How Hotel Reservations Work | Hacker News Reader

How Hotel Reservations Work

thecodist.com

172 pointsFrisette45 comments

Hotel yield management is about a decade or so behind the airlines, but they're catching up, so this is probably going to get a bit harder from the OTA's/customer's perspectives before it gets easier. (Yield management means the art/science of room pricing such that you sell out nearly all of your inventory while maximizing profits. It has gotten radically more sophisticated over the years -- loyalty programs, for example, are only about 30 years old, and they virtually transformed the industry. (They're patterned off of airline mileage programs, which debuted in 1979.)

This is, by the way, one of those boring fields where people who look like programmers, talk like programmers, and act like programmers take home bonuses like bank presidents. (Interview question: "To within an order of magnitude, how much money does it make Marriott if we increase systemwide occupancy rates by 5 basis points this year?")

Why on earth would you want to sell out all rooms as a goal?

There's an implicit "for the best price each room could possibly bring" attached to that goal. The reasoning is basically that the cost of a hotel room (construction, upkeep, etc) is not affected too much by whether it is occupied or not. (In that way it's very similar to a seat on an airliner.) So considering your costs as a hotel operator are nearly fixed, you're quite obviously better off selling 50 rooms at $200 and then 50 more rooms at $100, than if you had kept the price at $200 and only been able to sell the first 50 rooms. Of course in the real world the pricing strategies are a lot more complicated than that and depend on having good models for how many people are willing to pay what price, and when.

I must be missing something here. How does an empty room make any money?

In many occupancy-based price optimization algorithms (at least in the multi-family sector where I previously worked), it is not optimal to maintain 100% occupancy. It doesn't allow much flexibility. I've seen recommendations for apartments around 95% to cope with fee changing, turn over maintenance etc. I'm sure hotels are in a similar situation (but have even lower percentage due to guest volatility).

I first realized this concept on much simpler level. In college, my older friend and I were slowing waking up after a very successful party. There was one beer left. And I said, "That would perfect if we had zero beer left". He replied, "No, one beer is perfect. It means everyone got enough. If there were zero then that means someone could have tried to get a beer after we ran out. With one left, we know that everyone got exactly as much as they wanted."

If a guest knows they can get a better rate they will force you to drive down rates longterm. Also if you sell out, that means you could have charged more and made more in net

I've spent a lot of time researching the economics of information and one of the interesting things that popped out of that was when I looked at yield management, as a research topic, I realized that that is what information markets would look like when mature.

yield management is the exercise you do when your - the individual hotel's - distribution system is set up perfectly and humming along. The biggest gains for hotels happen actually when they set their mind to it that they want to shift the source of their business away from the OTAs to more direct business: direct can be the online booking tool on their own hotel website, mails, repeat customers, phone calls. yes even phone calls, that should be one of the goals when setting up a mobile hotel website.

To achieve this hotels should invest in a decent website, start blogging with a voice that they cultivate over time and go for better rankings in the SERPs. That, plus a humming hotel operation with systems in place to handle the direct business is the way to go.

This way even smallish properties can run circles arround bigger competitors. The bigger hotels have a hard time to copy this, as they are not as agile, and need meetings to figure out a "social media strategy".

Figure I know quite a bit on how to make hotels way more independent of OTAs as a booking source: this is actually the core of my SaaS offering: http://www.igumbi.com . It is an online pms, revenue/ yield management system and has a online booking tool for the hotel website. JSONP based and no sucky flash or iframe.

The most important aspect of the yield management architecture I find to be the use of nesting. It's a very flexible, value based approach on assigning fixed capacties to sellable allotments. This robust inventory algorithm, devised by MIT's Belobaba, is the thing that executes the yiled management settings. And I totally agree with the use of loyalty programs and direct chstomer interaction. That's why we also have a mailchimp list integration, so it becomes virtually a no brainer to keep your mailing list up to date in he daily business operations

who is your typical customer?

Small independent Hotels, Apartments & Fincas. Mostly in Austria, Germany, recently Spain and Switzerland. Properties between 1 and 30 units. Typically they are looking for a management approach that helps them to handle the reservations remotely. 1-4 Staff are dealing with the reservations.

