Enterprise software vendors have no taste | Hacker News Reader

Enterprise software vendors have no taste

medium.com

70 pointslelf97 comments

Enterprise software is often picked by committee, who will find it easier (at the same price point) to go for the product with more features rather than the product with a better design.

Hence, most of the negative feedback enterprise software vendors receive from the sales process is usually "We went with competitor X because they could do more for us.", which pushes the vendor to add more features at the expense of taste.

Spot on!

I work in enterprise software as well and unless the UX/features of an app is critical, you are spot on.

On the product I work on, one of the major GUI client is heavily used by power users with direct decision factor on the deals. It gets a lot of effort and polish and IxD folks go to client sites and maintain a feedback loop.

Other parts of the app are meant to be used by, well, "lesser" users (from a sales PoV) and get less IxD time/effort. It's all very rational.

More accurately the people who buy enterprise software are very rarely the same people who will use it. That is why they use other criteria such as tickboxes, and also why the UI design matters so much less to them.

yep and hence the constant uphill battle by companies to continually add new features to be able to advertise how much more the product can do for you and thus how much better it is. Its the checkbox mentality of the sales process which creates this race

The article author seems to have had very little experience actually in the positions that he criticizes. The truth is that actually good people in those positions are working very hard and that they have basically no resources available nor dedicated to "make it look good." The number of cycles that consumer-oriented start-ups put towards design is roughly similar to what enterprises would spend on compliance and security. Both sides are motivated by completely different business drivers - start-ups and consumer companies need design to drive customer acquisition, and enterprises tend to require sales teams to be able to engage their enterprise customers. This dramatically changes the output obviously.

Only the top-performing enterprise vendors that have been able to get the funding necessary to meet compliance issues AND design have a remote chance of doing decent UI.

Also, another factor to consider is that most enterprise software these days by the typical vendors are acquisitions, so that software has already been around for roughly... 5 years or so. After they're rebranded, it's been another 2 years and the original engineers and designers have left after displeasure with their new masters. And most of these new companies have very little interest in rocking the boat and just want to keep squeezing the existing customers for whatever and will take years and years to introduce the acquired software to their other accounts. So, by then you'll have really, really old software that you just bought.

>The article author seems to have had very little experience actually in the positions that he criticizes

The article author runs a company for 10 years that creates a single product for B2B market.

They may not look pretty, but they work a lot better than the over-designed crud with pretty UIs. As a SysAdmin, I hope none of these enterprise vendors ever hires a single designer.

This. This is software that is used to make money, not do something overly simple with pretty colors.

I think if you spend enough time appreciating the core functionality over the superficial form, your aesthetics change, and the superficial form that denotes quality functionality is what is considered 'better designed'.

It's a bridge between design and engineering. They are often two halves of the same coin.

well said. Also the author ignores the fact that changing design in enterprise sw is difficult because it impacts the user familiarity of the product they have become used to.

In reality, I consider that a problem in the consumer space too, but designers don't really care. Which is now why I just prefer to use enterprise-grade products in the home.

I laughed when I read the complaint about the "embarrassing lack of spaces"... Add the generous amount of space any fancy à la page UI needs and try to keep the all the functionality these apps have, you'll need way more than a 4k screen for it to fit the screen.

No, less is not always better.

> Add the generous amount of space any fancy à la page UI needs and try to keep the all the functionality these apps have, you'll need way more than a 4k screen for it to fit the screen.

Exactly. Reading Tufte's books I learned that if you are aiming for actually useful instead of just pretty, you'll end up with denser information display.

So you think this is OK? http://cl.ly/image/0a0r1P1m0b2U

I don't think that is quite what the parent was referring to. I think it was more "space" as in "have a clean simple uncluttered UI with ample padding around the edges" (see point 3 here: https://medium.com/@erikdkennedy/7-rules-for-creating-gorgeo...) not literally a lack of   spaces between elements.

I mean exactly that in the article. Spaces between forms and labels.

