Resort Availability on DVC.com...when?

But the business logic is set up to be run by someone with months of training - and they STILL screw up even simple reservations - my last reservation had the dining plan completely screwed up. Some of it would need to be greatly simplified to front a workable UI.

Then there is the additional control issues - this opens up Disney to additional opportunities for someone to defraud their customers (us!). Additional regulatory concerns (that they currently have on the CRO site, but would expand over to the DVC side when you start working with transactions. It would probably also make renting point easier, and Disney seems to be doing everything they can to discourage that cottage industry.

What happens when John and Jane Doe log in to discover that they took a trip they didn't take on short notice last month because someone managed to hack their account? What happens when they CLAIM that?

That was the point I was making. The technology is there. It is not like they are writing this from scratch. There are people claiming earlier that it would be too complex.

The question becomes, how much control do they want to give up and can someone do this on their own. I am thinking, no they don't want to give up control. And no, people cannot do it on their own.

Personally, I think a great comprimise would be to have the availability from the previous day searchable. Then you are moving from an OLTP database to datawarehouse database which should be less resource intensive. If they ran a job from the night before and just stuck all of the availability information into a new database that was read only, it would give you an idea of what is available.

The problem with that is that availability would be subject to change, and I am not sure if everyone is going to understand that. :sad2: It is just an approximation. But, it could certainly give everyone a general picture if the room they want is available or not.

But, on the other hand, people are probably going to call just to double check anyways, so it is probably not worth it. ;) It can be hard to come up with technology solution to fix social problems.
 
Personally, I think a great comprimise would be to have the availability from the previous day searchable. Then you are moving from an OLTP database to datawarehouse database which should be less resource intensive. If they ran a job from the night before and just stuck all of the availability information into a new database that was read only, it would give you an idea of what is available.

The problem with that is that availability would be subject to change, and I am not sure if everyone is going to understand that. :sad2: It is just an approximation. But, it could certainly give everyone a general picture if the room they want is available or not.

But, on the other hand, people are probably going to call just to double check anyways, so it is probably not worth it. ;) It can be hard to come up with technology solution to fix social problems.

That's my thought on availability. You never get the ROI you expect on these projects because you never move as many of the calls there as you want. People will discover they get "more accurate" information by calling - so they'll call. Or they'll call to complain that "it was there yesterday."

With some rooms in such limited supply and relatively high demand (AKL conceirge - can you imagine what its going to be like to get those rooms for Christmas?), it will be tough. When DCL allows you to book online, you can't do Suites - you need to call DCL for those.

My company just spent half a million dollars web enabling something for end users - something really simple - so we didn't have to send those calls to an admin. And its unusable - calls have INCREASED as a result of the project - the users can't understand the choices they are being given.

Another social concern - make it easy to change reservations and people will change them - and every change they make increases the chance for error - and the resulting complaints.

There are pieces I think it should be easy to web enable - for instance adding Dining should be a couple of check box clicks - Magical express should be the same - click that you want ME, pick your date, flight and time. (Why, oh why, can't you print out a copy of the confirmation letter?) Rooms, however, I think they will do eventually - but I think it will be error prone and disappointing for the first three releases - and smart people won't use it (like the current waitlist - couldn't get that one right either).
 
My company just spent half a million dollars web enabling something for end users - something really simple - so we didn't have to send those calls to an admin. And its unusable - calls have INCREASED as a result of the project - the users can't understand the choices they are being given.

That is a usability issue. And frankly, using Disney's various websites right now, I wouldn't trust them to build me a site that tells me what time it is in Orlando. :goodvibes
 

That is a usability issue. And frankly, using Disney's various websites right now, I wouldn't trust them to build me a site that tells me what time it is in Orlando. :goodvibes

Actually, in addition to usuability, we misunderstood the requirements, had completely the wrong scope (the scope changed during development - think of it as "we started out trying to develop a system to book rooms a OKW, and ended up needing a system we hadn't designed to handle eight DVC resorts") and released a system that was buggy - the interface between the front end and the back end behaved unpredicatbly. Oh, and we stuck in a data feed that from another source that we were told was clean - and it isn't. In the admin run system, we didn't need the data feed because it was information the admins just "knew" - but the end users would answer inconsistantly.
 