Just a heads up, you and your clients are taking credit card information over unencrypted HTTP.

Here are two examples of clients that are accepting credit card info over plain HTTP: http://vox-hotels.com/ http://www.sonnseit.at/

Hey thanx for pointing that out. The booking tool is actually tied in via https and the transmission back to the system happens via https. I need to pester my clients to set up the cert for their domain. I will also the insert a another step in the dialogue - to collect that info.

I own a small hotel in Europe and we are looking for a PMS. Does your software sync the availabilities with the major booking sites? Is your software in English?

Yes the software is fully translated into English (The marketing site not yet). Syncing with booking sites works via an Austrian channel manager. I have two other channel manager tools in the pipeline, will be out later this year.

Is that interview question supposed to be for programmers or account managers? Is there a difference?

I can't imagine how answering that question correctly indicates a good candidate for software engineer.

It's a joke, patterned off the famous "Estimate how many golf balls fit in a schoolbus" interview questions that are just windows into how candidates would think about such things. (They're terrible questions for engineers. Don't ask them.)

The dual purpose of the joke is that, if you played it straight, you'd probably come up with a number in or near the millions, which I am suggesting is a reasonably achievable result from comparatively trivial optimization work applied at scale. (That's sort of a thing for me.)

This is a pretty good read. I REALLY hate booking both air travel and hotel rooms as I ALWAYS get the feeling that I'm getting fucked. It always just feels like I know someone is getting the same room on the same night or the same flight for cheaper than me. I never feel like I've won, so to speak. When booking air travel, if I can, I like to book with Southwest just because it feels better. I may be able to get a cheaper flight with some other airline by going through Priceline et. al., but I feel dirty and cheated (somehow) doing it. I just want to deal directly with the person who has the room or the flight and know that I'm getting a fair price. Is that too much to ask for?

You would rather everyone at your hotel pay the same price, which is higher than most people at the hotel next door?

Human nature is incompatible with a rational sense of fairness. Everybody could start off happy with the price they paid, but upon discovering other people are also happy, will become unhappy.

Most people would rather this, yes.

You have to understand that people are, despite economics, not rational maximizers.

Most people would rather have a fair deal than a better deal - for example, if you offer to split "found money": split $100 with someone 90/10, they will probably turn it down, whereas if you offer to split $10 50/50, they'll probably say yes.

When booking flights (and probably hotels) use incognito mode. I was booking a flight direct with united.com the other day - found the flight at ~$400 and was entering information. Got sidetracked and the booking session timed out so I started again. New price for the exact same thing: ~$700. Incognito mode (or just clearing cookies as I've heard of this happening but never saw it myself) and it was back to the old price.

So just watch out when booking direct as well.

Get a govt id or something that looks legit (most clerks don't want to piss you off during check in). Almost all of the big chains set their lowest rates to match CONUS rates or a certain amount over it and its almost always their best rate. I don't understand feeling fucked though, if you feel bad / can't afford an extra 10-20% over their lowest possible rate you should just stay at a discount chain. Some of them are a pretty good value, like the MicroTel chain

The article misses the fact that there is a roll over cost associated with room occupancy. So booking all rooms no matter the price can actually cost the property money. One has to factor in electricity use, housekeeping costs and complimentaries (as well as some other more hidden items like prorated taxes) on a per room basis and then tally a minimum rate for profit, 100% occupancy at any rate can actually be less profitable than 80% at a staggered rate with all %80 being profitable.

Also there's a consumer-perception / branding effect going on with too-cheap pricing too. If I know I can call the local Four Seasons at the last minute and get a room for $75 then I'm going to be much less willing to pay $500 the next time - even if I "know" I got a special deal ("Well can't you make me a special deal again?").

