posted on March 2, 2002 02:36:12 PM new
I'm tired of the meaningless stats that i see posted about "SELL THROUGH" all over by posters like bidsbids and the alter ego dimview. So i am taking a look at some specific categories that I am imterested in at Bidville. What I did was look at closed auctions for the past thirty days in each category. I compare the auctions that closed to the auctions that closed with bids. It must be accepted that many items are so called "warehoused" at Bidville, and Bidville also offers a wide variance of auction lengths. therefore, any stats based on current listings must be guesswork, and do not accurately reflect the "ACTIVITY RATE" at Bidville.
So here are the numbers I found, you're welcome to add your own findings.
posted on March 2, 2002 05:19:05 PM new
How about the books storefront category? One seller with 299,000+ books on permanent relist. Joined on 9/18/01 and has 8 feedbacks or 8 books sold so far. Looks like books are slow movers on BV.
A note: Some sellers other than booksconsignment.com have now somehow listed in the books storefront section. Either a change in BV policy or a glitch allowed this to happen or they are unusual hackers.
I thought I would never write this, but I have to side with Dimview's stats over yours in this case which I know is at the heart of this post.
First, stats can be calculated in many different ways and that is fine as long as the definition of the stats is clear. However when you start to compare one stat to another, they must use the same basis or the stats are meaningless.
In comparing sell-through rates for online auctions, weekly sell-through has become the unofficial standard. This is because most auctions at Ebay are 7 days and since Ebay is the standard in the industry, the sell-through at most sites are compared to it.
Since most other sites allow auction lengths of at least 10-14 days and autorelists, using closed auctions is not usually accurate in developing a weekly sell-through rate. Therefore the number of listings is usually substituted for the closed auctions. I have no problem with this because what you are wanting to see if the % of listings being sold in some defined time period. Also when Bidville posted its own "weekly" sell-through rate, they used this method so even the site accepted the method.
The flaws with your method are significant. Mainly because it ignores the vast majority of listings on Bidville. Since the vast majority are on some sort of auto-relist, these are never counted in your method until the last auto-relist ends. Why? There is no difference between an auction ending and not relisting and an auction ending with an auto-relist. If a bid had been placed on the auction in either case, it would have ended. With your method, an auction running for up to 21 months with the 20 auto relists is ignored until the very end. This is very misleading on actual sales.
What I have listed above is talking about comparing sell-through rates between different sites. If you wish to use a different method to evaluate your own listings, that is fine.
Please, just don't invent a very flawed statistic that further enforces the concept that Bidville chearleaders can't stand reality.
posted on March 3, 2002 06:14:06 AM new
This is not selling used cars. You can not manipulate the figures to suit your tastes. The reality is, for every 1,000 items that close at BV ( and to be fair, most other third tier auction sites ) only 15 auctions successfully close.
That is why there can never, ever be any form of listing fees at those sites. The owners of these sites can never, ever hope to make any money with at least a 5% FVF along with whatever they can get out of the sellers for listings extras such as memberships or features.
posted on March 3, 2002 09:23:29 AM new
stavecards,
I understand that the method dimview uses is the "unoffical" method that is generally accepted. The points you make in your arguement, are exactly many of the points I have about the flaws of that method. My method counts each auction once, not many times.
(unless of course it is a DUTCH auction, which the generally accepted methodology ignores as well.)
My method could reasonably be applied to all sites equally, while the current accepted method does not.
Are you surprised to learn that the BVCS may have erred in using that method? Is it possible that they saw the method's flaws? Maybe that caused them to discard it? I don't know.
I refuse to believe that because the PTB claim that the other method has always been accepted that it is right. History shows us otherwise in many cases.
In the US, are speed limits are posted in MPH. In other countries they are posted in KPH. Who is right?
You are of course welcome to accept the methodology you wish. Thank you for your comments.
bidsbids,
If you are going to make accusations please prove them.
>>>>>"You can not manipulate the figures to suit your tastes."<<<<<
These figures are the actual figures I got from the site. Please, show me where I changed, or manipulated them.
TO ALL,
The one thing I do not know, is if when an auction closes and relists at Bidville, does it show as a "closed auction"?
I DO know that when one of my auctions closes and relists, I get a "Results" notification by email.
[ edited by jimhhow on Mar 3, 2002 09:25 AM ]
posted on March 3, 2002 11:13:48 AM new
"...BTW, maybe they just know how to use the Bidville listing page. Need any pointers?..."
=======
sliding just a bit off topic to address Jim's post script...
Actually it shouldn't be possible to list in the storefront. That's one of Justin's many sweetheart deals, and is considered a site within a site, and regular users in theory shouldn't be able to access it for listing.
So I'm very curious about how this seller was able to do it.
posted on March 3, 2002 11:19:00 AM new
I have researched the sell thru rate at bidville and came up with a 1-3% sell thru across the board. Some catagories are ok, while others are basically dead.
My post script was a response to the post above it by bidsbids wondering how people do it.
As I tried to express to bids, you can list there from the drop down selections on the bidville listing page. That is a subcategory selection.
