I didn't see anyone mention the McMaster-Carr website [1]. It may not be the "densest" out there, but it's clean, functional, and nicely presents a lot of information at once.
- initial ecommerce site was a mess (basically a page-by-page recreation of the catalog?) which saw minimal usage
- the redesign, which focused on usability --- notably reduced cognitive load --- resulted in an immediate uptick in orders which grew markedly for a long while until it represented the vast majority of their business EDIT: and also optimized for repeat orders on a schedule
If someone could find that, or a better writeup, I'd be grateful (it's _not_ the Medium.com article) and this page: https://iacollaborative.com/work/mcmaster-carr/ is just a mentioning by the company which did the underpinnings, not the overall architect. This link is decent: https://www.bedelstein.com/post/mcmaster-carr
There was of course previous discussion of this here:
https://news.ycombinator.com/item?id=34000502
Video on why the site loads so fast:
https://www.youtube.com/watch?v=-Ln-8QM8KhQ
(which is from the Medium.com article)
A fly wheel of benefits.
Those numbers could be anywhere, on completely unrelated things. They are not a good search query.
For my McMaster example, google gives 9 results, none of which are the McMaster site. That not specific enough? To be fair, I believe McMaster to be fairly protective of their catalog.
At least their part numbers are fairly recognizable - they are usually about 10 characters long, all numbers, with an "A" near the end. That's usually enough to get me to check the McMaster site first.
On the other hand, if you train people that if you want to use McMaster part numbers, you have to use the McMaster site... once you have a customer, as long as your site and inventory don't frustrate them, you have a customer for life.
You're sacrificing inbound for retention, in a highly measurable and testable way, for your unique audience and/or subsets of that audience. I have no doubt this is by design.
Hypothetically, if you make $1 in profit on your product, theory says that some competitor will bid up to $0.99 to secure that sale and if you don’t bid this amount also, your sales will suffer.
The end result is that Google and Facebook end up consuming all the profits for a large number of businesses online that have to survive by advertising, which explains Google’s immense profit margins.
Assuming what you say is true, this is truly a ballsy move by McMaster. Betting that their website is unassailable by their competition and thus such a value-add that they can forgo playing the losing game that Google and Facebook has setup is brilliant. I have such respect for that.
<meta name="robots" content="noindex, noarchive, noimageindex" />> McMaster-Carr has a Systems department which handles internal software development, including the website. I actually worked on a lot of the front-end functionality (among other things) during my tenure there and we had 1 person who was essentially the UX lead (I'm sure the team / function has expanded since then). > > The design philosophy for the website was heavily influence by Edward Tufte and myself and several folks from my engineering team were enrolled in his course to familiarize ourselves with key concepts. > > When I interviewed with McMaster-Carr I distinctly remember a Director who told me about building the first version of the website himself and not realizing it would become "a real thing" one day. I cannot for the life of me remember his name, but he was a sharp guy and I'm sure he's off doing great things. We walked into work one Monday morning, and his desk was empty and his whiteboard said "poof".
Given the timing and his description of what he did, it seems unlikely any other McMaster-Carr insider would have been writing about it, and this has to be what you are remembering, if anything.
(And https://iacollaborative.com/work/mcmaster-carr/ seems useless here.)
Still seems worth it, just wondering
Now that I think on it, I haven't seen an ad for them since before the pandemic, so I guess not (in person).
You still get all his books included in the price of taking the course: https://www.edwardtufte.com/online-course/
The other thing McMaster does that's kind of annoying, but also kind of funny, is that they go out of their way to purge the branding of the items they stock. Very understandable why they do that, but sometimes they do it when it doesn't make sense. Want to buy a generic "graphing calculator" for $126 which is definitely not a Texas Instruments TI-83 Plus? Here you go! [1]. Look, you're not fooling anybody here.
On the other hand, on my desk right now is a bag of springs, the info printed on it says it was made by WB Jones, part number 4011. We ordered it from McMaster. Why not? They stock the item and ship super quick. If I want another bag of the same springs, it's not like I can go to McMaster, type in "springs 4011" and expect to find it. Instead, I'll have to search up purchasing requests I've made, maybe ask a coworker if they ordered them, etc. to find the mcmaster number again. If I didn't know Mcmaster sold it to us, I'd have no idea they sold it at all.
To be fair, if they sell things that are interchangeable, like screws, it would be a lot to list every manufacturer they use. They have 5 locations, and probably stock from a different manufacturer or multiple manufacturers at each one.
In my opinion it is superior to McMaster
On RockAuto, watch out, because they stock some hot garbage.
They will only put a brand on a product (example: 3M DP420) when it truly comes from a single source and has special meaning/implications.
That said, I order tens of thousands of dollars of McMaster Carr items each year. They almost always come in packages from the OEM with OEM part numbers. So if I want more bolts like that, I just look at the box they were delivered in. The info is just not on the web interface.
The books are fun to leaf through on occasion, or if you need to take up an extra 3 inches on your bookshelf with something yellow. If you have one, it makes you feel like a real engineer. But I greatly prefer the website.
It's easy to create a website where interactions simply fetch the necessary resources and update the DOM as required. But managers then insist on adding 20 trackers so every little click and interaction gets logged somewhere for analytics.
Or are frameworks REALLY that slow?
Of course they are. There's a significant overhead from a virtual DOM and reconciliation with the real DOM. Then there's the larger overhead from relying on JavaScript for everything. The JS VM in modern browsers is very performant, but it can't optimize poorly written code, whether that's from frameworks, the gazillion libraries modern web sites depend on for analytics, trackers, ads, shims, helpers, etc., and, of course, any custom JS specifically written for the web site.
Browsers can enable very rich and responsive interfaces, but web development is bogged down by the insane state of popular frontend stacks. There's a recent trend of rejecting this insanity (htmx, Nue, Datastar), which I hope gets us on a track where we optimize for user experience using native web technologies.
Even though it’s never failed to connect me with precisely the part I’m seeking, to this day their interface spooks me a little: where are they hiding the endless walls of text and part numbers, the kaleidoscopic wall of bins?!
If there is a UI design award somewhere, they should definitely get it.
example: https://uk.rs-online.com/web/c/?searchTerm=zync+7010
Absolutely no BS. It dumps you right onto the list of car makes.