That's where services like HotelTonight make the properties very happy - they fill the rooms at deep discounts but HotelTonight takes the "brand hit", not the hotel itself. The Four Seasons can plausibly say "Well sir we didn't give you the $75 rate - that other service did" even if they full well approved it :)

Its not only Hotel Tonight that does that, even priceline's negotiator, hotwire's hot deals and travelocity's last minute discounts. It wasn't something new that HotelTonight tried, but by creating an app for a narrow vertical 'discount rooms for tonight', they attracted lot of people's attention.

The problem with HT is that it can't do enough volume to actually build a profitable business. It's primary customer is someone bigger to buy them out.

Four Seasons selling blocks of rooms for pennies on the dollar to HT doesn't hurt their brand? Do you think HT is paying $500 for each of those rooms?

The trade-off is supposed to be that you can't always get a last-minute room at the Four Seasons when you book it for "tonight" -- sometimes they'll be sufficiently full that they won't have sold that block to HotelTonight, and instead of having a plush room right across the street from where you'll be working early the next morning, you'll need to settle for a "motel" right on the highway, 10 miles away.

I'm not sure how these things end up playing out (and if HotelTonight shows data that maps how often selected hotels actually have rooms available, that might undercut the hotels' stance a bit), but I know I wouldn't want to rely on it; even planning in advance, in my experience it's not terribly rare that hotels are booked up if you don't call early enough (and it's an in-demand area).

Edit: ah, yup -- that's HT's pitch to the hotels. From the section for "hotel partners": The key to targeting these unique and valuable guests is the IMPULSE DEAL, a rate so you good you only use it when you absolutely need to move rooms at the last second. Since you don’t have that need everyday, our display changes everyday, meaning guests can never predict when your hotel will be offered.

That is a big draw. Most revenue comes through loyalty programs and most properties are terrified if they give to much to OTAs their numbers will drop and they'll be fired. Plus, people who book online are usually price sensitive and indifferent to the actual service and amenities - their loyalty comes from price not past experience. To make matters worse, companies like expedia/hotels.com have HORRENDOUS customer service which makes better brands even more hesitant to help them out, because its trouble they can't control. The vast majority of GMs are promoted from within, which means working class and under educated. They're not going to take chances because jobs are hard to find and not one of them wants to go back to housekeeping etc

I thought this too at first, but my sister used to work in hotels, this is by and large wrong. Almost any price is better than none. Especially if you're out in the sticks and they'll buy dinner.

Think about it a little more. Electricity & heat will mostly be on anyway. Complementaries are bought in massive bulk and are almost worthless. They'd have the same number of staff on if the hotel was 80% or 100% anyway. Depending on the size of the hotel maybe you might have to hire one more minimum wage worker.

Hiring the room for a fairly nominal amount would cover all those costs. Upsell them one product, such as a meal, and you've got profit.

You have to remember the majority of the costs are fixed when calculating the cost per guest in a hotel.

The article does not mention the potential impact of Roomkey:

http://www.roomkey.com/

Based on the articles that the CTO of Roomkey has written, I am under the impression that their technology is very cutting edge:

http://www.colinsteele.org/post/27929539434/60-000-growth-in...

http://www.colinsteele.org/post/23103789647/against-the-grai...

It is important to note that the big hotel chains had realized that they had fallen behind, in terms of technology, and they needed to catchup. Unable or unwilling to build the technology themselves, they acquired Roomkey (or rather, they acquired Hotelicopter and renamed it Roomkey).

The biggest downside to a system like RoomKey's is that they only carry rates directly bookable through the hotel. This makes sense from a consumer perspective because you get some benefits from booking directly such as 1) pay at hotel option vs. prepay 2) earn your loyalty points 3) priority when receiving upgrades since the hotel didn't have to pay a commission to an OTA.

On the downside, you have to believe in the "best rate guarantee" and at our startup we simply don't see that playing out. When we go out and search rates across all of the OTAs and secondary channels (distributors, wholesalers, etc.), we find a significant amount of price disparity in the market so those who don't shop multiple sources will invariably pay more for their hotel room.

I realise this isn't quite your area but I'd really love to know how far in practice hotel managers take the idea that "a room with guests paying anything is better than an empty room". Do they typically offer significant discounts to bring in last minute guests? How does this works in practice?

