CongyingChen
commented
6 years ago Thank you Simon,
Initially, I thought the beginning, middle and end headings should not be changes. Thanks for the suggestion, I'll change them.
As for the pricing, my main point is to use the existing data in the hotel to make the hotel have better understanding of our guests. I agree with you that when making the dynamic pricing, we definitely need to check the competitors pricing too. I'll add this to the data source.
I agree with you. In practice, although most hotel groups have policy that one account can only for one person, but individual hotel have their own way/different ways to deal with this issue. It is hard to control this issue at the hotel group level, which is the object of this data science project.
I like the idea a lot - the problem is clear, and the solutions make sense. It sounds as though there is existing data which could be put to better use, so it seems like it would be practical to implement.
I would change the beginning, middle and end headings to something more specific. Even if you just replaced them with Introduction, How Data Science Can Help, and Expected Outcomes, it would flow a bit better.
In the section that covers differential pricing, I think you mention that hotels could set the price depending upon what the customer could afford, but would you also need to look at what competitors are charging too? I'm just thinking that you may set a price that is high as possible while still ensuring that the customer would book, but that could be undermined if other similar hotels charge a lot less.
In the final section, you mention the problem of multiple users for a single loyalty program - that is a great point that hadn't occurred to me. I think you could move that up to the problem statement at the beginning (i.e. hotels should use more data to maximize returns from their loyalty programs, while ensuring that only genuinely loyal members take advantage of the benefits). Maybe you would be able to use the cameras to check that the person matches the actual member. I guess it depends how big a problem it is, and whether hotels are OK with account sharing, because it means they still get bookings.
Finally I liked the idea of matching people to services, since that is presumably very low cost, e.g. if the customer likes a certain type of pillow, or distance from the elevators - hotels can accommodate that easily so long as they know about them