge was to integrate an AI-based solution into their existing process, which heavily relies on email communication between clients and Tour Partner Group travel specialists.
The company aspired to significantly reduce response times to client inquiries by equipping specialists with meaningful information derived from AI, enhancing both efficiency and client satisfaction.
Chat with our experts and discover what STX Next can do for you.
Considering the client's needs, we proposed and implemented a solution using a custom Machine Learning tool integrated with a Large Language Model.
The implementation followed several steps, resulting in a seamless pipeline where incoming emails were processed and response suggestions were provided to sales personnel:
Mailbox extraction & integration
We began by reading email data directly from exported mailboxes.
Emails preprocessing
Unnecessary information (add-ons, system information, etc.) was removed, and emails were filtered to retain only relevant content.
Language detection & translation
The primary language of each email was detected, and all emails were automatically translated to English.
Email classification
We classified emails to determine whether they contained specific client requests.
Information extraction
Essential information was extracted from the emails.
Busy dates & events verification
We enabled verification of whether any events might impact the booking.
Hotel allocation
The solution generated the best possible hotel allocation based on the email content and current hotel availability.
Generate response
Finally, the tool was capable of generating response suggestions based on the initial email and the extracted information.
Considering the client's needs, we proposed and implemented a solution using a custom Machine Learning tool integrated with a Large Language Model.
The implementation followed several steps, resulting in a seamless pipeline where incoming emails were processed and response suggestions were provided to sales personnel:
Mailbox extraction & integration
We began by reading email data directly from exported mailboxes.
Emails preprocessing
Unnecessary information (add-ons, system information, etc.) was removed, and emails were filtered to retain only relevant content.
Language detection & translation
The primary language of each email was detected, and all emails were automatically translated to English.
Email classification
We classified emails to determine whether they contained specific client requests.
Information extraction
Essential information was extracted from the emails.
Busy dates & events verification
We enabled verification of whether any events might impact the booking.
Hotel allocation
The solution generated the best possible hotel allocation based on the email content and current hotel availability.
Generate response
Finally, the tool was capable of generating response suggestions based on the initial email and the extracted information.
The incorporation of ML and LLM technologies provided concrete evidence that Tour Partner Group can significantly enhance their processes using AI. The new tool is expected to yield numerous benefits for Tour Partner Group, their employees, and their users:
The implemented solution aims to shorten the communication loop with clients and reduce idle time for specialists analyzing various documents to respond.
With the support of the AI solution, Tour Partner Group’s specialists are equipped with valuable information early in the process of addressing client requests.
The lessons we learned from each workflow step during the Proof-of-Concept will shape future processes, potentially saving significant time in upcoming projects.
Due to the email classification step and optimized process pipeline, salespersons were instantly ready to engage with clients. The ML solution summarized and provided relevant information immediately upon receiving a client request, significantly shortening response times. As a result, sales representatives gained hours of preparation time for each request, as the entire process was completed in just a few seconds.
Schedule a call with our experts to discover the benefits of partnering with STX Next.
Our customers love to work with us
ul. Mostowa 38
61-854 Poznań
Poland
C.11 310 Col. Santa Gertrudis Copo
97305 Merida
Mexico
77 Farringdon Road EC1M 3JU
United Kingdom
c/o Rödl RAe, Taunus Tower, Mergenthalerallee 73-75,
65760 Eschborn
Germany
City Place 1401 Lake Plaza Dr, Ste. 200
Spring, TX 77389
United States