I suspect there's a whole lot more to it and it's probably quite a delicate balance between capacity maximisation and less tangible qualities like the brand reputation and what effect any visibility of discounts can have on normal full-price paying customers.

I believe this is (supposedly) the idea behind hotwire. That the hotels want to fill their beds by offering them at a discount but don't want their regular customers to take advantage of it - so they disguise their identity. Anyone know any more? (or why it doesn't seem much cheaper than other OTAs that name the hotels?)

Thanks for the insights!

You nailed most of it. For a chain, the GMs boss is going to be looking at two things: revenue per available room, and average daily rate. These metrics control the rate, within the constraints of the brand standards (so for example a mid tier place like a Holiday Inn isn't going to charge less than a Choice property like a Sleep Inn because the brand managers don't want the brand diluted)

Sounds like this would be a great market for a company like Hipmunk, Adioso, or Square to develop an app targeted at indie hotels.

If I'm Hipmunk and I'm running into the sort of issues described in this post (people w/o computers using fax machines to manage reservations and making me look bad when I misquote a customer because of it), I build a best of breed tablet app that electronically keeps track of all the complexity the little guys use paper for. It syncs with my cloud so I always have the most current information, and it integrates with my competitors too to make sure that no hotel has any reason to keep me from having the most current information.

Then, I start a rollout program where I'll send your hotel a free tablet /w the app preinstalled in exchange for $500 worth of booking credits at your location. I target the hotels in the most popular areas that currently give me the least accurate data.

Sounds like a plan!

Remember that the context for many of the hoteliers you're talking about (old paper systems) is very different to the solution you're proposing (easier electronic systems and more profit).

Many of these hoteliers are running a lifestyle, not financial or business, context. Invest in new systems (and learning new systems)? Not worth the hassle. I even stayed with one hotel (7 rooms) - their problem with OTAs was being sent too many customers. They rarely wanted to be full, and they also hated 1 occupancy nights.

It would be nice to know which systems just hand over your creditcard information to the hotel owner (plus even the 3-number validation number), such as booking.com, and which do not, such as hostelbookers.com.

Booking.com is the biggest that does, all the Big OATs don't pass on your info.

400k hotels * 200 rooms/hotel average = 80M hotel rooms in the U.S. That's one for every ~4 people in the U.S. That seems way too high... even if these rooms are only filled 25% of the time (which the article indicates is way below the 70% these businesses desire), that would mean the average American would be using a hotel room one out of every 16 days (and that assumes that people don't share rooms). Am I missing something, or are these stats wrong?

You're not counting the rest of the world that may be traveling in the US =)

I think the article is substantially overestimating the average hotel size. I'd bet it's much closer to 10-20 rooms than 200. You're forgetting a huge number of smaller roadside / rural establishments.

For San Francisco (a large tourist-oriented city with many larger hotels), there are 215 hotels, offering 33,642 rooms (source: http://www.sanfrancisco.travel/research/) or an average of 156 rooms per hotel. Given that smaller cities and rural resort areas will tend to have smaller facilities, I see the rooms/hotel being markedly lower.

SF's population-to-rooms ratio is 24:1, or rooms for about 4% of the population (and that's in a city with huge tourist and business travel traffic).

That would give you a high-end of 8 million rooms, or enough for about 5% of the population (assuming double occupancy, 2.5% for single).

You think hotels are hard? We are building software for small local tour providers to help them manage their availability and get their tours online next to hotels and flights.

It sounds like we are doing the right things in terms of implementation (providing api endpoints for bulk data and more fine grained api endpoints for real time availability) and ideally providing a new revenue stream to OTAs and others who are interested in a not yet cut throat margin business.

I'm sysadmin of one full reservation system (400+ hotels around the world) and I can say this an accurate read.

We are integrated with 20+ OTA in different ways, although 90% are pull systems, they call us for data, with the Open Travel Alliance protocol (OTA again), and it is a complex system (which it is not bad).

Maybe I can answer doubts you have :)

Reply on news.ycombinator.com