Actually, in addition to usuability, we misunderstood the requirements, had completely the wrong scope (the scope changed during development - think of it as "we started out trying to develop a system to book rooms a OKW, and ended up needing a system we hadn't designed to handle eight DVC resorts") and released a system that was buggy - the interface between the front end and the back end behaved unpredicatbly. Oh, and we stuck in a data feed that from another source that we were told was clean - and it isn't. In the admin run system, we didn't need the data feed because it was information the admins just "knew" - but the end users would answer inconsistantly.

Not to get too far off topic, but: So, a normal project? :rotfl:
 
Not to get too far off topic, but: So, a normal project? :rotfl:

Exactly - which is why I doubt this is as "easy" as the kibbitzers here on the DISBoards think it is. In my experience, a project is never as easy as it looks from the outside. They almost always cost more, take longer to develop, end up with requirements that were "musts" on the cutting room floor in order to get it released at all, and are released buggy.
 
The other problem I see is right now there is an artificial barrier to making reservations. There are only so many cast members answering phones. When they are busy, you are put in a queue on hold. With say 5,000 people trying to make reservations at a time vs. 50 people, that could really put a different spin on things. ;)
 
The other problem I see is right now there is an artificial barrier to making reservations. There are only so many cast members answering phones. When they are busy, you are put in a queue on hold. With say 5,000 people trying to make reservations at a time vs. 50 people, that could really put a different spin on things. ;)

Quint....I think we're going to need a bigger server! :scared1:
 
We Have Made our last 4 reservations on line, Gotten what we wanted each time. All made on weekends and always had answer by Monday afternoon.
 
Quint....I think we're going to need a bigger server! :scared1:

It isn't the processing....its the number of people trying to get reservations vs. the reservations available. The phones provide a capacity bottleneck that, in this case, is actually needed.

Right now there are stories about people going to book a room and discovering that between the time they cancel the reservation they had and the time they go to book the existing room, the room disappears. Imagine how much worse that problem gets when you have 4,000 members dialed in when reservations open for Food and Wine at seven months to get the Boardwalk View room that showed availability yesterday. Add delays in browser refresh from home and the differences in connect speeds - and you have a nightmare of "what happened to my reservation!"
 
It isn't the processing....its the number of people trying to get reservations vs. the reservations available. The phones provide a capacity bottleneck that, in this case, is actually needed.

Right now there are stories about people going to book a room and discovering that between the time they cancel the reservation they had and the time they go to book the existing room, the room disappears. Imagine how much worse that problem gets when you have 4,000 members dialed in when reservations open for Food and Wine at seven months to get the Boardwalk View room that showed availability yesterday. Add delays in browser refresh from home and the differences in connect speeds - and you have a nightmare of "what happened to my reservation!"

And that is the kicker. We don't plan far enough in advance to ever hit the 7 month window much less the 11 month window. I don't know what type of call load there even is at that time. The most I usually wait on call to speak to a CM is about 5 minutes. I don't know what kind of increase you would see; a factor of 2, 5, 10?
 
And that is the kicker. We don't plan far enough in advance to ever hit the 7 month window much less the 11 month window. I don't know what type of call load there even is at that time. The most I usually wait on call to speak to a CM is about 5 minutes. I don't know what kind of increase you would see; a factor of 2, 5, 10?

And even if you knew that, you wouldn't know enough to extrapolate behavior once its on the web. Having the information on availability and having it self serve will change behavior.

You'll also be open to people who write bots that grab prime reservations for - gasp! - rentals. The aforementioned husband has been on several bot killing missions for game console releases - and some other really ingenious ways to hack the systems to gain advantage.
 
