Revenue Management

AI predicts Airbnb prices with 69% accuracy

Jul 31, 2019 / Airbnb
Getty Images
Getty Images

One needn’t be an oracle to forecast Airbnb listings prices - AI models fed customer reviews and rental features will do the trick.

That’s the conclusion drawn by a team of graduate students at Stanford, who leveraged machine learning and natural language processing to anticipate future Airbnb rates. To train their price-predicting system, the researchers tapped the public Airbnb data set for New York City, which included 50,221 entries with 96 features in total.

The team tested several price-predicting machine learning techniques, including linear regression, tree-based models, SVR, and neural networks. But they report that the best-performing model - Support Vector Regression (SVR) - achieved an R2 score (a measure of how well the predictions approximate the real data points) of 69% on the test set.

Download: Airbnb Price Prediction Using Machine Learning and Sentiment Analysis

Distribution

Best Western is using AI to personalize ads

Jul 31, 2019 / Best Western
Best Western
Best Western

While the online travel market is cut-throat, and brands struggle to keep people's attention, Best Western thinks it has found a way to beat the odds with the help of AI.

Using IBM Watson's advertising products, it's running a campaign that spits out personalized travel recommendations and tries to get people to click through to its site and book rooms by asking them about their travel plans.

Best Western also used AI last summer with an ad that supplied discounts and tips in response to questions like, "Can I stay at the beach?" and "Can I bring my dog?" People spent one to two minutes per session with those ads - twice as long as the average time spent with other IBM Watson ads, according to Best Western.

Revenue Management

Travel advisors book their clients away from resort fees

Shutterstock
Shutterstock

Travel advisors now have a new task: ensuring that clients know up front about any resort fees they may find on their hotel bills.

With more hotels - including many in non-resort locations - tacking on charges for amenities like gym access and housekeeping services, the need to protect clients from unwanted surprises that can add hundreds of dollars to their vacation cost is essential.

While not hopeful that the fee practice will end anytime soon, many travel advisors are steering their clients toward hotels and resorts that don’t charge resort fees. They also say the resort fee situation is one more way in which travel advisors are demonstrating their value to consumers.

Related: Hotels face growing backlash against resort fees

Digital Marketing

Travel surpasses consumer packaged goods in digital ad spending

Jul 31, 2019 / Advertising
Shutterstock
Shutterstock

CPGs will rank fifth this year in US digital ad spending at $11.12 billion. But the travel industry will overtake CPG by 2020.

Travel-related companies will spend $12.97 billion on digital ads next year, while CPG firms will spend $12.80 billion.

Mobile is a key driver of online ad spending among hotels, airlines and other travel industry services. In fact, the travel sector spends more of its digital budget on mobile than any other industry–70.1% in 2019.

Digital Marketing

The top 5 things you need to know now in hotel digital marketing: July edition

Shutterstock
Shutterstock

This month, the latest digital innovations include a new “Swipe to Visit” feature in Google image results, A/B testing tools in Microsoft Ads, and the roll-out of advertisements in Instagram’s “Explore” tab.

Google has launched a new “Swipe to Visit” feature in Google image results that will drive traffic to AMP-enabled websites. The feature displays a preview of an AMP-enabled website’s header after a user selects an image result. The user can then swipe up on the preview to load the image’s corresponding web page.

The new feature is a great way for publishers to drive awareness of editorial content through discovery in Google image results. For example, if a hotel brand has AMP-enabled editorial content, such as “Top 5 Hidden Gems in New York City,” the corresponding image for that article appears in Google image results. Users can then swipe directly from the image to view the article, providing additional exposure.