Hotels.com - How much to replicate

yop

Guys,
Someone asked me the last day if he wanted something like "Hotels.com which can handles bookings and reservations but also be capable of expansion into other booking areas such as concerts, restaurants, cinemas etc"


Now I was thinking to myself about this and well it seems like a big job, but I can't quantify it in either time or even money.
To me its a hotel bookings system to start with but then its also a system for concert and restaurants and cinemas, so its really 4 seperate entities.

It could use the same booking engine but would require 4 different web interfaces.

But since I am not a very experienced web programmer (12 years of programming but not a lot of web ) I would appreciate your ROUGH and I mean ballpark figures.

Thanks

ntlbell

for what?

the development? the design? the infrastructure required?

the on going maintenance?

maybe if you're a bit more specific

yop

sorry, Monday morning, my brain is not turned on yet

The comment he was was "to develop" so I will take it that they will look after the hosting but that I would have to look at the development and the future maintenance.

So on top of the development we would be looking at a domain name, hosting etc.

I would just like a ball park idea of time and cost. I mean without doing too much thinking would he be looking at 10k, 30k, 50k that sort of thing.

I think he is testing the water more than anything.

Evil Phil

Off the top of my head if you could replicate hotels.com for 50k he'd be a very lucky man. Doesn't mean you can't provide a solution within 50k though.

ntlbell

yop are you looking for ball park figures of what you should charge him?

or are you just looking at what he might have to pay someone else?

trying to quantify the time if you're doing it is very difficult as you don't have much web dev experience

You could definitely build a scaled down version of it for well under 50k

as long as the foundations are done right it can be hosted on a cloud platform which would give them plenty of flexibility with regards scaling as they grow.

yop

Pure ball park. He asked me to look at it and give him an idea of what it would cost.

I wouldn't take on the project as I said as I dont have enough web dev experience to handle the developement of this.

So I would go back and say you are looking at around 50/60k euro. I dont want him to think he can get it done for 10k and then say "Ok" I will go looking for someone!

Thanks for the info.

ntlbell

yop wrote: »

Pure ball park. He asked me to look at it and give him an idea of what it would cost.

I wouldn't take on the project as I said as I dont have enough web dev experience to handle the developement of this.

So I would go back and say you are looking at around 50/60k euro. I dont want him to think he can get it done for 10k and then say "Ok" I will go looking for someone!

Thanks for the info.

Personally I think 50k is far too much. the days of build it and they will come are long gone.

if he started off fairly small there's no need at all to be spending 50k

plus if he gets it built somewhere like russia or poland he will get a hell of a lot more bang for his buck.

if he puts a bit of effort into finding the right dev house or a decent freelancer he could be up and running for under 20k

John_Mc

Is it for a load of hotels or for one in particular? I've done a booking system for a single hotel and am in the process of redoing it. There's a bit of work involved but not as much as one would think. That said, I have the UI's and workflows already done so I just need to code around them.

If it were for a load of hotels you'd need to consider how you'd handle retrieving the availability of a room and it's rate for each of the hotels. That could require quite a lot of work if it were to be done properly.

yop

It would be for a load of hotel, hostel etc as per Hotels.com

The India, Russia route sprang to mind for me yes, I would be able to PM the job for him but have had varying experiences with working with Indian developers from good to very poor.

ntlbell

yop wrote: »

It would be for a load of hotel, hostel etc as per Hotels.com

The India, Russia route sprang to mind for me yes, I would be able to PM the job for him but have had varying experiences with working with Indian developers from good to very poor.

so go back to the good one's? or get recommendations I'm sure with your 12yrs exp in dev you'll have no problem getting plenty of feedback from others.

colm_c

The coding on a project like this would be IMO one of the easier tasks, no real challenges, making the UI intuitive, getting the marketing sorted as well as getting hotels to sign up and give your site preferential rates is the big tasks here, otherwise the whole project is pointless.

jmcc

yop wrote: »

To me its a hotel bookings system to start with but then its also a system for concert and restaurants and cinemas, so its really 4 seperate entities.

It is really a large database backed system with a number of interfaces. It is complex and expensive to do well. You could probably throw cheap students at the problem but it really needs the back end database work to be done by professionals and getting good database people (database architects and database adminstrators) is not cheap. You also need a rock solid infrastructure that can take the data from the sources (the hotels, cinemas etc) and integrate it with the databases. And the billing system needs to work. Then you've got to have a well designed set of websites and marketing.

But since I am not a very experienced web programmer (12 years of programming but not a lot of web ) I would appreciate your ROUGH and I mean ballpark figures.