An on-line reservation system:

Part A: Resort availability. A 'live' database that keeps track of every villa at every DVC resort:

Resorts: (There are currently 8, with more coming)
Villa: Mostly Studio, 1-B/R, 2-B/R, 3-B/R (at selected resorts. VB also has Inn rooms which are different from Studios)
Views: Mostly Standard view at resorts. (BWV has two categories, standard and preferred, and also has Boardwalk View). AKV has four categories for each of the villa types, such as Value, Standarad, Savanna, Concierge. VB has Inn Ocean View or Garden View.
Other: HCA rooms dues to medical reasons. First floor rooms due to medical reasons. Elevator accessible due to medical reasons.
Requests: Near pool, near elevator, near front desk/hospitality house, etc. (Smoking non-smoking is no longer applicable)

Part B: Making a reservation, initial information:

1. Check the member's account, are their dues paid in full? If not, they are not allowed to make a reservation.
2. Check the member's account. Do they have a loan? If they do, is it overdue? If it is, they are not allowed to make a reservation.
3. Check how many memberships the owner has. (This is the blue membership card. Most have only one, but many have multiple memberships)
4. Check the resorts owned and group them by membership numbers and by add-on resorts.
5. Get how many points they own under each resort and under each membership number.
6. Get the number of points they have remaining in their account.
7. Classify all the points remaining into five groups: Regular Points, Banked Points, Borrowed Points, Holding Points, Developer Points
8. Get the reservation request from the owner. EG they want a 2-B/R BWV boardwalk view check-in October 10 for 5 nights.

Make the actual reservation:
1. Is the room available. Yes, then proceed. No, then inform user. (If desired, the system could also develop a wait list with more questions such as day-by-day waitlist or full reservation waitlist, waitlist cancellation date, waitlist notification: Automatic book or call first?)

2. Present the user with all the points eligible to use for this vacation reservation. This means examining all the point possibilities against the reservation itself to see what is allowed. If the reservation is more than 7-months away, then only home resort points are allowed. Special rules apply to Developer points and to Holding points which must also be checked. If it's less than 7-months away, then all points under the same membership numbers 'might' be eligible to use for this reservation.

3. Determine the use year the vacation falls within, and then determine what points the owner has that are within that specific use year. In the case of multiple membership numbers, multiple use years are possible. It's possible the user might be presented with allowable points from two different membership numbers and several add-on's under each membership number, and with points in different use years.

4. The reservation may require 220 points. The user starts by saying use 50 points from membership number "X", Add-on number "A", and must select which of the point categories to use (Banked, Borrowed, Current, Developer, Holding) and determine if those points are eligible to be used for the particular reservation. The system must then not allow any further points be assigned using membership number "Z" since you cannot combine points from two different memberships to make a reservation. The user can then say use 50 points from Add-on contract number "B", again from only the point categories allowed.

5. Currently if someone owns OKW for example and has multiple contracts, MS takes care of using the points from the add-on contracts as required. EG a master contract of 150 points, a 30 point add-on, a 50 point add-on, and a 70 point add-on. (To the member this is the same as a 325 point contract, but internally it is 4 separate contracts, as seen on your annual dues statements) The member wants to make a 220 point reservation. MS will take the 220 points from the above contracts. In an on-line system the user would have to define the contracts. EG say use 100 points of the 150 point contract, use all 30 points from add-on #1, use all 50 points from add-on #2, and use 20 points from add-on #3. (Unfortunately that's only 200 points and the reservation requires 220. The system has to tell the user and require additional points be assigned to this reservation). In other words, the system will require numerous checks and balances. Currently a member can specify to MS which contracts to use when multiple Same-Resort add-on's exist. Normally this is not done and MS does this for you. In an on-line system you would have to do this yourself. (Or make the program even more complex by doing it automatically, or more complex yet, by offering you a choice)

6. If the reservation is less than 7-months away, then also factor in that points can be used from different Add-on contracts at different resorts. Tell the system to use 100 points from OKW contract, Add-on Number 2, use 50 points from BCV contract, Add-on # 5, and use 70 points from AKV contract, Add-on #8!

7. The system needs to check and prompt if it thinks the user is making a mistake. EG someone has both banked and regular points in a use year, and selects to use regular points for a reservation, it would be useful for the system to prompt and ask if they really want to use the banked points instead. Image the outrage if members book on line and didn't use banked points, which subsequently expired.

8. There's more yet, including getting all the other information concerning room occupants names, etc. An additional check here should be added that if someone tried to put in 6 names and is trying to reserve a studio, the system does not allow it. And how to handle other requests?

Compare all the above to a hotel reservation where you can go on line, request a room, give a credit card number, and be done. Simple. DVC reservations are NOT simple.