Ah sorry, I take that back then

I obviously meant general page whitespace; but you are right, the particular sentence I quoted was referring to the screenshot you posted, in which there is gratuitous (and horrific) lack of space.

I still think that my comment is appropriate for the general stance of the article, but the downvotes were undeserved (I upvoted you, for that matter)

Whitespace is downright inefficient. If I bought a screen with over 2 million pixels in it, any one of those not serving information or functional elements is wasted electricity.

If I wanted whitespace, I'd get a 800x600 screen from the 90s, and create a border around it with white paper. Lot more power efficient, same effect.

> I hope none of these enterprise vendors ever hires a single designer.

There are many aspects of design other than visual design (aka prettiness, as you call it). Usability, workflows, integration with other tools, onboarding process, documentation, etc, etc all fall within experience design. I'm not familiar with the tools mentioned in this article, but if a UX designer has never worked on them, I guarantee there's lots that can be improved without making them any more pretty.

The root cause is that enterprise software is not sold to the people that use it.

As for salesforce.com you should see how much worse the things it replaced were.

I would also frame it in terms of usability rather than beauty. As a programmer one of my pet peeves is issue tracking systems that force you to fill out 25 irrelevant fields and wait 40 seconds several times to create a ticket. Not to mind that when you say you spent 4 hours on a ticket you might find that 4 hours got added to the estimate one time and the estimate got zeroed another time. A tell is that during standups the PM is moving post its around because it is too slow to use electronic tools.

In an environment like that what should take 30 seconds takes more like 5 minutes.

>The root cause is that enterprise software is not sold to the people that use it.

That is changing I think. Quite many companies to pilot projects and compare several vendors in parallel to make a final selection. I agree that is not a majority though.

Anyway, people who buys software should take aesthetic into consideration.

P.S. I am the author of the article

Yes the article seems to be based on screenshots which are deemed to be too 'busy' or have too many gradients, rather than on actual usability.

The article is about taste, not usability. It is intentional. It is possible that usable solution still ugly. But in ideal world we should have both.

I'm not sure if that's possible. The screenshot in the article shows a lot of screens with a lot of information and a lot of possible actions to take. For software that I need to use to interact with large amount of complex data, I do not want options to be hidden behind 4/5 clicks.

When I was a part of a team that did enterprise-ish management software, one of the complaints of our initial design is that it's too simple and it takes too long to perform actions (>1 clicks from the main UI), and actually making it far more complex (users had to learn where the buttons are).

Also, "Taste" is pretty subjective. Personally, I find software that's easy to use to be of good taste. A high learning curve (lots of options) is not necessarily bad as long as I can be efficient when I'm familiar with the software (think: Vim vs Google Maps)

It may be possible but it is also difficult and adds one more constraint.

There certainly are cases where usability falls by the wayside for looks, and plenty of cases where the people involved are oblivious to everything and are just trying to get something based on mainframe or first wave client-server technology to be accessible by the web.

The learning curve issue is a real one and it has a few dimensions. One is the case of an app you use every day where click minimization is the number one thing. The other is the app that you use only occasionally and the user's long term memory is the bottleneck. One trouble is that even the app you use every day probably has some corner that you go into every six months, so even that kind of app has learning curve issues that don't go away.

In serious software for serious work there is a simple way to cut down the learning curve - training sessions. Instead of dumbing down the UI and slowing everyone down, you can just take new employees to a few hours long session where they get familiarized with the parts of the tool they need for the job.

Exactly. The feedback loop doesn't drive more usable, tasteful software. Enterprise software is a top down model. There is no competition inside the corporate firewall for any given solution so it's a monopoly and, like most monopolies, they don't enjoy continuous innovation and higher quality.

Apologies in advance for my lack in eloquence, this is off the cuff...

I have been thinking this is true of the information services organizations being organized as ITaaS. There is only one help desk and this leads to IT behaving as a monopoly. As a monopoly there are few incentives to perform better. In order to preserve the organizations value, layers of bureaucracy are added to measure the performance of the organization. These grow organically but manifest themselves as status reports and status meetings. Small committees make broad mandates which operation-wise are questionable. Looking at the CRM this creates chains of approval for resolving tickets, forms with 25 fields which contain hundreds of selectable values.

