#1
biddingtraveler , Dec 16, 2010 10:48 am
Sep 8, 2013Hey FT folk,
Some buddies of mine and I wrote a site a while back specifically geared towards Priceline bidders. We've been lucky enough to have our own set of enthusiastic beta testers, but I'd like to kick it up a notch by seeing what you guys have to say. This board has a ton of grizzled veterans, a real tough crowd, so I'm fully expecting to get flamed. But you can't improve without criticism right?
Basically, if you hate filtering out all the crap you need to make a bid (e.g. Los Angeles bids when you're bidding in San Francisco), or you hate doing the whole permutation calculation for rebids, or you just wish the bidding experience wasn't so frustrating, then check us out.
The Bidding Traveler - Priceline hotel bidding help
The Bidding Traveler - Priceline hotel bidding help
Some buddies of mine and I wrote a site a while back specifically geared towards Priceline bidders. We've been lucky enough to have our own set of enthusiastic beta testers, but I'd like to kick it up a notch by seeing what you guys have to say. This board has a ton of grizzled veterans, a real tough crowd, so I'm fully expecting to get flamed. But you can't improve without criticism right?
Basically, if you hate filtering out all the crap you need to make a bid (e.g. Los Angeles bids when you're bidding in San Francisco), or you hate doing the whole permutation calculation for rebids, or you just wish the bidding experience wasn't so frustrating, then check us out.
The Bidding Traveler - Priceline hotel bidding help
The Bidding Traveler - Priceline hotel bidding help
#3
I'm confused exactly what your autobid system does (the help window on what it is, well is of no help). How long will it take, will it just keep submitting, or will it submit once a day at a higher rate? I think you need to lay out the expectations a bit better so I know what to expect. My assumption would be if I tell it I want a 3 1/2 star hotel at a certain zone in Newark on some date between $50 and $80 dollars I would find out in a few minutes if that worked or not, but it's totally possible you put in a $50 bid today, $55 tomorrow, etc.
Airport codes should be allowed to be used as search terms on the front page
Airport codes should be allowed to be used as search terms on the front page
#4
biddingtraveler , Dec 16, 2010 12:13 pm
Hi Mike,
thanks for checking us out. To answer your question, the site will calculate all your rebids and execute them all in sequence, immediately. To use a simple example, if you bid for two zones 1 and 2, with a lowball bid of $80 and a max bid of $85, we'll first bid this way:
zone 1: $80
zone 2: $80
zones 1 + 2: $85
That way you don't have to work out the math to develop the optimal bidding strategy. Once you add more zones the math gets pretty complex. I wrote a blog post about it here.
Airport codes on search. Got it. It's on the list!
thanks for checking us out. To answer your question, the site will calculate all your rebids and execute them all in sequence, immediately. To use a simple example, if you bid for two zones 1 and 2, with a lowball bid of $80 and a max bid of $85, we'll first bid this way:
zone 1: $80
zone 2: $80
zones 1 + 2: $85
That way you don't have to work out the math to develop the optimal bidding strategy. Once you add more zones the math gets pretty complex. I wrote a blog post about it here.
Airport codes on search. Got it. It's on the list!
#6
Very interesting blog, by the way. I've always known that I could do more bid permutations that I had been doing, but I've been to lazy to analyze it. Your blog post lays it out in detail.
Admittedly I didn't bid using the site yet, but if your site does the work of figuring out (and applying) the optimal bidding strategy, I look forward to trying it out.
Admittedly I didn't bid using the site yet, but if your site does the work of figuring out (and applying) the optimal bidding strategy, I look forward to trying it out.
#7
Mountain Man , Dec 17, 2010 8:24 am
Your site looks very useful - I've used Priceline extensively over the past 10 years and have probably accumulated 30-40 successful bids, most for myself, but also for friends who are not familiar with the ins and outs of the bidding system. The tool would definitely help the Priceline neophyte or someone who deosn't have the time to work out a complicated bidding strategy.
I'd second some of cordelli's comments from above.
P.S. I found the blog entry very interesting too. Fascinating to realize that with 5 zones and 2 free rebid zones there are 24 possible bid permutations. I've often used Priceline to stay in 4 star hotels in New York - that's the sort of city where bidding strategies such as the one outlined would really work.
I'd second some of cordelli's comments from above.
- It might be helpful to have more navigation instructions on the bidding page - I only discovered by chance that by clicking on the little green bubbles on the map I could get a list of all the winning bids with their dates for a particular hotel.
