To validate the problem, we investigated through user shadowing, data analysis, and interviews. First, we needed to understand, who are Guestline's users and which challenges they are facing.

The hotel industry faces a shortage of experienced staff, leading to many new hires who need training on both the software and main tasks. These are typically tech-savvy users, who prefer straightforward software and find Guestline's PMS unintuitive and outdated. Experienced staff, used to the PMS for years, rely on routine and may resist changes.

Responsibilities vary by hotel size. Smaller hotels have less defined roles, while larger ones are more structured. Common roles include:

• Front desk: 70% of PMS users are focused on reservation management and front desk tasks
• Revenue Managers: Optimize revenue through online agencies, and analyze trends and bookings
  
• Hotel Managers: Track and evaluate hotel performance
• Housekeeping: Use the PMS to organize work
  

Guestline operates in multiple countries, each with unique challenges. Most clients (80%) are in the UK, followed by the DACH region (Germany, Austria, Switzerland) at 5%. Guestline has a strong reputation in the UK after two decades, but in newer markets like Germany, users find the software complex and stagnant. 

Finally, we were able to identify some key challenges:

• Navigation is unintuitive, with no breadcrumbs or page names, and the personalization through hotkeys hides the underlying issue of excessive complexity. Searching is overly complicated, with 50+ pages using a different search tool, and 7 different searches for reservation types. On average each user performs 19 reservation searches per day.
  
• Multitasking is problematic, i.e. checking in someone while replying to a phone call to modify a reservation. The system does not support multitasking well, as every flow starts on a different page.
• Inconsistency in design and workflows makes common tasks like searching, creating, and editing difficult.
  

The previous lack of a product design team led to some internal issues. 

• Customized features for specific hotels increased architecture complexity, resulting in over 180 subpages.
• Different development teams without a common design system created inconsistent designs, making maintenance harder.
  

To tackle the problem of navigation, we came up with some simple yet effective solutions. The most important were:

The first validation happened through the usability testing platform Maze.co.

Testers were asked to perform 3 tasks in order to assess the ease of use, the workspace set-up, and the customization:

1. Navigate to the canceled reservations report
2. Switch user account
3. Add a hotkey

Results were clear: the new navigation concept is easier

• Efficiency improved by 9%
  
• Perceived quality improved by 12%

One frequent error is not understanding that the sidebar is vertically scrollable. Follow-up interviews also show unclear differences between role-based actions and hotkeys, and difficulties in finding the reservation search. Comments also highlight the improved look but note concerns about horizontal screen space.

Searching, especially for reservations, is one of the most common tasks and involves many different pages. To improve this, we focused on two main solutions:

After validating the reservation tool through interviews and usability tests, we launched it to a small group of customers with an opt-out option, replacing the reservation ID search.

The results were impressive: less than 12% opted out, mainly because they were in a rush and didn't have time to learn the new feature. However, many soon asked to opt back in. Feature adoption jumped from 5% to 95% in three months, and user feedback was positive, leading us to gradually roll out the functionality to all customers.

The unified search, a new feature, was launched as an MVP with limited functionalities for a group of BETA testers. Their feedback helped us refine the concept and prioritize features. It was interesting to see what users were searching for and how they navigated the tool. After several iterations, the unified search was launched in September 2023 to a limited number of users with an opt-out option. The initial results were very promising, with an initial 26% opt-out rate and a 65% adoption.

After identifying usability and accessibility issues, along with customer complaints about the outdated look, we knew the direction to take: