Lots of people make a quick judgement on your trustworthiness based on whether you at least have adequate design. If they turn away in 200 ms because of poor design, they will never be able to evaluate what your true value proposition is.
Many decisions in a B2B context are defensively made, so as to avoid the risk of looking stupid in the future if things don't work out. Like the old saying, nobody got fired for buying IBM. If your branding and design isn't at least adequate, again there will be no trust and a fear of being blamed in the future.
I've even had clients tell me that design really doesn't matter and they just want to see functionality right now, only to be shown something with full functionality but poor design and then walk away.
The trick is to not be a perfectionist but to spend the 20% effort that will get you to the 80% which is good enough.
Is it poor advice though? I don't think the article is suggesting to skip designing a nice landing page -- the article is saying that customizing the _login page_ is not worth the effort, which I agree with.
I can say that at my last company, redesigning the login page was bottom of the stack of things we wanted our first designer to work on. We just had a centered "Login with Google" button, our logo, and a email + password form. What else do you need?
Your conclusion is hard to argue against, but I'd like to pitch in that some customers explicitly want something that's not well designed.
In particular, those customers that think adopting an early-stage prototype will give them a competitive advantage, they expect something that's not well designed. If they get something shiny when they've asked for a prototype, they'll be worried you're trying to hide a turd behind the polish, whereas if you give them something that doesn't look great, they'll know you're serious about giving them the latest, rawest, earliest experience of whatever it is you're innovating on.
(Of course, "early adopters" is a rather small market niche, so you might not grow very far doing just this.)
It pulls at a compelling thread about the differential possibilities of design and the identities it's coded against. Anti-design is an inherently valuable approach to work. It can let us think about how to achieve the right client fit, the right product motivations, and the right overall system incentives. This isn't to agree (per se) that only bad clients want glossy design in prototypes, but rather to suggest that we as people self-segregate into particular visual cultures that may seem inherently bad to outsiders. Design, or the perceived lack of it, can be a form of (highly valued) in-group signalling that allows us to sieve out the wrong kinds of interactions.
I had a contractor do house repairs before we put it on the market. We didn't ask him to, but he took his orbital sander and sanded the stair railings that lead to the front door. He said "people really notice this."
The very first prospect we had put his hand on the railing, looked up at the house, and said "this place seems to be well taken care of." And made a decent offer the next day.
So I'm in favor of detailing the sign up page with care.
The advice on polish applies mostly to venture-funded startups trying to get users to move from existing $SOCIAL_APP to new $SOCIAL_APP; in this instance, sure, polish what you can at the top of the funnel.
For B2B products it's a little different. When you're holding an MVP in your hands, it's really hard to tell whether you're grasping a rusty old anchor that will drag you to your death, or whether you're looking at lightning in a bottle.
One quick way to tell is if businesses sign up and user it even though it's as fugly as can be.
If, when pitching the product to a potential sale, the subject indicates that they need particular cosmetic changes before higher-ups would approve it, then you're likely pitching a dud product.
If, otoh, they interrupt you during the pitch of that fugly product, and offer to trial it immediately, or buy the prototype, or call in a higher up, then you have the proverbial lightning in a bottle.
Cosmetics is, I think, a good indicator for B2B products - if people want it even though it's ugly, you're selling painkillers. If people want it only when it's pretty, you're selling vitamins.
You could just as easily get the opposite effect: if the login page is polished and snazzy, the user may well have the same expectation of the product itself, and every sharp edge will be that much more visible.
If the login page is simple and generic, the message is way toned down. Similar to the advice of not polishing the front-end too much if the backend barely works because clients will expect that a finished interface means a finished product.
Not only is the comment more nuanced and helpful than the original article, it also quite easily holds up against ye olde yOu dOnT kNoW sTufF argument.
I too once thought "Why don't girls like me for who I am, why do I have to dress nice and make myself presentable, this is just a façade and I want them to love me for who I am inside", but then I grew up and understood that appearances matter and there's no point in fighting people's feelings. You might say "don't worry buddy, my startup is really good and effective, don't mind the messy styling and presentation" but why place that hurdle on your business when just a tiny bit of effort can make a big difference.
There's no need to be pixel perfect, but some eye for colors and proportions goes a long way, just like there's no need to perfectly groom every single hair, but a nice shave and deodorant go a long way.