P.S. I found the blog entry very interesting too. Fascinating to realize that with 5 zones and 2 free rebid zones there are 24 possible bid permutations. I've often used Priceline to stay in 4 star hotels in New York - that's the sort of city where bidding strategies such as the one outlined would really work.
#8
biddingtraveler , Dec 17, 2010 1:12 pm
Hey, thanks for the comments and the feedback. I'm writing a reply in a timely manner, but it'll be awhile before you see it, since I'm still under moderation.
Anyway, by the time my original reply to cordelli gets posted this will just be a reiteration, but.. yes, the site automatically calculates the entire bidding strategy and does it within minutes. Waiting 24 hours would suck. People won't want to use software that sucks.
Also, what is probably not clear and that we need to message more clearly is that the bidding is actually done by your own computer (it's just driven by some automation software we wrote). So there's no issue of personal information getting stored on our servers.. all we do is set it up so that your own computer can do it. This makes it safer for everyone, since we don't have to store credit card numbers -- everything goes between you and Priceline. Unfortunately, we've been struggling on how to tell users through the web page. The standard reaction is "why would I give these unknown guys my credit card".. doesn't matter what you say, habits are hard to break!
To generally answer your questions, this site is pretty algorithmic (we're all from a very quant/software background). All the stuff shown to you - recent bids, rejected bids, the hotel popups, the bid data.. it's all been specifically tailored for your query.
Let's use an example. One thing that annoys me about seeing raw winning bid info or hotel lists is that there's no guidance behind that information. Say you see that someone won the Venetian in Vegas for $49 a night. But then if you look closely you'll see that they won on a Wednesday. Then what sucks is that the reader thinks that information is useful and goes to bid $49 (or worse, $45) for a 5 star on a SATURDAY night. So in this case the past information was actually harmful... both for Priceline, who doesn't sell the room, and the user, who goes away frustrated. In contrast, we've built statistical profiles for your query, which vary by day of week, and other factors, and present directly relevant information. A particularly interesting sidenote is that there are areas which suffer from the inverse problem (high weekday rates, low weekend rates), particularly business-centric areas. We identify them algorithmically and show only information which will help you.
Now, regarding the bid increments, this has been something of a dilemna. Yes, if you did a 5 star in Vegas, or another area where you theoretically have thousands of bids at your disposal, and you could theoretically start a bid at zero and walk it up a dollar at a time until you hit the target. But this will just piss off the user, who has to wait, and Priceline, who sees useless bids clog the system.
Being somewhat of a data guy I believe the solution lies in the data itself. What I mean is that once we crunch enough numbers for you and present you relevant information, we distill for you a market value of what an opaque hotel costs at a particular moment in time. At that point, the permutation bidding is really just a crutch -- wiggle room, if you will -- to just get a little more precise... since you should know approximately what it's going to cost, already! So given all the inputs: zones you've selected, past wins, rejected bids, what you've decided your lowball and final offers are -- the software calculates the sweet spot for what each bid should be. And by sweet spot I mean sweet enough for both parties: Priceline, who sells the room, and the user, who does not grossly overpay and gets a damn good deal. Put another way, we're just trying to help the bidding along. It does no good to anybody if you waste time bidding far under market value, and you feel really like a sucker if you overpaid by $50. There's a happy medium in there somewhere!
Hope that addresses some of your concerns. Now if only you can see my replies as I write them...
Anyway, by the time my original reply to cordelli gets posted this will just be a reiteration, but.. yes, the site automatically calculates the entire bidding strategy and does it within minutes. Waiting 24 hours would suck. People won't want to use software that sucks.
Also, what is probably not clear and that we need to message more clearly is that the bidding is actually done by your own computer (it's just driven by some automation software we wrote). So there's no issue of personal information getting stored on our servers.. all we do is set it up so that your own computer can do it. This makes it safer for everyone, since we don't have to store credit card numbers -- everything goes between you and Priceline. Unfortunately, we've been struggling on how to tell users through the web page. The standard reaction is "why would I give these unknown guys my credit card".. doesn't matter what you say, habits are hard to break!
To generally answer your questions, this site is pretty algorithmic (we're all from a very quant/software background). All the stuff shown to you - recent bids, rejected bids, the hotel popups, the bid data.. it's all been specifically tailored for your query.
Let's use an example. One thing that annoys me about seeing raw winning bid info or hotel lists is that there's no guidance behind that information. Say you see that someone won the Venetian in Vegas for $49 a night. But then if you look closely you'll see that they won on a Wednesday. Then what sucks is that the reader thinks that information is useful and goes to bid $49 (or worse, $45) for a 5 star on a SATURDAY night. So in this case the past information was actually harmful... both for Priceline, who doesn't sell the room, and the user, who goes away frustrated. In contrast, we've built statistical profiles for your query, which vary by day of week, and other factors, and present directly relevant information. A particularly interesting sidenote is that there are areas which suffer from the inverse problem (high weekday rates, low weekend rates), particularly business-centric areas. We identify them algorithmically and show only information which will help you.