Even comparing to the IRS system is apples to oranges. IRS code is set for taxes for the PREVIOUS year, and not filed until the following year. Everything is static by that time. Your income for the year is done, your exemptions are done, your itemized deductions are done. In other words there are no changes to the information that will take place between the end of the year (Dec 31st) and when you actually will file (by April 15th). It's a STATIC system and thus programs can be written to incorporate all the rules for this static system. An IRS system is not an on-line system. Even when you use a canned program or an on-line (fill out the form type of program), it is not a 'live' connection to your SS account number. You complete the information and send it in. It is review AFTER that. Preliminary by computers and possibly by auditors. The IRS system is more similar to using the DVC e-mail system to request a reservation where you fill out a form and send it in. It is not a 'live' system.

H&R block can write a tax program and sell it to a million people for $29. That's $29,000,000 in sales. How much are members willing to have their dues go up to have an on-line reservation system. Marriott can afford a relatively 'cheap' on-line system for their hotel chain of I believe Thousands of hotels, and millions of customers. DVC has about 150,000 actual memberships. If 5% of them would use an on-line system, that's only 7,500 members. Do we really want DVC to spent a million dollars (or maybe much more?) for a system that only 7,500 people would use?

DVC is a live DYNAMIC system, not static. Things change all the time. Not pay your dues? That's a change. Bank points? That's a change. Buy another Add-On? That's a change. Make a reservation just 5-minutes ago? That's a change. An on-line system has to cope with ALL of this. It has to be error-proof. It has to have built in all the things a person could do wrong and catch it before letting it process through.

Could a system be developed? Certainly, the technology is there. Would it be expensive? Absolutely! Would it be cost effective? Big unknown. Most likely not, at least for a long time. The system is complex. The key is to make it simple so that people will use it. That's not easy. And if people won't use it, you spent a ton of money for no reason.

Finally, a system like that would not be a 'create it and forget it' situation. DVC is constantly changing. The system would require constant updates. Things like adding new resorts, changing banking rules, etc. All that is not free. Those of you using my preliminary DVC-Tracker program are probably aware of the loop DVC threw me when AKV was announced, with their combination of villas that became available. No longer the simple Studio, 1-B/R, 2-B/R such as at BCV, VWL, or even a 3-B/R available at BWV, OKW, SSR, but had to throw in Value, Regular, Savanna, Concierge!. It's great for members, but adds complexity to computer programs. Imagine a DVC on-line system already set up which has to be modified within a year to accommodate GCV's, and only a few years later to accommodate DVC-Hawaii. Not to mention that CRV's may be in the works.

The initial cost of a program: HIGH
The maintenance and upkeep: Unknown
Program Updates: HIGH

Cost Effective?

Just .02
Caskbill
 
When it is argued that an online system is too difficult to mange one thing that is ignored is that a system already exists today. MS uses this system. MS is able to book rooms from a "live" database. MS is able to book a room using multiple contracts, multiple use years, multiple resorts points, etc. MS does this. It is a PROVEN system. It works. Buying event tickets online pose the same live database problems on a FIXED scale. (And ticket sales will swamp a network much faster than DVC reservations) It has been done. rsinj is correct in saying that the technology is not the gating factor.

One reason MS needs to ask and confirm many questions is not because of the complexity of the software it is simply to make sure they understand what the member is asking them to do. People make many assumptions when a transaction is taking place (such as, of course I want my banked points used first, of course if I book at the AKV I want my AKV points used, etc). These assumptions change from member to member. An online system would take these assumptions out of the equation.

Another reason MS needs to ask questions is that they have the information in front of them. As a member on the dumb end of the phone we can't see the screen to understand what will happen when submit is clicked. They are simply verifying they got it right. An online system would allow the user to confirm this himself.

I believe that confusion between what the member request and what MS interprets as the request is the cause for the vast majority of problems. An online system where the member can self verify would surely reduce these errors. I know I would triple check to make sure I got it right before I hit submit.

As to the argument it is not fair, it is no less fair than the forcing us to queue in a phone line. What If I get Tom, the new MS guy who was out partying until 3am. His ignorance and hangover slow him down so much that 18 calls placed after mine are completed before mine. Unlikely? yes, but no more so than an online systems unfairness. Also, no one is advocating that ALL MS be eliminated. There will always be an option to call into the same system we have today.

