Telegraph Travel: Hotels – Case study

Brief

Telegraph Travel is the world’s largest professional hotel reviewer. The current reviews allowed readers to book the hotel by directing the users to 3rd party booking agents.

In order to improve the experience for readers, investigate improvements in the booking path.

Process

Strategy

The hotel strategy at The Telegraph was well defined – to provide readers with quality journalism – in this context, professional reviews of the worlds best hotels. Any commercial strategy must complement the editorial strategy, and not bias it in any form.

Research

Knowledge Kanban

We commissioned a session with key stakeholders to fully understand their existing knowledge of the task and users. We wanted to know what they knew. What they didn’t know. What they thought they knew. And what they wanted to know.

From this, we were able to understand what additional insights were required. Should we talk to users to understand more? Should we investigate existing analytics? Should we improve analytics tags?

User testing

We quickly uncovered a number of questions we didn’t know the answer to. These would help us understand our customer in greater depth – and produce a product to best serve their needs. These questions included: –

• What value does the professional review give readers?
• What information do they get elsewhere?
• What process do they go through to purchase a hotel?
• Is the purchase a joint decision?
• What affects changes in behaviour?
  • Hotel location (Beach, city break, wilderness)?
  • Flight duration (domestic, short- or long-haul)?
  • Familiarity with the country?
  • Duration of the holiday?
• Which competitors sites give the best experience?

Analytics

We had a large amount of data available through existing dashboards. However, we wanted to dive deeper into understanding what our users looked at, what they booked, and how they completed the process.

• What is the existing booking path? How do they get inspiration? How do they narrow down options and then choose just one?
• Where are the pain points in the booking process?
• What lead time is there for booking? Does this vary across user types? Or a destination? Or time of year? Or holiday duration?
• Are there other factors (e.g. family) affecting booking?
• What destinations do they book? Which are purely aspirational?
• Does The Telegraph serve the correct destinations? Or the correct hotels in those destinations?

Analysis

Research

Research showed a number of factors that dramatically affected the booking of a hotel. The specific destination was a strong trigger. Driven by recommendations from others. Driven through wish/bucket lists or pure curiosity. Driven around specific events; Business meetings, friends birthdays, sports events. Driven specific location – such as visiting friends or family. Offers and deals were secondary to their desire to book a hotel – although good deals could often secure a stay in one hotel or another – or one destination over a different one. It was rarely the primary decision maker.

Research by the user was key to their final choice. The Telegraph review was key for our users, and the key draw to our site over our competitors. The review was treated as trusted – much akin to a restaurant or film review. The review was given by an expert – someone paid to review hotels for a living – and as they had no affiliation to the hotel, was trusted. In contrast, the amateur reviews from our competitors were treated as purely ‘advice’ – with some users looking at the 2* & 4* reviews only – enabling them to cut through the misleading ‘Great hotel’ or ‘Worst hotel I ever stayed in’ reviews – usually heavily weighted to the latter.

Readers wanted to know details about the hotel – it’s location, style, room types etc. – the information varying drastically between users. Some needed to have easy access, some wanted a bath. Some needed particularly facilities – such as wi-fi, or a hairdryer. They also wanted information about the hotel’s surrounding. What could they do when they got there – was the hotel served well by public transport? What were the nearby key attractions? Could the reader walk out safely from the hotel & explore the destination, maybe to visit nearby restaurants & bars?

Basically, people wanted to be reassured. Reassured that the hotel met their needs. Reassured that the destination met their needs. Reassured that they were spending their money wisely.

Analytics

We produced a number of spreadsheets, graphs & dashboards to understand the data available. As these are commercially sensitive, I am unable to show them here.

In order to fully explore trends in which destinations were popular, we plotted each hotel & destination onto a world map and visualised the data spatially.

Design

Production

Each requirement was analysed, and where possible, broken into a number of phased releases. Risk assessment allowed us to prioritise based on risk vs. impact. Requirements that had the biggest risk but the biggest impact were particularly analysed to ensure that a phased approach allowed for quick learnings.

We worked closely with the Product Design, Product & Engineering teams to support them through the production & development phases, aligning to their sprints and ceremonies. This allowed us to quickly adapt designs which proved difficult to engineer, and also to improve the User Interface designs where possible – especially simplification of enhancements which had been de-prioritised due to the additional effort.

Launch phases

Each release was tested & adapted based on insights from analytics. This allowed us to quickly adjust and hone the requirements, and prioritise those giving the biggest customer or business impact.

The phased approach also allowed us to de-prioritise requirements that had a little customer or business impact. This saved the team both time (and ultimately money). Due to users inherent nature, there is always inertia to any changes. Success was, therefore, any change that did not negatively impact our KPIs.

Evaluation

All phases, as highlighted above were carefully analysed. Further rounds of user testing proved that the design changes had been successful – our users felt more confident in purchasing hotels directly from Telegraph Travel – as the information each user required had been clearly displayed to them, in a format that was easy to digest.

Ongoing user testing & data analysis subsequently highlighted a number of additional insights. These had not been identified in previous phases and verified once data reached significance.