Can we finally follow a more conventional sorting system?

Enza3D · 2016-10-18T22:24:44.000Z

My issue is that this new ranking doesn’t take into account what the customer wants, only the speed a Hub can deliver it at. If that’s all the customer wants and they need the print asap they should be able to filter by that. The sorting should not by default be based only on speed, and rank Hubs with one-day turnaround automatically on top. If a customer is just looking to print something with little care for how long it takes, you could very easily loose that customer because now your Hub is no longer one of the first few options. But what if a customer cares more about how well the Hub delivers orders, or their price? This is what I take issue with.

This sorting issue really is getting ridiculous. This is not a novel idea that Hubs is pioneering; Amazon, eBay even Etsy all allow you to sort by price, delivery time, number of reviews, and rating with the default sorting being a combination of reviews and ratings. Here, it would make sense to be able to sort by cost, distance (for pickup), rating/reviews and turnaround time. I don’t understand why this filtering functionality can’t just be applied here, and leave the default sorting as it was originally.

In addition, if this is how it’s going to be Hubs needs to implement some kind of one day cutoff time. It’s not fair to a customer or a Hub if a customer uploads a model here at 7 pm local time expecting to get it the next morning. A Hub should not be penalized for not being able to fulfill that expectation (as it currently stands, if someone submits an order to me know, Hubs states it should be done by tomorrow), and the customer should not have that expectation in the first place. Standard cut-off time for next day at any company I’ve worked with is noon local time, meaning if an order is not placed and paid for before 12 pm on 10/19, next day turnaround means the print’s expected completion date is 10/21.

Left · 2016-10-19T08:15:52.000Z

I totally agree.

I closed my hub some weeks ago, and this news isn’t changing my decision at all.

At first 3dhubs was a place were people could experiment with 3d printing.

Customers were happy if you could help them with their projects.

Nowadays everything has changed. It feels more and more like an enormous pressure for a very small price.

It is not worth it anymore if you ask me.

Hoffa · 2016-10-19T08:56:13.000Z

Thank you for posting this! I just noticed the same, and it’s a bit sad. Sure I can usually fit a customer job in and get it done even for same-day pick-up, but I’d rather play it safe and surprise them with an early completion than promise 1-day and have to let them down because I had another print that took all night, or something.

wirlybird · 2016-10-19T11:05:09.000Z

I agree, under promise and over deliver.

cobnut · 2016-10-19T11:12:14.000Z

@Enza3D I totally agree with the ranking issue, but I also agree with @AndreasDesmedt that if a customer needs a fast turnaround, they should be able to find Hubs that can deliver it. As others have suggested, it would seem logical to provide sorting options - an additional “question” beyond layer height and material, asking for urgency.

Having said this, I think most people will want things as soon as possible, even if they don’t need them as soon as possible. I like using Amazon Prime because I get everything the next day, but in truth, many things I order sit in my office for days before I have the time to actually use them.

However, I do think that if there can be only one sorting order, it should be based upon reviews and number of orders, not speed. If a customer needs next day delivery, they can look through the suggested Hubs to find one that offers this, but their primary sort order should ignore turnaround time.

Filemon · 2016-10-19T11:15:00.000Z

Hello fellow Hubs,

Thanks for the comments, some good ideas which I’ll feedback to the product team. I also understand the concerns, so allow me to quickly elaborate;

The shipping service update being discussed here is only about transit time, from shipping to delivery. So print quality in relation to overall turn-around is not part of this update.
The default sorting is “Best Match” which takes into account many different aspects of being a Hub (more here).
We do plan to allow customers to choose to sort by “Delivery” time as well, next to “Best match” and “Distance”. But we’re not replacing the Best Match ranking in the checkout. Sorting by price is also on the roadmap.
It’s true that some customers are not sensitive to turn-around time but more to price or service, they can keep using the “Best Match” sorting and later on sort by “Price”.
Finally, in order to keep 3D Hubs competitive and grow the platform, and driving more orders to Hubs, we have to keep adding new services. One heavily requested service by customers is fast shipping.

Hope that explains. Please let me know if any questions remain.

Best,

Filemon

Not1ofthepack · 2016-10-19T11:34:02.000Z

From a customer view point, “fast shipping” translates to fast printing. “I see you have 24hr shipping, here is my order… ‘response your print request will take 38hrs to print.’ What!?! It says 24hrs… OK do it…” Order is completed and rated down because it is not with in customers hands in 24hrs. Or they are not willing to pay $44+ for over night shipping. That is not even taking into account the poor designs or incomplete files. Or 2 hrs later oh I made a design change. I want this instead. I could go on, but seriously the system is not even close to being balanced at all.

Not1ofthepack · 2016-10-19T11:38:00.000Z

Agree

Filemon · 2016-10-19T12:05:03.000Z

Ok great, glad to hear the first part is sorted.

About variability. I guess you can capture a lot of the distance variation with different specified shipping options (dedicated to regions). Weight is a bit more tricky. However, the majority of prints is below 1kg. For very big prints you should have a bit more margin to play with. If an order is really massive, I recon the customer will understand special shipping is needed. For me, that has worked so far.

Cheers!