You could probably get an amateurish piece of junk (a few hotels and cinemas) for under 100K that would fold under the first major load. A decent solution would be upwards of a few million to do well.

Regards...jmcc

jmcc

Evil Phil wrote: »

Off the top of my head if you could replicate hotels.com for 50k he'd be a very lucky man. Doesn't mean you can't provide a solution within 50k though.

Lucky as in a man who has won the Euro Lottery every time since it started?

Regards...jmcc

jmcc

colm_c wrote: »

The coding on a project like this would be IMO one of the easier tasks, no real challenges, making the UI intuitive, getting the marketing sorted as well as getting hotels to sign up and give your site preferential rates is the big tasks here, otherwise the whole project is pointless.

Coding as in web design or coding as in real coding? Getting the database updated in near realtime is the most complex part of this project. It has to take various types of data, parse them, and then integrate them into a database which will then be published out. If there is a common data interchange specification, then it will be fine. However if the input data is not standard then there will be a lot of coding to get it into an acceptable data format. The web side of the site would have to interface well with the database(s). The database backend is what makes sites like hotel.com the icebergs of the web. The users only see a small percentage whereas the bulk of the site is beneath the surface.

Regards...jmcc

jmcc

ntlbell wrote: »

if he started off fairly small there's no need at all to be spending 50k

The problem is that small sites do not necessarily scale well. A simple e-commerce site with a small MySQL backend is very different to Amazon.com in scale and concept. If you are going to do it right then you've got to design a scalable site from the start. Otherwise you will end up having to junk the smaller site when traffic overwhelms it and redesign everything.

If he could outsource much of the database element by using existing booking databases and services, then the project would be cheaper.

Regards...jmcc

colm_c

jmcc wrote: »

Coding as in web design or coding as in real coding? Getting the database updated in near realtime is the most complex part of this project. It has to take various types of data, parse them, and then integrate them into a database which will then be published out. If there is a common data interchange specification, then it will be fine. However if the input data is not standard then there will be a lot of coding to get it into an acceptable data format. The web side of the site would have to interface well with the database(s). The database backend is what makes sites like hotel.com the icebergs of the web. The users only see a small percentage whereas the bulk of the site is beneath the surface.

Regards...jmcc

Both - there is nothing ground breaking here, it's just takes time, knowledge and money.

What I'm saying is that if you're going to spend a lot of money on this project you better make damn sure you're getting your profit and users using the thing, otherwise it could end up like many government projects, abandoned.

In terms of complexity and scalability -- starting off with a project to build something like hotels.com IMO is a mistake, you need to prove your revenue model on a smaller scale -- and see if you can in fact complete with hotels.com or similar services otherwise you'll be embarking on a 2-3 year project for nothing.

Hiring a small team of agile developers initially to get a concept up and running quick (1 month) to prove the model, then if it takes off look at porting it to a more scalable model/design.

tomED

Ok, well we've just been there and there are a few fundemental issues which people are over looking. The main one is where you will source the content from.....

Generally sites *like* hotels.com will have a feed of data from somewhere. There are many places that do feeds and each is different. The most basic one I've seen has been IAN.com - they have some good simple integration ideas and then they have more advanced options, but you need to prove a certain level of traffic first.

The one we just completed looked at two different options, because of client confidentiality I can't say who they were. Both were completely different, so different that we ended up having to ditch one of them because it didn't allow us to build a site the way the client wanted it. So researching where you will get the data from is imperative.

IAN.com is free, but most of the really good ones, with better margins can cost quite a lot per annum - possibly even more than it costs to build the website itself!

In terms of coding, it's completely down to the feed. Some feeds have a complete well designed API that you need to make calls with XML to get an XML response, while others are simply just XML data feeds and then you are sent somewhere else to confirm bookings and payments etc.

So if it's a hotel.com he wants, I would say there's no way he'd get it for under €50k. A slimmed down version, yes definitely - but don't forget the feed cost. Although the coding is not rocket science, it can be very time consuming, especially when working with a third party API.

Hope this helps.

tomED

colm_c wrote: »

Hiring a small team of agile developers initially to get a concept up and running quick (1 month) to prove the model, then if it takes off look at porting it to a more scalable model/design.

Although in general I would agree with this, sometimes people have enough money to spend to make something like this a success. If you are willing to spend the money, it could work.

If you are on a tight budget, that's a different story. In that case, you need to make the best of the money you've got and in reality, most of your money will be needed on the marketing side of things. So a simple site that does the trick can grow organically and be built better in phase 2.

yop

tomED wrote: »

Ok, well we've just been there and there are a few fundemental issues which people are over looking. The main one is where you will source the content from.....

