ATJ Itinerary Management
current process & STAKEHOLDER NEEDS
ATJ’s itinerary management process involves many layers of communication between multiple stakeholders: travelers (also called “clients”), the ATJ sales team (“travel specialists”), ATJ support staff and third-party suppliers (overseas tour operators, airlines, hotels, restaurants, etc.) To understand how the ATJ sales team currently builds and manages itineraries, I constructed the diagram to the right.
I interviewed stakeholders throughout this process (travelers, sales staff, support staff and suppliers) to understand the rationale for the current process, what was working well and where there was room for improvement:
“It was frustrating that our ‘finalized’ itinerary from ATJ was not completely accurate, The timing of activities and meals was often different than specified in the itinerary.”
- Gina & Edward S., ATJ Travelers
“I see a lot of money going out the door because of the inefficiencies of the current itinerary system. It’s slow, requires a whole department to check for errors and results in customer service problems.”
- Donna Galland, ATJ CFO
“The sales team doesn’t like to waste time building the perfect itinerary since many leads do not convert. I see the benefit of a more efficient itinerary building process so that we train new salespeople more quickly, but I also need to keep my current team happy.”
- Eric Kareus., ATJ Sales Director
“I love working with ATJ, but I wish the itineraries they send me to confirm were easier to understand. They are so long and wordy, it is hard to know what is important.”
- Tony Nong, ATJ supplier
ATJ itinerary creation and confirmation flowchart, revealing the complexity of the itinerary management process.
PROBLEM FINDING
Visualizing the process in this way allowed me to map stakeholder frustrations to three key problem areas:
1. The sales team frequently recycles old itineraries, only updating text as necessary. (They use this method because it is less time consuming than building an itinerary from scratch.) This causes old (often outdated) information to be continually reintroduced into the database of current content and makes it easy to overlook typos, duplicate text or other details that are embedded in the previous paragraph-style text document.
2. To populate a new itinerary with content, the sales team must search the text database by keyword and then manually copy and paste the relevant information into the itinerary they are building. This is slow and cumbersome, incentivizing the reuse of previous itineraries.
3. The sales team sends long, paragraph-style itineraries, often containing out-of-date or irrelevant text, to their suppliers to confirm (and reconfirm) travel services. Many of these suppliers do not speak perfect English, thus the suppliers may not notice important services to be confirmed because they are lost amid unimportant content.
IDEATION
I held a workshop with internal ATJ staff to brainstorm solutions to the problems defined above. In some areas, all staff was in agreement on necessary changes, for example that providing a more streamlined version of itinerary for suppliers to confirm would reduce errors and confirmation times. But in other areas, such as how personalized itinerary text should be, perspectives varied.
Eventually, we were able to converge on a basic solution that everyone felt comfortable testing out. The current paragraph style itinerary would be replaced by a bullet point version. Verbiage would only change minimally, but each activity (tour, meal, transfer, etc.) would have its own bullet point, so that travelers and suppliers could easily see, and confirm, what activities would take place each day at a glance.
Example of ATJ’s old paragraph-style itinerary
Example of new bullet-style itinerary
Customer-facing PROTOTYPE & TESTING
To reconfigure the entire itinerary creation and management system around this new structure would require significant changes to ATJ’s highly-customized ERP system and internal processes. While these changes would be possible, it was important that new itinerary format be validated before additional work was invested in redesigning the back office functionality and sales’ itinerary building process.
As a test, I worked closely with the salesperson most excited about this change to manually restructure the itineraries for three of his clients (two groups of repeat travelers and one group of new travelers). The new bullet-style itineraries were sent to the travelers and used to confirm services with the suppliers for these trips. Feedback was then collected from the travelers, internal staff and suppliers. (Note: while a larger test group would have been nice, the time-consuming work of manually creating each itinerary in the new format was a limiting factor.)
Based on their post-travel feedback, travelers using the new itinerary felt more positively about the travel planning process than average (though, the small number of testers made a rigorous quantitative comparison impossible). Qualitative data suggested that these travelers found their itineraries “easy to follow” and “user friendly” (one repeat traveler specifically said that he prefered this new version to the paragraph style version he has received during previous trips). There were also no complaints from these travelers about their daily activities while traveling not matching their itinerary, suggesting that suppliers found these new itineraries easier to confirm correctly.
Internal support staff and suppliers greatly prefered the new version, finding it easier to grasp important details quickly. These initial positive findings suggested that it was worth investing the time to redesign the back office functionality and sales’ itinerary building process to accommodate the bullet-style itinerary format.
BACKOFFICE PROTOTYPING
I worked closely with ATJ’s sales team and development team to understand what steps would be necessary to create the new bullet-style itinerary using ATJ’s existing ERP system. We were able to agree on the following basic steps, which would allow for a highly standardized basic format, while also providing sufficient opportunities for salespeople to edit and customize itineraries as needed:
Split the current paragraph descriptions of activities into individuals cells.
Assign the task of maintaining standardized text for each activity to one person (encode editing permissions in DB).
Tag each individual activity cell with its destination and activity type (touring, meal, overnight, etc.).
View all activities per destination when creating itineraries. Select those desired, per day.
Functionality to order activities and add custom activities, as desired.
Insert activity text into itinerary automatically when selected.
Allow manually editing of text to customize itinerary, as necessary.
The process would be as similar as possible to the current itinerary building process used by the sales team, with the new “select an activity” functionality simply replacing the existing search/copy/paste text process.
The old format (one single cell of content per day, populated by text cut and pasted from search) would be replaced with the new structure. Sales would now begin to populate the day’s content by selecting the destination (and sub-destination) from a dropdown menu. Individual activities to take place that day would now be selected from the menu. Each item corresponds to a block of standardized text in the database.
Old itinerary building process, which relies on the salesperson searching for and copy/pasting text into the daily text box.
New, proposed itinerary building process where the salesperson checks which activities will occur that day and appropriate bulleted text is auto-populated into the daily text box.
Once selected, each activity would pull the corresponding text from the database into a layout that allows the sales team to reorder the day’s events to fit. The text for each activity, in the correct order, would then appear as a fluid daily description, which could then be manually edited, if necessary.
Salespeople would have the option to reorder daily activities before populating the daily itinerary text.
Additional manual editing capabilities would allow salespeople to further customize the itinerary or add non-standard activities (in the same way that they are added currently).
I worked closely with ATJ’s database administrator and development team to construct a working prototype of this process to be tested with the sales team. To limit complexity, we selected Bhutan as the destination to use for this prototype as it contained a fairly limited number of destinations and touring options.
BACKOFFICE TESTING
We tested the new itinerary building process by giving volunteers from the ATJ sales and sales support team a simple, predefined itinerary for Bhutan that included standard and customized activities. They were then asked to build the itinerary using our current process and the new process while I observed and timed them. While there were a few points of confusion with the new process, every tester was able to complete the test and built the itinerary using the new system more quickly than with the old system (about ~10 minutes with the new system and ~15 minutes while the old system). Based on this success, the ATJ management team voted to move forward with implementing the project.
Once the basic new itinerary building structure had been set up in ATJ’s ERP system*, the bulk of the implementation work lay in standardizing and coding the activity text per location. This work was underway when I left ATJ.
*Due to the highly competitive nature of the tour operator industry, ATJ has requested that the finalized design and structure of their itinerary building process not be shared.