Enza3D · 2016-10-19T12:31:57.000Z

You are correct and I think we’re on the same page here. My issues was never with speed being a sorting option (it should be) just with how it was implemented. The default sorting issue appears to have been corrected, and maybe it was just a bug as the update was rolled out. I just want customers to have the easiest time getting what they want, and Hubs to find a balance.

rkelly · 2016-10-19T13:22:20.000Z

Why is it that “keeping XXX competitive” always means loosing sight of the main user ship in favor of more profitable options?!

Please take a page from Makerbots book. They had a fantastic user base who loved their product, they WERE dedicated to putting out the best product they could, and they did a great job with the rep 2 and 2x, then they lost sight of who there users were and started going after the more professional and high end pro-sumer markets because they thought it was more profitable. They completely alienated their main user base that they had established and were unable to penetrate the professional market. No professional shops use Makerbots because they are too small and not reliable enough, and they’ve alienated the maker community to the point where this year at the World Maker Fair just the mention of a Makerbot was scoffed at by everyone unanimously.

It feels as though 3D hubs is approaching a fork in the road, one side is makers and the other is professional prototypes. I don’t think you guys are even in the same stratosphere as professional prototyping shops, The company I work for does hundreds of thousands of dollars in prototypes per year, and we don’t, and never will, use 3d hubs. Cost is not a motivating factor, neither is speed. We want accountability and quality. I can call up my contact at protolabs or any other professional prototyping company and talk to the engineer that is going to be making my part. On hubs, its kind of a crap shoot to whether or not a hub is run by an engineer with some free time or just some person who bought a printer.

For most personal projects this distinction is not a big deal but for the professional world that is a show stopper. And to be frank its not worth my time to sift through the hubs to find and try one that looks professional, I would just rather pay the premium and not have to worry about it.

Please pick a user base and stick with it. You have a loving maker community that leans on you for low risk prototypes and personal projects. IF you seriously want to go up against the big boys of prototyping and professional work you need to up your game by several orders of magnitude.

Enza3D · 2016-10-19T14:21:59.000Z

While this appears to be the case now, I know yesterday (at least during the afternoon) default sorting was based on shipping speed. This was brought to my attention by a customer who was trying to place a repeat order and couldn’t find my Hub, as well as by another Hub. I’m happy that’s not the case anymore.

I also have to agree that if a customer sees overnight delivery, they will expect one day turnaround. You guys will have to make it explicitly clear to them that they are mutually exclusive and selecting overnight delivery does not guarantee one day turnaround.

Regarding shipping, its impossible to set a hard cost on overnight delivery, because it’s based on weight and distance. Is it ok to just offer it but not set a hard cost? Otherwise, I’m going to have to set it ridiculously high and that probably won’t be right for a lot of orders.

Enza3D · 2016-10-19T14:25:43.000Z

I suppose it can be handled on a case by case basis as well; a customer who lives in New York can get 1 day delivery from just USPS, which is much cheaper, whereas someone in California would need to use FedEx/UPS. I’ll have to play around with options, but it may be beneficial to notify customers when they upload a print to the main search page that quoted shipping rates (unless for flat rate) are subject to change.

kolbers3D · 2016-10-19T23:22:55.000Z

Hello!

I think you are completely right Enza3D. It should be sorted by reviews and prices. Who cares about one day delivery or order completion if the price is too high and/or the quality isnt good? Of course its good to add the option to sort by completion and delivery time, so people that need it fast can find a suitable hub. But the standard way should be sorting by somekind of reviews in combination with pricing!

Thanks

Perry_1 · 2016-10-20T21:51:41.000Z

“Best Match” - whenever I am not the best match, I feel insulted. Especially when I actually am the best match, but someone else is labeled best match. Rude.

That being said, can someone explain what is going on here? Is the ranking now changed so faster shipping increases a ranking? Can someone explain it to me? I have no idea what the actual change is…

Iceman24k · 2016-10-21T01:35:50.000Z

Hey wait the problem I noticed is the clock seemed to start when I accepted the order!!! so the client paid a DAY LATER and BOOM the part was DUE???

We need the "countdown to start after the payment clears please.

Enza3D · 2016-10-21T01:55:03.000Z

I noticed this too, and (hope) it’s a bug that they’ll work out. If a customer doesn’t pay for an order, that print is not being started.

AndreasDesmedt · 2016-10-21T06:01:29.000Z

I still think the best match option sucks. When I look up Belgium, I’m way under more expensive hubs, with way less reviews, further from the center of Belgium and longer trunaround times. Thats not really a better match than me I think…

Perry_1 · 2016-10-22T14:44:30.000Z

I do not like this statement that you have to pick two of three- quality, speed, cost. This is not necessarily true. My hub does all 3.

Perry_1 · 2016-10-22T14:56:53.000Z

Please stop linking to this page of what makes a hub a “best match”.

1. The criteria is weak. What makes a hub a “best match” has far more factors than are listed on this page. If this is all the criteria used in “best match”, then the algorithm is very weak. Additional factors, such as material offered, colors offered, prints completed, and a host of many other factors are not reflected on that page.

2. I have spent an inordinate amount of time ranking pages by available information, based on what you show on that page. Based on the limited factors shown on that page, I have ran multiple scenarios on hubs. Almost always, based on just those factors, 3dhubs recommendation for “best match” comes back differently with my simulations than what shows as “best match”. Now, I know 3dhubs algorithms are “secret”, but that does not make them more accurate for “best match”.

As you know, I hate the “best match” terminology, and will continue to push for this important change in terminology.