For that I had the quote saved to a text file of comments, with the link.
If it were something like a web page which I found worthwhile it'd be saved as a MHTML, categorized and with notes in the filename for quick searching via Everything. I don't use bookmarks for anything except regularly accessed pages.
20 years ago there was an article[1] that argued anything worth sharing was worth saving, on the benefit of keeping a copy of content locally vs linking/bookmarking (which is prone to link rot). Ironically that site itself no longer exists.
> If you want your links to be worth anything in two, three, or five years, download all the pages you're linking to to your hard disk.
This is a nugget of something useful wrapped up in a way that turns it into naive advice that feels like a contrarian talking point.
The biggest risk to your funded startup is opportunity cost. You can (hopefully) find more funding and that can give you more time, but you cannot rewrite the time lost on the wrong thing.
This idea that your first cohort should be religious believers regardless of the public image of your startup is a nonsense bubble dream. If you need customers then your time is best spent on the details that could favour acquiring them. If you're running a B2B and you're spending resources on granular refinement of your public image, then yes that may well be highly inefficient vs going out to network and get LOIs / deals signed.
The summary of what was a good point is:
Really actually understand your business, your customer and your success criteria and chart as close to a straight line to that point as possible with minimal diversions.
This is terrible advice. The author is essentially discounting the value of aesthetic, which is nonsense, and then suggesting that if you do value aesthetic you automatically have a bad product or a bad investor, again utter rubbish.
I argue it’s the other way, discounting aesthetic in this way to me says you are either lazy (and therefore I should be concerned about credibility and other gaps) or you are a contrarian in which case you will likely be high maintenance and trouble.
Horrible horrible advice. Agree on the login page etc.. but your sales stuff needs to be on point. I see this from the other perspective, working at a big company. The smaller players never get their foot in the door because they don't know how to talk to us. And they don't optimize their decks toward the audience.
If the talk is not about they customer, why should your customer care?
"The perfectionist dilemma is when a creator values the quality of a finished product such to the extent that it inhibits their ability to iterate, change, and even produce." - Adam Miller
I agree that this is a hard truth, especially among us developers, but I believe it has roots that can't be easily ignored. The amount of fear of failure when thinking about releasing a software product to the public (whether it's a website, a mobile/desktop app, or a game) is big and scary, not because your product stinks [0], but because you got afraid that there is something wrong out there that will push-back the user in his first 500ms of trying your product, hence, you Start Customizing Your Startup's Login Page again 'n' again..
I think the general sentiment (spend time on things that matter) is valid, but to some extremes it might bring more damage than value. In this specific case, either you have a disruptive product such as GPT that sells itself, or you will need to promote the s*t out of it before it has some traction (the common fallacy of "build it and they will come").
So I agree, spending a ton of time customizing a signup or landing page, a powerpoint presentation, etc might be at the same time a waste of time, and the most important thing to do - depending on the circumstances.
You might argue that the idea of an MVP has shifted more towards "quality", but if you're not in a crowded market (I am, OnlineOrNot got significantly more users when I started caring about landing pages), and fixing expensive problems, you can get away with a lot.
In a previous gig I worked at a startup so early stage they had hardcoded their logins in the backend, and users could not change anything without emailing us. The problem we were solving was worth tens of thousands of dollars to the users, so they didn't care.
If you spend your life filtering out for the "right kind of customer" based on some Platonic ideal you have of a correct customer you are going to throw 95% of them away
All of the examples given are about trust. Does the user trust that the product is worth their time to use, does the investor trust that the company will make them money.
There are many ways to gain trust, and you often need some of everything. A website that looks like it was built in the 90s likely won't gain sufficient user trust to even get to proving the value of the features. An investor deck that doesn't pass a minimum bar won't even get a reply, let alone a chance to defend the product.
This is not because people are "shallow" and want their login pages or decks to be pretty, it's because we pattern match to save time. I don't reply to every terrible recruiter email I get because I know from experience that it's unlikely to be of any benefit to me – I don't need to do some job interviews every time to make that reasonable conclusion.
Know your audience, know what their bar is, don't spend time going above and beyond on details that don't matter, but do spend time reaching the bars that are necessary, there are many of them.
There's a reason we wear tuxedos to weddings. Not because it's intrinsically better than a tracksuit, but because it sends a specific message. Social conventions have meaning whether you like it or not, and a nice landing page or PowerPoint have meaning too. Also, as van der Rohe used to say "God is in the details"
> then you can eventually think about your companies branding.