[ edited by jimhhow on Mar 3, 2002 11:26 AM ]
I don't believe that this is a case of who is right. As long as the math is done correctly and honestly, and the parameters are clearly stated, then both numbers are "right". The difference is which method provides the more relevant data. To me, the weekly sell-through rate provides much more meaningful comparison data than your method.
When I look at a weekly sell-through rate, I read it as the percentage of total listings that closed during the week with a successful bid. A number like that with a defined time period gives me useful information.
Before I address your method again, I will answer your question above. On Bidville, if an auction closes and is auto-relisted, I don't believe that it shows as a closed auction if you do a search. The auto-relist keeps the same auction number and no end of auction e-mail is sent except an e-mail is sent if you had a feature enhancement on the first listing. I also don't believe that an auction shows as closed if was relisted using one of the two manual relist methods. Using one of the methods also does not generate a new auction number. As long as the auction is not issued a new ID number, I don't believe it shows as a closed auction in a search.
Because of these facts and others is why I find many flaws in your method. The first flaw is that there is no time period for comparison. I don't know if the average length of time of the closed auctions was one week or one year (which is still possible on Bidville). Using the Bidville extreme (30 days listing times original list plus 20 auto relists), a listing could run for almost 21 months without showing as closed. If it happened to receive its first bid on the day before the expiration of the last relist, it would count the same in your comparison as an auction that ran for only 7 days and closed successfully. Also theorectically your method could always show 100% sell-through. If every listing was either auto-relisted or manual relisted keeping the same auction number, then only auctions with bids would show as closed.
As far as Bidville stopping the posting of sell-through rates, I believe it was another example of the site starting something and then quiting (like the weekly newsletter). If was only a problem with methodology, they could have made the change and kept on posting the numbers.
I know you want to show Bidville in a good light, but it doesn't do any good in the long term for that good light to be false.
posted on March 3, 2002 02:44:15 PM new
Stave,
>>>"To me, the weekly sell-through rate provides much more meaningful comparison data than your method."<<<
That is fine. You are entitled to your opinion and I can respect that.
>>>"When I look at a weekly sell-through rate, I read it as the percentage of total listings that closed during the week with a successful bid."
I disagree, flatly. what you look at is the number of auctions that closed with bids. If there was a reserve not meant, it does not mean that the auction was sold.
I believe there are as many flaws with dimview's compariaons as with mine when applying a formula designed for one sight to all of the others. I believe that my approach could be used on all sites in a more equitable manner.
Due to the wider latitude of auction lengths allowed at Bidville alone is one reason that dimview's comparisons are not equitable, in my opinion. But, if every auction site offered exactly the same thing there would be no need for more than one.
As far as the relists, there is one common factor with the auctions. All auctions close, eventually, not all auctions relist.
If someone wants to "warehouse their auctions for 21 months, that should not be reflective of "activity" at a site. My numbers look at something different then what dimview's numbers do.
for Example, dimview uses the current number of listings overall at Bidville versus the number of auctions sold in a prior period, that is not a stable basis for comparison.
I am not trying to argue dimview's numbers for what he is trying to show. That is a sell through rate versus the entire listing body of the site.
What my numbers look at is the activity level in indiviual subcategories over a thrity day period. I think if someone is considering an auction venue, they want to know if there is or is not activity. The numbers being skewed by the warehoused items fog that.
In one subcategory alone,
Sports Stuff : Trading Cards : Baseball : Singles : 90's
Which is considered one of the "warehouse categories, almost 5000 auctions received bids in the last 30 days. Whether they sold or not due to reserves, is another question. The point is that is the activity that occured in that time period.
Bidville is cetrtainly not the largest site, but almost 5000 bids in that warehoused subcategory, in a 30 day period should not indicate a "DEAD" site as some have pronounced it.
Dimview's figures are what they are, and show what they show. My figures are meant to show a different aspect. You and everyone else must decide the yardstick you will use to judge.
>>>"I know you want to show Bidville in a good light, but it doesn't do any good in the long term for that good light to be false."<<<
I want to show Bidville in the true light.
There are too many willing to jump up and point at every little thing to say the site is dead, and they must change, the owner is absent. I mean come on, no wonder the cheerleaders get upset. These people purport to be showing things like they are, but there are many positive things, and I never see them point that out.
They show a negative BV thread, the cheerleaders show a positive one, they look for something in it to apply a negative spin.
WHY??? I don't know. Are they on the payroll of another venue? At least that would make sense.
Rob, I understand you think my method is flawed. So is dim's method.
As far as the method to calculate the sell-through rate, we will just have to agree to disagree. Too many philosophical differences to resolve.
As far as Dimview's method, I have never had a problem with his/hers (does anybody know for sure?) calculation of the weekly sell-through rates. With the data available to an outsider, I thought the calculations were pretty accurate. However my problem with Dimview was the interpretation of the meaning of the stats. He/she reached, and still does, many conclusions which were not logically supported by the numbers reported.
Now, if your aim is to show that there is activity on Bidville, it might be more impressive to just report the raw successful closed auctions numbers like the 5000 number listed above. When Bidville was reporting the numbers, I probably paid more attention to the trend in the raw numbers than I did to the %.