The perverse design of enterprise CRM I feel is greatly a product of the monopolistic/bureaucratic behavior in a large organization.

Many of these responses hilariously demonstrate why enterprise software interaction design is so poor: the market simply doesn't care.

The root cause is not that enterprise software vendors have no taste, it's that the stakeholders involved in the procurement process don't value interaction design. Like any market, it's likely that the vendors that are successful value those things that their customers value, and interaction design is not one of those things.

The vendors that are passionate about design eventually do something else, because their work isn't valued by the enterprise market. The vendors that remain in the market either don't value design or aren't committed to it. And, by extension, the best designers don't stick around in companies where their talents aren't valued.

I see two problems:

1) Beautiful design takes lots of effort. Making sure that texts and buttons align, choosing color schemes, takes a lot of time. A large fraction of the code is UI code, so minimizing the amount of time spent on coding the UI is a quick way to safe time and money.

2) Beautiful design requires authority. When you make a twitter client, you can just scrap that confusing feature that noone will use. In enterprise world, you need to implement every stupid feature the client wants even if it makes the product worse. No amount of taste or design can fix an interface that is broken in principle.

Enterprise software is designed to demo well (that means lots of well exposed one dimensional features) and differentiate against competitors in demos/shootouts. Everything else is secondary - design, usability. This makes it easy to sell, but leads to high levels of dissatisfaction - (i.e. the 80% failure rate of CRM or ERP or ATS software).

I will take it one step further. Enterprise software is designed to "sell". In enterprise software, the buyer and the user are not the same person and the buyer has the power. That is why enterprise software sales is so bloated.

We build and sell enterprise software and our strength is design. The way we work around this problem is that we are selling only to early adopters and have also open sourced our product. We are not getting the big bucks, but its starting to show result.

The non-design is often a feature and a selling point for these vendors.

There are some simple reasons that these might go against current trends in design - for example, these need to be high-information density displays, not minimalist showpieces. This gives plenty of room to ship something that is suboptimal, but to market it to the customer as clearly more functional, BECAUSE it's ugly.

I really think that UX can be improved, and better design could squeeze more efficiency out of these systems, but this thought process - that ugly directly means functional - actually does move a lot of software.

I literally was at a talk from an AXIS Communications salesperson yesterday. They advocated their never-changing web UI as a feature, and everyone agreed it was one.

If you mean the company that makes the network cameras, it is kinda nice that you don't have to change your code for every new camera or version of software.

Sometimes what works does not need to be changed,* especially when your talking physical security, where a camera system may be installed for upwards of 20 years. Oh, and guards who still don't know how to operate it.

* In this case I think they have made substantial changes behind the scenes to fix security holes, but the UI remains the same.

Yup.

I've developed many enterprise software solutions over the years. In my experience, there's a fundamental tradeoff in enterprise software (and software in general) that has to do with the number and sophistication of users.

Basically, if you have lots and lots of unsophisticated users, you HAVE to spend time on design to make it intuitive so you are not overwhelmed with customer support issues. If you have a few, more sophisticated users (that often require a more complex interaction with the data) you can get away with a simple interface and make up for it in training or customer support.

It's a simple cost-benefit tradeoff that enterprises with limited resources make all the time.

Given that, I agree that much enterprise is needlessly badly designed. I can look at systems and tell that nobody even mocked up the screens before the programmers started coding.

Mocking the screens would require that there be a design. Having a design would require...well, requirements.

Too few internal enterprise development environments manage to do these things. So an enterprise product with beautiful design is at a disadvantage for all the reasons other have noted, but also because it breaks the branding of the suite of enterprise software in which it becomes embedded, makes the rest of it look bad, and draws attention to the fact that the local IT department had to go out and buy something. :-) .