I work in the software industry. This is not a complicated problem. Walmarts system to monitor inventory is much more difficult problem. A different set of problems, yes. But more difficult none the less. It has saved them billions, literally. I can't believe the ROI is not there to enable MS's version of the system to web based.

And security concerns can easily be resolved. An easy security feature would be to allow email addresses to be changed only by snail mail. And once a reservation is made a confirmation must be made by responding to an email sent by DVC or the room is returned into the pool after 24 hours. Rooms cannot be booked online withing 72 hours of check-in.

The current phone systems security is less than that simple procedure. If I find a wallet with a SS# and DVC # I am in. I can call and make a reservation for a quick trip. The member would not be aware of this trip until the confirmation letter arrived. If I was really mean I could change the members address to avoid the letter.

The hardest part of this is developing a user-friendly front-end. Disney is notorious for leaning towrds the complex. Bad choice for this application. Google provides a better solution. A very simple interface.


I have yet to read any reason this could not or should not be done. Caskbill's description, while a verbose description of one of the more complex situations, is still a rather simple problem. If you can calculate this using pencil and paper it is not a hard problem for software.
 
Steve,

With all due respect, there is a world of difference between a system that can be used by full-time employees who have undergone extensive training, and a system that the general public can just log on and use.

I work in the financial services industry, and have worked a number of self-service projects. I assure you they are very, very hard. You know the guy who can't stop his VCR clock from blinking 12:00 (and, yes, he has a VCR and not a DVD player)? He's one of your customers. He has to not only understand what he's doing, but not be able to do any harm. That's hard.

As for Walmart's inventory system - what do you suppose they spent on that? I don't think I want such a thing paid for out of membership dues. A self-service website may be cheaper, but it could very well cost tens of millions and still not work right. It's not uncommon for large companies to spend a fortune on a system that doesn't work and have to eat the cost.

Maybe DVC they could pull it off; maybe they couldn't. Do you really want to take a chance with our money? If it were a matter of buying something off the shelf, I might feel different. But this is large-scale custom development. Very expensive and risky.

When Disney can get their own on-line user interfaces working great, I might trust them to build us a system. But not now.

PS - I do agree that the security and queuing problems aren't the hard part.
 
I agree the hard part of this is the user interface. It is a difficult problem to present data in a way that makes it intuitive. While difficult it is not impossible. A confirmation page that describes what transaction will take place would eliminate the confusion of the page with the options.

If the member can understand the print confirmation he receives today he can understand the confirmation page.

As for those who have difficulty there will be the phone.

Example:
2-B/R BWV boardwalk view check-in October 10 for 5 nights (Caskbills)

2-BEDROOM BOARDWALK VIEW
Oct 10 Tue 32 32 from Contract 1
Oct 11 Wed 32 32 from Contract 1
Oct 12 Thur 32 32 from Contract 1
Oct 13 Fri 61 4 from Contract 1
-----------------30 from Add-on 1
-----------------27 from Add-on 2
Oct 14 Sat 61 23 from Add-on 2
----------------38 from Add-on 3


This is the type of work I do. Just because it is difficult and has been done poorly in the past does not mean it can't be done well. I can be done and done well. This is not an overly difficult problem.

I would love to see this (if you can't tell.) But the current system needs to remain in place as well. I think both would be a great benefit to all.
 
Then you obviously know that without understanding what data is being pulled real time, which data is batch processed that the CMs are able to override, what the processing capabilities of the system are, what the current hardware capacity is and bandwidth, what the regulatory requirements are to permit customer access, what the process will be for backing out an end user error, its impossible to say its as simple as putting a user interface on an existing system.
 
You are right crisi I should have never said it was "as simple as putting a user interface on an existing system". Oh wait. I didn't say that.
 
Well, I still don't think that technically it is much work. Like I said earlier, the business logic already exists. The CMs use it right now.

The problem is the social aspect. With all of the questions that are posted here,....

I just don't think that disney would want to give up the control.
 





New Posts







DIS Facebook DIS youtube DIS Instagram DIS Pinterest DIS Tiktok DIS Twitter

Add as a preferred source on Google

Back
Top Bottom