On a slightly unrelated point, I would support efforts to curb the ability to "warehouse" items on Bidville. I have always been in favor of reducing the number of autorelists to five or lower. Myself, I only use three. I only use that many to maximum use of the photo gallery $$. It has been my long-term belief that the vast number of nickel and dime items create so much clutter that better items are lost in the shuffle. This is definitely the case with sports cards. I made one proposal to the site to create a special way to list the commons for set completions that would both enhance their sales and reduce the clutter from the more high valued cards. So far, I have not received a response.
posted on March 3, 2002 03:49:30 PM new
I agree that the sell thru rate is hell to figure if Seller A uses 3 day durations and Seller B uses 30 day durations. There needs to be a common time period to figure out a universal rate.
BV has 2 very unique listing situations. The 30 day duration and the 20 auto-relists. The ease of using the 30day duration and the 20 auto-relists together at the same time is very appealing to sellers with very commonplace items that do not want the hassle of constantly relisting their auctions. It's almost like one of those radio stations that is totally automated and doesn't have disk jockeys and very few engineers.
posted on March 3, 2002 04:45:36 PM new
I guess I personally would not be ooposed to limiting relists to five. I rarely use any more than three myself anymore. And I don't even set all of my auctions to autorelist.
Now, I am not trying to nitpick, Stave, but let me ask you something.
How would you feel if someone said this about the items you list?
>>>>"It has been my long-term belief that the vast number of nickel and dime items create so much clutter that better items are lost in the shuffle."<<<<
I understand that you are in favor of finding a way to enhance the quality for both types, but I see many speak with disdain, at least you didn't call it junk.
Errors have been found with some of dimview's numbers sometimes. but anyone can make a mistake. That is all beside the point, though. My aim is not to replace the SELL THROUGH numbers. I believe the ACTIVITY RATE actually would help define them.
The ACTIVITY RATE should only be used for subcategories. I have seen numerous examples already where trying to apply a SELL THROUGH rate to a subcategory would be foolish as there can be less items currently listed than were bid upon. This can skew the numbers. I would not use the ACTIVITY RATE to try to show as an example site wide, that again would not be an accurate reflection to portray.
THe SELL THROUGH rate was developed for EBAY when there was basically only EBAY.
Now that there are other players and varieties, other methods of figuring activity are needed.
Niche sites should be able to count their Niches using a yardstick that does not discrimminate against them being a Niche site.
Unlikely, but conceivable that maybe a certain category does better on Bidville, or some other Niche site than it does on Ebay.
I have no problem with your opinion, Stave, as you say, we can agree to disagree. I just have the feeling that everyoneas yet doesn't understand what I am doing, hell, I ain't sure I understand it all, but I knew something was wrong by seeing the same stats posted continuosly, and they just don't make sense.
posted on March 3, 2002 09:17:25 PM new
jimhhow,
I have reviewed your stats and they are quite flawed. On thing that you have left out is the auctions that ended without a bid and relisted. Those auctions do not show up on a closed auction list as they keep the same auction number until it is truely closed. If you really follow the numbers, you will find that less than 1% of the auctions except junk commons sports cards end with a bid. To make your numbers correct, you would need to follow all of the relists as well and record them. Bidville also tracks them and admits to the low rate as well. I can see at the surface where you get your facts, it is quite clear that you don't have them all. You do report factual information, just not all of it.
bp
posted on March 4, 2002 01:18:24 AM new
Wow barparts, you managed to mimic enough of everyone elses posts that you sounded like you knew what you were talking about for a minute.
Jim, I see your point. But maybe micro-analyzing sub-sub-categories is just a little too deep of an analysis?
Personally it bores the hell out of me, all the stats talk back and forth. One has only to glance at BidVille to know the sell through rate is extremely low. What else really needs said?
1. BidVilles sell through rate is low. 5% or less per category.
2. BidVille does not advertise nor does it sell advertising space on it's auction site.
3. Yes, there are thousands of 5 and dime sportscards on BidVille. But, what can you do? They have as much right to list as anyone else.
4. Yes there is a books storefront.But, what can you do? They have as much right to list as anyone else.
People can support or bash BidVille and it doesn't make one iota of a difference. It just sits there, plugging along, day after day. Bidders and sellers come and go and nothing ever changes.
Nothing that matters anyway.
It's all a matter of personal choice, hang around..build up a repeat bidder base ( or trusted seller base) and don't expect much in the way of perks or pep talks from BVCS. Or, don't.
posted on March 4, 2002 02:50:51 AM new
There must be many wise tribal leaders in your family tree Kodi. Well said.
Perhaps Orlando is just trying to hang on and make a few bucks on his PayPal lawsuit? With all the rumors about a new PayPal online auction site starting soon it seems like there were would have to be a settlement of some kind to that old lawsuit first. Even that could work against the BV Faithful. Let's say PayPal gives Orlando $250,000 to settle the suit. No more reason to keep the site running as it only brings in a couple of thousand a month at best and does not have a rosy future. Out comes the 'Out of Bussiness' sign on the door.