I'd love to see some "better" designs. Good design is hard, and it's worth paying a good designer to do the work, but often designers will make something look nicer whilst making it harder to use or a lot more resource intensive. I regularly have to load hundreds of kbs of stuff just so someone can show me a design feature that has no function. to be clear: I'm not talking about line spacing (which makes things easier to read) or font colours (which can improve contrast) or some javascript that actually does something.

tldr designers need to start calling out the bad design, not just un-designed software.

> Rational Team Concert has home icon in top corner. Why?

So many product blogs make the mistake of having the big icon linking to the home of the blog, not the home of the product. Link to the product.

This is really quite simple. It's not worth it from a cost-benefit analysis.

Good design takes a lot of effort and a lot of money. Good UI is really hard, and means spending salaries on extra people.

Good design is important if that's a competitive differentiator, which is usually the case in consumer apps/objects.

For enterprise software purchases, buying decisions are made on functionality, not design. So why would an enterprise software company waste resources on design that could go toward additional functionality, bugfixes, etc.?

Saying:

> Is it so hard to add correct spaces between form element and labels? No, it is not, but nobody cares.

is ridiculous. First of all, yes it is hard, because most developers are not trained in design, and then programming good layout actually takes a lot of effort as well. And secondly, many people do care, but it's just not a smart business prioritization of resources.

Not sure why people are surprised by it. This is capitalism - when you are in competition, the product will develop towards being as crappy as possible while still sellable. You can see thins everywhere - from software to phones to electric kettles to tools to furniture to toilet paper.

If these products were cheap I could buy this argument. But I see stuff licensing for 6 or 7 figures with UX from hell. The problem is that the people who are forced to use this crap are not the "customer" who pays for it, is sold on it, meets about the purchase, or cares about usability.

Just because the customer isn't the end user doesn't mean they are some disinterested party. Often they are writing requirements that the end user doesn't even care about, because this software has to fit into the existing software ecosystem at the megacorp. Stuff like LDAP, SAML, and ERP integration. And also provably meet corporate standards, and work on the corporations reference hardware architecture, etc.

Information-rich screens can reduce clicks and save massive amounts of time, but it's very hard to design information rich interfaces that don't clog your forest for the trees. Add in a requirement that the screen needs to be responsive and work well on mobile, and what was a heavy lift before becomes massive.

For example, https://Userify.com has a very simple (and rather prosaic) interface with not a lot of tools, but it's easier and faster for people to pick up than trying to hook authentication in their Linux boxes to Active Directory or LDAP.

The next release, by necessity, will be more complex and powerful because it's getting an approximate metric ton more functionality. We'll try to balance that by fixing the blah design and color scheme, but balancing info-rich interface against functionality on varying size screens is just really hard.

Even so, the author is right -- but form follows function still applies. I'd rather have an ugly, but fast and functional, interface over a beautiful and weak interface any day. The app that succeeds in both will be most successful.

Ultimately it's an optimization problem.

I can confirm, that, at least at some companies, more time has been devoted to ux, and now ui.

If I am correct, I think they've always shot for good ux, but ui was usually secondary. Now they have focused on ui a lot more, and companies have even begun incorporating an extensive styleguide for the user facing components (something that would noramlly be done for a webapp or website by a front end dev, but is usually neglected in enterprise apps). ux has also been refined.

Legacy products are probably not getting extreme make overs any time soon(because of existing customer base, but you never know), but I would not be surprised if we begin to see other developers of enterprise level software begin to pay more attention to UX/UI and likely mimic the look and feel of web apps created by what were originally smaller companies. We are at the point where enterprises are not only using domain specific applications, but have employees familiar with webapps like twitter, facebook, etc., all of which blow a lot of older enterprise products out of the water design wise.

Essentially, customers expect more on the design side these days, and I think enterprise devs will begin to attempt to meet that expectation to the extent that they can.

