Clear information regarding shipping time

bluedragon

This is related to the shipping time issue but I have an independent suggestion that warrants its own thread. In my opinion, if implemented, this would:

1. Better inform the customer about the average processing time and what to expect
2. Allow customers to choose whether they use or not stores with long processing times without the need to hide those stores from the public view
3. Partly bypass the need for feedback and mitigate the fact that few customers provide feedback

What I propose displaying a very objective piece of data in a prominent place on an individual store page (below feedback)- AVERAGE PROCESSING TIME. This is calculated by the website and not chosen by the store and it shows the average time in days between an order being placed and being shipped. This information would also be available on the "Stores" web page next to feedback and customers would be able to sort based on it just like with feedback.

This would clearly benefit stores with better processing times and encourage others to up their game, ultimately benefiting the customer. It would also be more informative than the "ships in x days". I have the 2-5 days option selected just in case but most of my orders ship in 1-2 days.

Would this be possible/desirable? @Lawrence

Stragus

I can see a few problems with this.

Twice, I have had a buyer send a message similar to: "Wait, don't ship the order yet, I'm going to place a new one and we can combine shipping".

Another one, after placing an order, requested tips how to build a gear train for high torque, and we exchanged message for a long while until he agreed his order was complete.

When some parts are missing due to inventory errors, some other buyers prefer to wait until I acquire the missing parts rather than take a refund of (2*MissingValue)+$1. That also can take some time.

Basically, if what you propose were implemented, a seller would be very reluctant to provide good customer service (like in the examples above) in order to protect his Average Processing Time rating.

bluedragon

Problem 1 could be solved by excluding combined orders.

I suppose problem 2 is uncommon. The advantage of AVERAGE is that uncommon events have little weight overall.

I would be very surprised if any seller would prefer to alienate an existing customer by providing bad service- he would risk negative feedback. Would you, as a seller, be reluctant to provide good service in order to protect APT? I would not.

pumbaugh

What if a seller is marking the items before the actual ship date to improve his score? Not all methods are tracked or able to be verified of ship date.

bluedragon

There is fraud everywhere it appears....

95% + of my orders have the shipping tracking number uploaded to BO.

Fraudulent unscrupulous sellers doing what you described would be caught over time because of negative feedback from buyers. Buyers receive an email letting them know the order has shipped. If the order arrives late and they see a discrepancy between stamp date and shipment notification date they will not be happy. I suppose unhappy customers generally provide negative feedback. I wouldn't know of course ;-)

brix4kix

this is the same issue that the other thread is having debated and abused, there is already stores that to get there inventory out there they have put 3-5 shipping and changed there front page to say expect delays for some bogus reasons. so what ever is done there will be those that will cheat and deceive, to get around the system.

that's all I have to say.

bluedragon

As I mentioned earlier, sooner or later the abusers will get caught because the postage and BO shipping dates do not match. The current system, where stores pick a processing time frame, is completely based on trust as any store can set it to 1-2 days and ship 3 weeks later.

Lawrence

I do actually use this internally, I have previously considered exposing it, but there are too many variables that can cause it to be inaccurate, like customers asking for the order not to be shipped, for various reasons, or payment not being received ( manual payment status is often inaccurate)

If stores do not keep to their displayed dispatch times, this is often relayed to me through negative feedback or issue report alerts.

BrickItOn

95% + of my orders have the shipping tracking number uploaded to BO.

It must be nice living in the USA.
95 of mine orders are shipped WITHOUT tracking, because it simply costs a fortune here (1 minifig to the USA=16 euro with tracking as example)

Nice to have another tread about the same topic, but I still wonder what PROBLEM has to be solved in the first place? Why should shops that ship in 6+ days or more be punished when a costumer is probably happy to wait some extra days? I ship almost all orders within 2 to 5 days, but I would like to have my setting to 6-7 or 8 (14 is a bit to long) just to make sure the customer knows I am not the fastest shipper (abroad a lot)

Again: I don't see anything that needed to be fixed at all.

But anyhow: I make good sales here, so that makes me am happy to be here. I can't say I am happy with the direction this site is going however (and the catalog). I will look at the site updates to make sure my settings are correct to be visible at all, probably eat some neutrals about silly things without have the option to respond or something and give all my customers positive feedback because it is the only option.

other then that: have fun all, I won't be participating in site improvement or things like that

ps: the news on the front page is now 1 year and 2 days old.

Markyd7

I do actually use this internally, I have previously considered exposing it, but there are too many variables that can cause it to be inaccurate, like customers asking for the order not to be shipped, for various reasons, or payment not being received ( manual payment status is often inaccurate)

If stores do not keep to their displayed dispatch times, this is often relayed to me through negative feedback or issue report alerts.

Could you not remove the worst 10% times and best 10% from the equation. This would help with the odd order that might take longer.

bluedragon

I do actually use this internally, I have previously considered exposing it, but there are too many variables that can cause it to be inaccurate, like customers asking for the order not to be shipped, for various reasons, or payment not being received ( manual payment status is often inaccurate)

As stated before, I understand that there are variables that could affect the accuracy of the data BUT (and I say this with no stats), I am positive that the number of orders affected are a small percentage and therefore can not significantly skew the value. I have had about 450 orders and may have had 5 orders delayed by one of those reasons. Others may have different experiences but ON AVERAGE the number will reflect well how fast a store ships.