Now, regarding the bid increments, this has been something of a dilemna. Yes, if you did a 5 star in Vegas, or another area where you theoretically have thousands of bids at your disposal, and you could theoretically start a bid at zero and walk it up a dollar at a time until you hit the target. But this will just piss off the user, who has to wait, and Priceline, who sees useless bids clog the system.
Being somewhat of a data guy I believe the solution lies in the data itself. What I mean is that once we crunch enough numbers for you and present you relevant information, we distill for you a market value of what an opaque hotel costs at a particular moment in time. At that point, the permutation bidding is really just a crutch -- wiggle room, if you will -- to just get a little more precise... since you should know approximately what it's going to cost, already! So given all the inputs: zones you've selected, past wins, rejected bids, what you've decided your lowball and final offers are -- the software calculates the sweet spot for what each bid should be. And by sweet spot I mean sweet enough for both parties: Priceline, who sells the room, and the user, who does not grossly overpay and gets a damn good deal. Put another way, we're just trying to help the bidding along. It does no good to anybody if you waste time bidding far under market value, and you feel really like a sucker if you overpaid by $50. There's a happy medium in there somewhere!
Hope that addresses some of your concerns. Now if only you can see my replies as I write them...
#9
As an avid PL user (75-100 nights/year) I am always looking for tools that will assist my bidding, both to lower the price I pay and reduce the amount of time I spend researching and rebidding. The site looks pretty fantastic so far. Some feedback:
- The site is excruciatingly slow to load. The design is obviously meant to mimic PL itself, but it ends up feeling like a slower, clunkier version.
- The listing of recent wins and prices is brilliant. Today I have about 3 screens I toggle between to get this information, and getting the price data out of BetterBidding is difficult at best. Love the way this is all at my fingertips here. When looking at the prices, especially in zones with many wins, it seems that maybe a graphical representation of the price range would be helpful. Would love to see a line with the winning prices/ranges to get a better understanding of the underlying data.
- It's entirely unclear what's going to happen when I put in my credit card, and that makes me not want to go any further. There needs to be a lot more explanation upfront if you expect users to put in a CC number. I would perhaps suggest collecting all other info (name, address, contact info) first, and then a separate page indicating what's going to happen with the credit card when the payment information is collected. "Autobid" is unclear, and this creates some worry that I'll be on the hook for half a dozen PL reservations if something goes wonky. I'd be more comfortable if the bid structure were laid out first and my confirmation was requested before bidding started, or even perhaps between every bid until I got comfortable with the automation.
- Solving this problem as a web app/service is definitely the right way to do it. Others have produced solutions that function as standalone apps, but I've never been interested enough to deal with installing them. I love the low barrier to entry here.
#10
rmiller774 , Dec 17, 2010 9:52 pm
I have made many dozens of PL bids. I want to be very close to the lowest PL acceptable price - not a higher "sweet spot" amount agreeable to and selected in the blind by Priceline and others. For instance, if I entered a low offer ($60) and a final offer ($80) how could I know that I would be treated correctly and receive the actual lowest acceptable PL room price when I am putting blind trust in two forces who strive to maximize profit from my bid, rather than to do their best for my wallet. I would need assurances.
Edit: BIDDINGTRAVELER's modifications to his system have assured me for what that's worth.
Edit: BIDDINGTRAVELER's modifications to his system have assured me for what that's worth.
#11
Quote:
Let's use an example. One thing that annoys me about seeing raw winning bid info or hotel lists is that there's no guidance behind that information. Say you see that someone won the Venetian in Vegas for $49 a night. But then if you look closely you'll see that they won on a Wednesday. Then what sucks is that the reader thinks that information is useful and goes to bid $49 (or worse, $45) for a 5 star on a SATURDAY night. So in this case the past information was actually harmful... both for Priceline, who doesn't sell the room, and the user, who goes away frustrated. In contrast, we've built statistical profiles for your query, which vary by day of week, and other factors, and present directly relevant information. A particularly interesting sidenote is that there are areas which suffer from the inverse problem (high weekday rates, low weekend rates), particularly business-centric areas. We identify them algorithmically and show only information which will help you.