Generally sites *like* hotels.com will have a feed of data from somewhere. There are many places that do feeds and each is different. The most basic one I've seen has been IAN.com - they have some good simple integration ideas and then they have more advanced options, but you need to prove a certain level of traffic first.

The one we just completed looked at two different options, because of client confidentiality I can't say who they were. Both were completely different, so different that we ended up having to ditch one of them because it didn't allow us to build a site the way the client wanted it. So researching where you will get the data from is imperative.

IAN.com is free, but most of the really good ones, with better margins can cost quite a lot per annum - possibly even more than it costs to build the website itself!

In terms of coding, it's completely down to the feed. Some feeds have a complete well designed API that you need to make calls with XML to get an XML response, while others are simply just XML data feeds and then you are sent somewhere else to confirm bookings and payments etc.

So if it's a hotel.com he wants, I would say there's no way he'd get it for under €50k. A slimmed down version, yes definitely - but don't forget the feed cost. Although the coding is not rocket science, it can be very time consuming, especially when working with a third party API.

Hope this helps.

Excellent information thanks for that.
That IAN.COM is a good site, never knew it existed.

I will be going back today to that lad, will be interesting to see what he says!

tomED

yop wrote: »

Excellent information thanks for that.
That IAN.COM is a good site, never knew it existed.

I will be going back today to that lad, will be interesting to see what he says!

No problem - best of luck with it!
By the way, some things I didn't make very clear when I read it back.

IAN.com is an affiliate program, with small margins. The other feeds were more like a travel agents feed if that makes sense, so you add your own margin.

Also - we ditched one feed HALF WAY through the development!!! Documentation wasn't clear and their feed was terrible, if you are a programmer, you will understand when I say none of the hotels had a unique id!!

Evil Phil

Hotels.com is partnered with expedia so I presume their getting their feed from them. If somebody asked me for a quote for a site like hotels.com I'd say at least 500k.

jmcc

Evil Phil wrote: »

Hotels.com is partnered with expedia so I presume their getting their feed from them. If somebody asked me for a quote for a site like hotels.com I'd say at least 500k.

Apart from the hotels feeds, it would also have to integrate cinema feeds etc. The entertainment.ie site integrates cinema data. However I don't think it has booking facilities. Getting the design right from the start would be very important if it is to be expanded in the future. This is why I'd be wary of a try it and see solution. If the funding for a real operation isn't there then it is not worth doing it at anything even approaching the scale of the hotels.com model. Sticking with feeds and very little else might be the simplest and cheapest solution.

Regards...jmcc

tomED

jmcc wrote: »

Apart from the hotels feeds, it would also have to integrate cinema feeds etc. The entertainment.ie site integrates cinema data. However I don't think it has booking facilities.

Regards...jmcc

The feed we were working with had tours and theatre tickets etc for the US obviously...

yop

Evil Phil wrote: »

Hotels.com is partnered with expedia so I presume their getting their feed from them. If somebody asked me for a quote for a site like hotels.com I'd say at least 500k.

SERIOUSLY!!!! That seems a bit extreme does it not Phil?

Why would you quote 500k, there is no way a customer would go with a 500k quote!

I would have thought that as the lads said here you would get a slimmed down version for about 40k, with a full blow system for about 150k.

Why would you use Unique ID's, TomED, they are for girls

What sort of time scale then for something like that, based on what you have done TomED

tomED

yop wrote: »

Why would you use Unique ID's, TomED, they are for girls

hahahahahaha

yop wrote: »

What sort of time scale then for something like that, based on what you have done TomED

It really comes down to how far you want to go. The basic IAN integration, you could have a really simple (now I mean simple!) system up and running in a couple of days.

With a proper XML API, where you have to handle transactions, set up crons to retrieve hotel information, credit card payment integration, search engine friendly, image handling, etc etc - easily 2 months. And that would give you a good basic working system. There's a good bit to do on the backend for handling bookings and cancelling etc.

You could easily spend 3-4 months on the one we developed, which doesn't have half the features Hotels.com would have.

jmcc

tomED wrote: »

The feed we were working with had tours and theatre tickets etc for the US obviously...

Which would be great for Americans and people in the US. The initial spec is a bit nebulous. There is no defined market or target audience so it looks like a tyre kicker request.

Regards...jmcc

jmcc

yop wrote: »

SERIOUSLY!!!! That seems a bit extreme does it not Phil?

To do it properly from the start, 500K would be cheap.

Why would you quote 500k, there is no way a customer would go with a 500k quote!