Such information (real data) has to be better than the current situation: a statement by the store announcing how long it takes. Right?

bluedragon

Could you not remove the worst 10% times and best 10% from the equation. This would help with the odd order that might take longer.

or provide a measure of variance- 2.5 +/- 0.3 days, for example

BrickItOn

1 last thing: Please keep in mind that this (as proven by the other topic) 1 "small" change put some stores completely and 100% out of business. They will probably turn their back on BO as well if they don't read the forums.

bluedragon

Hence my suggestion. All stores would be visible but customers could sort by processing time and choose what was acceptable to them on that particular day.

One day a customer may be be buying parts for a slow moving long term project and not care about shipping time. Next day they may be buying for a show and need stores that process orders in less than 2 days.

pmitch

Using the median (middle) rather than the mean (average) would minimize the outliers' affect and would be the most meaningful stat in this case. This is just another function that can be called, so it is available with minimal extra work.

I would think having the store's stated objective (e.g., 3 - 5 days, etc) along with the actual median would provided the most info for the prospective buyer.

My only hesitancy is that the median statistic would not automatically consider only business days. That is, to exclude weekend days, some custom code would be needed. Since holidays vary widely by country, excluding these days might be beyond the reasonable scope.

bluedragon

Median- you are right. Weekends, holidays, snow days, etc... In my opinion, these things should not be factored in as you suggest. The advertised processing time is absolute and does not refer to holidays and such. Besides, although some countries have more holidays than others, weekends happen everywhere so the playing field will be level.

pmitch

Yes, and if that is explicitly stated, behind a help screen - not clutter on the main screen - then everyone is equally empowered. Buyer / seller both have full info. One problem with the current setup is that the weekends don't seem to be directly addressed. I have that on my own info page, but it would be nice if everyone counted the same way - again, better for both buyer and seller.

pmitch

Please note that I am not meaning "empowered" as a social positivism. When one party has more power because the data (of performance in this case) is not transparent, and rather the seller can STATE what the policy is without being (transparently) accountable, then the buyer is at a disadvantage in making a decision. I am using "empowered" to suggest that transparency approaches balance for the buyer AND for the seller who wants to provide exceptional service. Showing the median service time with a timeliness indicator rewards the seller who ships in one day and exposes the seller who STATES a 1-2 shipment, but performs at a 4 day service level. Again, this helps everyone who SHOULD be rewarded (empowered).

unbrickable

why not use a standard deviation where you only use a certain range of the confidence interval to determine the "average" shipping time ? btw i would like this funtion.

pmitch

Yes, that is fine, but it really just achieves what switching from mean to median does. That is, the impact of outliers is lessened. It would more accurate, but possible more confusing for most. It also has the benefit of being able to be expressed with one number rather than a number and a +/- range. The correct (and careful) use of statistics is BEAUTIFUL (imho), but not necessarily appreciated by the general audience.

Teup

I fully support the ideas but also agree with some of the problems. There's one more problem that I haven't seen mentioned yet; If a seller is temporarily away/busy/hospitalised/whatever and they set their shipping time to like 2 weeks for a period of time, there will be some problems with the statistics. When they're back in business and the seller claims to ship every order the same or next day, customers will look at the statistics and think the seller is lying. And reversely, if a seller is away for a while their statistics will still display they ship the next day so a buyer would expect fast shipping but won't get it.

I've been away for a while and shipped orders in an average of like 10 days time for a while but usually when I'm fully in business ship the same day. So when I'm set to 6-14 days I ship in 10 days, and when I'm set to 2-5 days and fully in business I ship the same day. I think I deserve some creds for that but a statistic like that would not provide it..

Just wanted to add this problem point to the agenda, still really much like to see some form of this implemented!

Possible solution: Make the statistic not about how many days it takes, but make it about in how many % of the cases the seller meets the shipping window specified at the moment the buyer placed the order!

bluedragon

Unless you are referring to a new store, a couple of weeks of slower processing time should not affect the median too much. In the cases where there was significant deviation from the median for a period of time and a warning is displayed as suggested above (or not), the store can address the issue by having some text on their page.

Absolute data are generally more informative than relative data, therefore I would still prefer actual time to better inform the customer.

Teup

Unless you are referring to a new store, a couple of weeks of slower processing time should not affect the median too much. In the cases where there was significant deviation from the median for a period of time and a warning is displayed as suggested above (or not), the store can address the issue by having some text on their page.

Absolute data are generally more informative than relative data, therefore I would still prefer actual time to better inform the customer.

True, true, but I think it's safe to say that most buyers don't read so well, and it kind of defeats the point of the statistics if a custom message on a splash page overrules it (after all, the statistics were supposed to show us what's "really" going on with that shop ). I think that if a shop has a statistic saying "this guy ships in 1 day!" and/or is favoured in the search results in some way because of it, people will go for that and not read so well. Also if a seller would change their shipping term regularly (say, every other month because that's just their business model, they work off and on) the statistics would not do them justice either as praise or warning.

Trying to think of some more ideas.. maybe exclude the statistics for orders placed while the shop has the 6-14 days shipping setting, and shops that currently have that setting are not getting (favourable) statistics displayed. So, leaving 6-14 day shipping out of this whole innovation.
Or maybe record statistics per setting, like, if a shop is set to 2-5 days shipping, show info on how long the shop takes on average to ship orders that have been placed while the shop was claiming to ship in 2-5 days. If the seller then changes his setting to shipping 6-14 days, show how long it takes on average for past orders placed while set to 6-14 days.

It all sounds a bit complicated perhaps, but I guess we all agree some sort of data modification has to be made to inform the buyer accurately, and it's all invisible calculation..