The whole premise of GPT-3 is to make services more efficient without decreasing their quality. This is particularly good news for tourism and hospitality businesses which are characterised by highly complex systems. Don't be shocked when GPT-3 will plan your next trip to the Mediterranean.
May 15, 2025
"Quality is more important than quantity. One home run is much better than two doubles" – Steve Jobs
“Quantity has a quality of its own” - Napoleon Bonaparte
"I can do both well :)" - GPT-3
The introduction of GPT-3 happened at the perfect time. The industry is facing a lot of challenges, not the least of which is the struggle to maintain good customer service. Some of the times due to a lot of demand and other times due to labor shortage. Or cruelly because of both. So, the GPT-3 in this case helps companies by dealing with direct and easy questions, data analysis, offering a creative standpoint when writing content and much more. This way employees are free to deal with the complex tasks.
Traveler, the guest and the tourist prefer to stay in touch with the company throughout the whole customer journey. In the dreaming, researching and planning phases they want to get as much information to reduce uncertainty and the risk of bad decision making - which can cost a lot of money. This comes up because the services cannot be tested in advance - they are produced and consumed in a single act. Therefore, a lot of contextual content must be produced in several touching points. On the other hand, in the post-journey world, user reviews and user generated content is predominant. Here, a proper analysis can bring a lot of good insight on customer sentiments. GPT-3 does a decent job in this area.
Before getting too excited, GPT-3 still can't do everything and indeed is better not to. But it has proved to perform well on several cases.
Customer Service
GPT-3 can be trained to take into account a large number of variables. That can be (1) external - number of rooms, rate, location, distance to the airport or (2) travel preferences - preferred type of hotel or favorite season to travel. This way, the response can be more personalized - like a human. By having such knowledge, GPT-3 is capable of answering a lot of common questions but also suggest action. For example, if a traveler wants to know the cheapest hotel on 15 November, the GPT3 can point out the answer and offer them to book. The same thinking applied to refunds or room availability. Recently Amadeus has partnered with Aimendo to automate the communication of agents with travelers.
Content Creation
By content I mean blogs but also emails, seasonal offers and hotel descriptions. What is fascinating is that GPT-3 can adopt to different tones of writing and creative depth. Maybe you want the tone of voice in next email to be a little bit funnier. You can do that. I've read people that suggest having multiple tones targeted to multiple target groups. I think that is dangerous. Because it may confuse the reader and it is subject of category bias. Can't a business traveler also be a family traveler? So, just because you can doesn't mean you should do it.
Now, GPT-3 can be a life saver if you want to translate pieces of content into several languages. A hard hit to translators, isn't it? Mack Grenfell has built a tool - byword.ai - which allow you to generate high quality articles at scale. I've used it to write some of the travel articles in my blog - traveltokorca.com. These tools can also optimise texts for certain keywords or long tail keywords.
Data Analysis
First and foremost, GPT-3 can help marketers understand customer sentiment. This is powerful. Now we can extract all the reviews from third party websites - Tripadvisor, Booing.com, Expedia- and preform a simple sentiment analysis. What do customers think of us? This way we can determine if they have a positive, negative or neutral opinion. We can use this data to improve customer service in the future. If you want to know exactly how to do it, Michael Taylor - Vexpower - has built a free course around it. Secondly, GPT-3 can generate countless ad copy variations which you can test.
However, one thing should be clear. We must carefully decide how much power we delegate to GPT3 and other tools. Should it be more concerned with tactical work or is it valuable to help in strategic planning at well? If we fail to do so, is like outsourcing key strategic capabilities to an external agency.
updates going wild: ChatGPT, text davinci 003
Thanks to:
Michael Taylor
Colin Lewis
Roman Egger
Mack Grenfell