(I've even seen enterprise apps that have started using what I would term 'cutesy' user facing messages. There's probably a technical ux term for it but a good example would be github's "github <3's you' or whatever they display when you sign up--stuff like that, or giving the user a pat on the back when they run a function like 'good work!' stuff you'd never imagine would appear in enterprise level apps is creeping in because of the sheer popularity of webapps like twitter, etc.)

Are there examples of enterprise software with "good taste"? I'm sure someone out there must be doing it correctly.

I agree that the interfaces from the screenshots looked cluttered, but I think a contrasting shot of a properly designed enterprise app would have made for a stronger argument.

Atlassian products are pretty and functional to boot.

SourceTree is obviously the exception to this rule.

Tableau's design is pretty tasteful.

Shameless plug for Workday, a cloud HCM and Financials company.

http://workday.com/

Promo of our mobile app. https://www.youtube.com/watch?v=ZjRNZyRkojE

We challenge ourselves to make beautiful UIs by comparing ours to our "design competition", the modern, consumer-facing tech companies out there (FB, Twitter, Google, etc) that our customers are already used to using.

Getting snazzy UI into enterprises is not easy especially those that still use IE8!

We compete with Atlassian Jira, Rally, etc. and have taken the time to craft a different UI/UX:

https://www.countersoft.com

Splunk. Slack. Zendesk.

I wouldn't be shocked if a lot of these companies literally have no designers working on these products. Just engineers laying out UI similar to what is already in the product, based on the requirements. Any art assets come from stock libraries.

Yeah, you pretty much nailed it.

When a boss sends an email and says 'use an image like this' - you use that image.

I suggested hiring a designer, just for some of the most used features and all I got was 'yeah, okay, hmmm'

And these companies are better off for it.

Most people have no taste, that's all there is to it. Minimal and clear visual design has just started to get mainstream, just pick any piece of amateur GUI software at random and there's only a 1/1000 chance it won't be ugly. Even the most used GUIs in the world (Windows/OSX/Android/iOS) are full of little imperfections and nonsensical design decisions. One day software will be important enough that everything we use everyday will have been designed carefully, this isn't the case yet.

It's quite simple.

1. Enterprise needs vast amount of custom, specific features

2. It must be universal, to be able to do any business operations.

3. Good design costs, businesses want features, not designs (reasonably good/usable design is enough), thus save money

In my experience, specs change on the go, due to how much customization business want, this is really hard to manage on the back-end, if you'll need to do this on the front-end, development cost raises dramatically. Reasonably good, and extensible design usually is more than enough.

It's true.

My concern is that well designed, tasteful and thoughtful services are seen as hipster. I'd imagine that anything with a font size that is readable can be dismissed as hipster bullshit by enterprise clients. Even gov.uk was derided for having a clean design.

But, we've seen many times that our enterprise clients are drawn towards the design of somewhere.com, they want to use something that was built to be enjoyed. Taste is a barrier.

Sorry for the snarky nit picking but...

One of the critiqued UIs includes the following annotation:

Rational Team Concert has home icon in top corner. Why?

Surely the author wouldn't be so indeterminate in their own design right?

Targetprocess 3 has a user gravatar in the top left corner. Why?

(Actually Targetprocess looks like it has a very nice UI, but then again I don't find the others all that horrendous either. There's just no accounting for taste.)

Targetprocess design is definitely not perfect and there are many areas that are ugly and outdated. We face the same problem as all other vendors. However, we do try to design better and better and almost all of our new solutions are better than the old ones.

Taste is personal. I think these applications have no taste.

Enterprise software procurement will shift thanks to SaaS and the ability of teams/departments to buy what they need quickly. More and more, this is a means to bypass the formal, long-winded procurement process.

Hence pricing SaaS to sit below team credit card spend limit helps.

That's when polished, 'tasteful' and usable software should win.

Unfortunately the mandated apps the company uses are still going to be pretty shitty as long as they are selling those to management instead of the teams.