So when I enter dates & zones and search, what you display as recently accepted bids aren't necessarily wins for those dates, but rather across all weekends (or weekdays)? That seems reasonable for some stays, but could be very misleading when looking for a high demand or special event date. For example, searching NYC, Times Square for 12/20-1/02, at the 4* level, you show a recent weekend win for $140. I'm assuming that's not actually for New Years Eve in Times Square; the cheapest Hotwire for that zone & level is currently over $300.Originally Posted by biddingtraveler
Let's use an example. One thing that annoys me about seeing raw winning bid info or hotel lists is that there's no guidance behind that information. Say you see that someone won the Venetian in Vegas for $49 a night. But then if you look closely you'll see that they won on a Wednesday. Then what sucks is that the reader thinks that information is useful and goes to bid $49 (or worse, $45) for a 5 star on a SATURDAY night. So in this case the past information was actually harmful... both for Priceline, who doesn't sell the room, and the user, who goes away frustrated. In contrast, we've built statistical profiles for your query, which vary by day of week, and other factors, and present directly relevant information. A particularly interesting sidenote is that there are areas which suffer from the inverse problem (high weekday rates, low weekend rates), particularly business-centric areas. We identify them algorithmically and show only information which will help you.
Is that right?
If so, how about adding an option for "search these specific dates only"?
#12
I gave it a try and it sort of worked...started at $30 and increased to $60, where it succeeded in buying a 4* in Anaheim. However, it never stopped and just hung at the $60 message. I had to close the window.
Also, no "contact us"?
QL
Also, no "contact us"?
QL
#13
biddingtraveler , Dec 18, 2010 1:02 pm
QuietLion, please contact me immediately via private message. I've also sent you email.
#15
biddingtraveler , Dec 19, 2010 12:00 am
Outstanding feedback guys. I'm humbled. And at the same time tired.. definitely a tough crowd to please.. 
Some updates.
- First, all you holdouts who are worried about credit cards: you talked, we listened. After one all-nighter marathon coding session, I think you'll really like the changes. In particular not only did we completely eliminate credit card collection from our pages, but we eliminated all address and personal info collection too. Now there's no excuse! Go try it!
- rmiller774's comments about bid increments: I should have been more clear explaining. Basically, bid increments will range from about $3 for really low prices (around the $30 range) moving up to around $5 for mid-priced ranges, (around the $70 mark), and if you're a high roller bidding at the $170 range of so for 5-star hotels, the increment is $10. Keep in mind that the increment may not be constant. Basically, the software would be doing pretty much what a human would do. We designed this to be a convenience tool for how normal people would bid, not an exploitative tool for Priceline. But I still stand by my statement that it will offer a superior experience to what you would do on your own.
- Congrats to QuietLion for nailing a $60 4-star. It was 40% off retail. I should have mentioned earlier that once you win we also calculate and display, inline, how much money you saved on a tax-inclusive basis. Thanks for reporting the bug too, which we believe we've fixed
- If you try us out: A user has reported that our software doesn't properly handle the case where you are already signed in to Priceline and you are using saved credit card information. As a temporary workaround, you might want to sign out from Priceline before starting. We'll be working on a fix for this issue as soon as we can.
I'll try to incorporate and prioritize all the other feedback that everyone else has been grateful enough to write!
Some updates.
- First, all you holdouts who are worried about credit cards: you talked, we listened. After one all-nighter marathon coding session, I think you'll really like the changes. In particular not only did we completely eliminate credit card collection from our pages, but we eliminated all address and personal info collection too. Now there's no excuse! Go try it!
- rmiller774's comments about bid increments: I should have been more clear explaining. Basically, bid increments will range from about $3 for really low prices (around the $30 range) moving up to around $5 for mid-priced ranges, (around the $70 mark), and if you're a high roller bidding at the $170 range of so for 5-star hotels, the increment is $10. Keep in mind that the increment may not be constant. Basically, the software would be doing pretty much what a human would do. We designed this to be a convenience tool for how normal people would bid, not an exploitative tool for Priceline. But I still stand by my statement that it will offer a superior experience to what you would do on your own.
- Congrats to QuietLion for nailing a $60 4-star. It was 40% off retail. I should have mentioned earlier that once you win we also calculate and display, inline, how much money you saved on a tax-inclusive basis. Thanks for reporting the bug too, which we believe we've fixed
- If you try us out: A user has reported that our software doesn't properly handle the case where you are already signed in to Priceline and you are using saved credit card information. As a temporary workaround, you might want to sign out from Priceline before starting. We'll be working on a fix for this issue as soon as we can.
I'll try to incorporate and prioritize all the other feedback that everyone else has been grateful enough to write!