Not sure that I can find common ground with this statement.
Branding is something that tends to be poorly understood (or supported) by "doers," (usually the tech people that are implementing the product).
The thing is, no one really appreciates a brand, unless it is successful.
I worked for a company that was (possibly still is) one of the premier brands in the world.
They had no cohesive branding strategy for about 75 of their 100 years, and ended up "retrofitting" a cohesive branding strategy, in the 1990s (my boss, who eventually became the Chairman, was behind that).
It was quite painful. It ended up working out, but there was a lot of agita.
I think we shouldn't "bikeshed" branding, but we'd be well advised to keep it in mind, from Day One.
Imho, any 50-100$ template would do. It's not about aesthetics being unimportant. The scale of measure here is ROI and the ROI of doing it yourself to get something good looking is probably bad for most startups.
Maybe if the thing you're selling is something "disruptive" enough that people will give you that pass. But if you're selling another variation of <insert new devops tool> or <project management software> you better have a baseline (aka good) level of polish. people absolutely judge you based on those pages.
i can't tell you the last time i went to a corporate site that had a shitty landing page and a wonderful to use product. this advice was sound 10 years ago, not now IMO.
Well, as usual it depends. It depends on who is your target customer and what they value. Here in Hacker News the login page is completely barebones and it's OK. In other cases that won't pass, but a login page using the stock tailwind components (or another library) will be enough.
In general, I'd say that picking a good library of components is necessary for most projects and it will take you to the proverbial 80% results with 20% effort.
In a world where everyone and their grandmother has a (on the surface) polished SaaS side-gig appearance does matter if you want to be taken seriously. It shouldn't be but it's the truth.
If time is an issue then you find a way to polish without spending much time. What does the login page need except for a logo, a form and a few links? Remove the bloat and keeping things polished becomes easier than fixing it after the fact. Less is more.
I was involved with a startup that ultimately failed because of the dot com bubble, but one major observation was that everyone who used our product was fantastically enthusiastic about it, but persuading them that they needed it in the first place was really difficult. Making the login page look pretty helps with that.
Complete BS. Early '90s, perhaps — Users have advanced in their judgmental capabilities. And these days, you only have .5 seconds to make a first, splendid impression. The details are not the details, they are the product.
I want to like this page, I really do. But please don't use Computer Modern on the web... It's a pretty ancient typeface that's made for paper first, and looks bad on a raster display, even a pixel-dense one.
> I want to like this page, I really do. But please don't use Computer Modern on the web... It's a pretty ancient typeface that's made for paper first, and looks bad on a raster display, even a pixel-dense one.
Looks okay to me (on desktop and mobile).
My take on readability wrt to blog posts is that the font doesn't matter as much as you think it does. Sure, don't use Comic Sans or some handwriting-equivalent, but any mainstream font should work (sans for headlines, serifs for content, maybe?)
The larger factors affecting readability:
1. The paragraph width. It's tiring to read a paragraph that's 250+ characters wide. Stick to 64-78 characters wide. I find setting max-width to 38em-42em best.
2. Line spacing - bump it up a little - 1.25em-1.6em is great at normal font sizes. You need to adjust as the font size increases.
3. Background color - muted colors are nice. Bright colors are not.
4. Foreground color - start with low contrast and increase it until the devtools in the browser tells you it passes the accessibility test.
Only after you have done all that, start playing with the font.
The thing is: We don't know what we should optimize first. Sure, We should make our core offering very good so that customer buys it. If We feel our offering is compelling enough, what's next?
Does optimizing landing page counts? or reworking that pesky login page?
Lots of people make a quick judgement on your trustworthiness based on whether you at least have adequate design. If they turn away in 200 ms because of poor design, they will never be able to evaluate what your true value proposition is.
Many decisions in a B2B context are defensively made, so as to avoid the risk of looking stupid in the future if things don't work out. Like the old saying, nobody got fired for buying IBM. If your branding and design isn't at least adequate, again there will be no trust and a fear of being blamed in the future.
I've even had clients tell me that design really doesn't matter and they just want to see functionality right now, only to be shown something with full functionality but poor design and then walk away.
The trick is to not be a perfectionist but to spend the 20% effort that will get you to the 80% which is good enough.