If a customer is serious and not a tyre kicker who has seen a few sites on the web, then they should have a proper budget and funding.

I would have thought that as the lads said here you would get a slimmed down version for about 40k, with a full blow system for about 150k.

Only a few people here have ever worked on large scale sites and really appreciate what a large scale website involves from a hardware and software point of view. This is why you are seeing wildly varying estimates of the costs. The hardware part, the servers, hosting and administration hasn't even been addressed yet.

Regards...jmcc

yop

You have it in one with the tyre kicker, I have found out since that this guy has "an interest" in a hotel, now whether he has an interest in a chain of hotels I am not 100% sure.

What has also transpired is that he is also interested in getting a full back office, front desk, stock control and night audit system put in place.
This I can manage as its familiar territory, what is definately obvious is that this guy wants one decent system with all areas from back-end, front-desk and web interface in place.

The zero's are quickly been added to the the figures

Timescales and "grasp of work" in a web site this size is where I am lacking so I do apologise for sounding like a right thick on that aspect.


Anyone fancy giving this guy his 1m euro product

tomED

jmcc wrote: »

Which would be great for Americans and people in the US.

Emmm no - what it's great for is people travelling to the US.

There are other tours for sale in the feed from other countries, but not theatre etc. I wouldn't say it will be long before the feed providers will have a lot on board.

jmcc wrote: »

The initial spec is a bit nebulous. There is no defined market or target audience so it looks like a tyre kicker request.

jmcc wrote: »

To do it properly from the start, 500K would be cheap.

If a customer is serious and not a tyre kicker who has seen a few sites on the web, then they should have a proper budget and funding.

Of course it's a tyre kicker - isn't everyone when they are deciding whether they have the funds to proceed with an investment? He's looking for ballpark timelines and prices. Yop also hasn't stated what kind of a client they are, could be a client doing something completely different so he can't be serious until he knows his risk.

I know all us developers hate the "how much would a site like bebo cost", but at the end of the day it's up to us to manage their expectations. There are cheap ways and there are expensive ways. Providing them with the knowledge to bring it to a defined level is better than saying No, you don;t understand what's involved so talk to me when you have €500k.

jmcc wrote: »

Only a few people here have ever worked on large scale sites and really appreciate what a large scale website involves from a hardware and software point of view. This is why you are seeing wildly varying estimates of the costs. The hardware part, the servers, hosting and administration hasn't even been addressed yet.

I would have to say if the site is properly coded, hardware wouldn't be an issue at this stage. Even if you throw out basic figures, if you made €50 (which would be a lot I know) per hotel room sold and your conversion rate of visitors was in the region of 5%, you would only need 20,000 vistors to your site over a year. Obviously that's quite ambitious, but you can see straight away that it's not a huge amount of traffic to gain a return.

So before you even think about hardware, you'd need to make sure you can cover the costs of building a great website and the costs of a marketing campaign.

If he had €500k to spend, I'd still spend €50k on a half decent website and the rest on marketing it!! When that money is made back, invest in getting something better.

By the way - I know of a holiday site that paid €18k for their site selling cheap holidays. They made their money back within months - but then stopped investing and now they are just ticking over.

jmcc

tomED wrote: »

Of course it's a tyre kicker - isn't everyone when they are deciding whether they have the funds to proceed with an investment? He's looking for ballpark timelines and prices. Yop also hasn't stated what kind of a client they are, could be a client doing something completely different so he can't be serious until he knows his risk.

And this is the heart of the problem - not even the client knows exactly what they want. Until there is a spec it is all just conversation.

I would have to say if the site is properly coded, hardware wouldn't be an issue at this stage. Even if you throw out basic figures, if you made €50 (which would be a lot I know) per hotel room sold and your conversion rate of visitors was in the region of 5%, you would only need 20,000 vistors to your site over a year. Obviously that's quite ambitious, but you can see straight away that it's not a huge amount of traffic to gain a return.

But is it up to the developers to work out the financial return on the project? Marketing should be part of the project but this would be moving too much towards a feasibility study and would involve a lot of work.

Regards...jmcc

ntlbell

jmcc wrote: »

The problem is that small sites do not necessarily scale well. A simple e-commerce site with a small MySQL backend is very different to Amazon.com in scale and concept. If you are going to do it right then you've got to design a scalable site from the start. Otherwise you will end up having to junk the smaller site when traffic overwhelms it and redesign everything.

If he could outsource much of the database element by using existing booking databases and services, then the project would be cheaper.

Regards...jmcc

but you don't have to spend 50k to build a good foundation which is what i suggested at the start.