For instance, the example with the home button immediately reminded me of Sharepoint. What's safer than imitating SP in the corporate hegemony? (bland, overcomplex for its purpose, incredibly frustrating workflows)

Could someone tell me exactly what was wrong with the home icon in the top corner? I'm not defending the design - it looks like crap, but I couldn't put my finger on why that particular bit was wrong. There's a home icon top left in my browser as I type this.

Just look at JIRA... It's ubiquitous, cloud based, ugly and bad user experience. It's still successful.

When JIRA started it was beautiful when compared to other available options (bugzilla, for example). JIRA's UI was a major competitive advantage in the beginning. These days, it looks and feels terrible, IMHO.

I think the UX and design of Jira and the rest of the Atlassian tools is pretty good. It's reasonably clean and consistent while being customisable as required.

If anything, I'd use Jira as an example of what enterprise software should be like! It's hardly perfect, but I s better than almost every other non-tech-oriented SaaS I've used.

I think it's the enterprisey configurability really. A lot of the instances I've had to use were confusing messes with local OK parts, I wouldn't be surprised if some combination of configuration yielded much better usability.

Bugzilla is obviously very ugly (and directly out of the 1990s) but it does work pretty well and it's fairly logical to use.

Further to my comment above, we (Countersoft) compete with Jira every day and win deals because people prefer our UI over theirs!

Another reason is needing to support old browsers in SaaS for enterprise. All the new sparkles and rainbows are great design, but often also require the newest browsers to really use.

I've never worked in an enterprise environment that didn't still have to support IE8. Enterprises are slow to move on this.

Yet another reason for this not already mentioned is that some enterprise customers actually expect these ugly interfaces. This horrible ie 6/7 aesthetic has come to signify what "enterprise" software looks like, and it's what's expected.

Quick, efficient, and functional. That's what I expect of enterprise software, yes. It should have lots of options, and no useless cruft like CSS, JavaScript, etc.

Have you ever changed the UI on a legacy Enterprise app. The reaction from the Windows 8 UI changes is a taste of it.

Customers hate change. They have been trained on the legacy UI. You change the UI you make their job harder (at least that's the perception).

It's not that enterprise software vendors have no taste, it's just that taste is not a factor in the business of selling the software. And taste costs, so why spend money on something that brings zero return. Enterprise software vendors are looking at their own product and going, gee, that's ugly, but hey, look at our bank balance. More of that!

Consumer software, totally different story. But this is enterprise. As an enterprise employee, you're not paid to sit there blissing out on how good your screen looks on a rounded corner tech gadget. You are paid to do your job, whatever that is. Whatever increases that metric is what wins.

Most enterprise apps are truly ugly but what matters isn't looks but how many clicks or clacks it takes to get a specific task done. If you use the same app every day it hardly matters how pretty or even intuitive the app is.

Actually this is true and it isn’t. I mean, for sure ‘how intuitive’ is not a good metric for enterprise products, which usually come along with a big bunch of days for training, because you have a small number of users, somehow forced to use your product for a long time.

If you’re investing in reducing the ‘time to learn’ (learn the semantics of your product) you’re probably focusing on the wrong thing.

Yet, all these products (and I work on those a lot) are not well designed as a symptom of missing care for the “cognitive cost” for the user.

Anyway, the products listed in the post are way better than others. Which is sad :\

While I agree with the premise of the post, I do think that the use of design is changing in enterprise software. All of those products have been around for awhile now and I think that the idea that you need designers and good design is just a relatively new concept for ESVs. I can see it happening at my own company, a couple of years ago UX wasn't really something we considered, but now we are hiring UX people for all of our products. So I think you'll see this change going forward, it's just the enterprise is always the last to join the game.

Or the reality is, you don't need designers, but the hipster crowd convinced companies they do, because they needed something to do with their art degrees they'd get paid for.

Hiring has something to do with this in the current climate as well. If you find a dev with good UI/design chops odds are against you hiring him/her to build a UI for an invisible B2B/enterprise company in the Bay Area. Most of the people fitting that description seem to want to go to a VC backed startup building the next world changing Uber-of-SoLoMo app something or other.

I really think all of these interfaces look entirely adequate. I can imagine being able to easily use all of them without much training. There's really no comparison to the stuff I was trying to use 5-10 years ago, which tended to the illogical and unusable.

I guess I have no taste? I'm glad I'm not an interior designer.

Enterprise software is just a reflection of enterprises in general. A lot of the left hand doesn't know the what the right is doing, duplication of work, etc.

Where a lot of consumer software maps towards what a human would do naturally without the constraints of business.

I'd say that B2C apps are designed by small commited team while huge apps for B2B are designed by committee and developed by many separate teams focused on different aspects of application.

Unfortunately another contributing factor is that many enterprise websites or applications have to support old versions of ie. My company is still supporting XP due to a particular partner.

I would like to see the OP do mock redesigns of these sites they've picked, showing by example how these designs could be improved.

What about the re-designed Quickbooks Online released last Spring/Summer? Is that not considered enterprise software?

Call it whatever you want... I love it. I think I (unintentionally) signed up for it shortly after the redesign. At the start it felt like there were a few things that should have been easier to find than they were, but it has quickly and subtly evolved.

I didn't think I'd ever say this, but the new QBO has been by far the least painful small business accounting software I've used. Very very impressed!

It was a bit of a learning curve from the old layout, and I still think there's a gap between what it is and what it could one day be, but I agree that it is certainly better than 99% of enterprise software applications. QBO also serves a market where the line between consumer applications and enterprise applications is less clear.

As a target process user, I have to agree with the assessment that Enterprise Software has no taste.

The design challenge behind a modern enterprise app is also way steeper than on any consumer app.

Enterprise apps like CRM, ERP, HR, etc. need to cover a myriad of use cases and local/regional/global intricacies. They achieve this through configuration and even customization options. Nobody, ever, uses enterprise apps out of the box.

So your iPad app now needs to readjust and reconfigure itself based on customer defined data models, page layouts, workflows, triggers. And by customer I mean some yokel at Accenture/CapGemini/etc who is at least three layers removed from the person who wrote the requirements, plus never had any real training on how to configure the app.

And as they can change even stuff like field labels, your beautiful and simple UI now needs to handle huge strings in random places, etc.

Source: Running a team that builds a very successful enterprise app in a specific vertical. And we're spending a huge amount of time on protecting the end users from all the tasteless forces between us and them.

Props to the Salesforce UX team, who are fighting the same fight. Go look at the Salesforce Wave analytics app - truly wonderful UI. Their way of drilling into a donut graph is simply awesome.

You haven't seen real horror until you try to design a UI to enter accounting codes that works for more than one company.

Design by Comitee = Death

It's not about making it "look good". Completely subjective. The app needs the capbility to switch out the CSS. For instance, we resell/implement a popular cloud app - just our company - we do dozens of large implementations.

Out of box, this app has a super modern look! It is not the SAP etc.. of days gone by. Even with this, some love it, some hate it. Waste of time trying to guess what people think looks good.

Put their logo on, probably 8 out of 10 ask for this. No brainer.

Need custom CSS, 1 out of 20 ask for this. App should be able to.

If they want to get crazy, the capability for complete custom pages. Maybe 1 out of 50 customers ask for this and after you quote them the price they'll not want it.

Isn't this an argument for developing bespoke applications using lighter/more rapid technology for each customer?

At some point, a programing language is more convenient than your configuration system.

Good design is subjective. What you or I might think is good design, someone else will think is wasteful because it has too much white space.

Believe it or not, the customers for enterprise software usually want the software to pack as much information onto a single page. It might look nasty to you, but it's what the customer wants.

That's not to say it couldn't be organized better or cleaner, everything can always be improved.

But you're mistaking your own opinion for universal fact.

Exactly. The designer obsession with whitespace is basically a cancer to tech products. I go out of my way to find software that was made by an enterprise vendor, ideally one that has no designers on staff. o_o

It's those people that want to remove gridlines from spreadsheets, because pretty...

Reply on news.ycombinator.com