Skip to content

Instantly share code, notes, and snippets.

@sleepyfox
Last active December 10, 2023 18:20
Show Gist options
  • Save sleepyfox/a4d311ffcdc4fd908ec97d1c245e57dc to your computer and use it in GitHub Desktop.
Save sleepyfox/a4d311ffcdc4fd908ec97d1c245e57dc to your computer and use it in GitHub Desktop.
'Users hate change'

'Users hate change'

This week NN Group released a video by Jakob Nielsen in which he attempts to help designers deal with the problem of customers being resistant to their new site/product redesign. The argument goes thusly:

  1. Humans naturally resist change
  2. Your change is for the better
  3. Customers should just get used to it and stop complaining

There's slightly more to it than that, he caveats his argument with requiring you to have of course followed their best practices on product design, and allows for a period of customers being able to elect to continue to use the old site, although he says this is obviously only a temporary solution as you don't want to support both.

This argument is both incredibly entitled and terribly egocentric, as well as being wrong-headed on several counts.

Firstly: humans don't resist change when it's something that they asked for, they resist things being imposed upon them against their will. There is an incredibly persistent cultural movement in product design that "we know best", this is a very parent-child style relationship: "Mother knows best", that both disempowers and disengages customers.

Let me be clear: when I buy a product I am paying for what the product can do for me now. It fulfils a need that I currently have. I am not paying money out of my own pocket for a faint hope that the product may do something in the vague and nebulous future.

So: Product does X. I find that valuable. I pay $n to buy X capability. The product probably does Y and Z too, but I don't care about that. I bought it to do X.

When you as a product manager or designer or PO or whatever decide that your product should do A, B and C too, I don't care. I don't want those features, I didn't pay for them.

When you as a product person change the way that I have to use the product in order to do X, you are asking me to spend time, effort and attention to change my habits around X in order to do something differently, which may (or may not) benefit me in the future. In all likelyhood you made it easier for new users to learn X. I don't care about new users. I care about continuing to use the product in the same way as I always do in order to do X, even if you have forced me to do it in a sub-optimal way.

Every change that you make to the product after I have bought it makes it more likely that I will leave your product and find something else that does X instead, because the cost to me to learn how to something different in your product is now not much different than the cost to learn how to do something in a different product.

The more times you force me to change my behaviour, the more badwill (being the opposite of goodwill) builds up. Eventually I'll become so pissed off that I'll move, no matter what the cost.

Secondly: Your change probably isn't for the better. Not for me, not for the majority of existing customers. As stated above, the real benefit is almost always for new customers, who will find it easier to learn to use X. That's even assuming that this isn't a 'branding' change, which actually benefits no-one other than the expensive branding consultants that you just paid.

The vast majority of the effort that designers spend on look and feel, typeography, colour palettes, image choice and placement, tone of voice, button placement, size and style and a host of other things are of marginal value at best. The really hard stuff - like ethics, accessibility and knowledge architecture are almost always neglected in favour of bike-shedding. The popular rise of apps like Pocket and browser features like Firefox's Reader View are proof that it is the functionality and the content that is important, not what colour the buttons are.

Thirdly: the idea that you can just tell your customers to suck it up is a relic of last-century marketing that relied on captive customer bases and lack of customer knowledge, awareness and community. Modern customers are, in the majority, well informed and highly vocal with other customers in their community. Unless you have a significant barrier to exit you'll find that your established customer base leave the moment your competitors make it easy enough for them to migrate. Even the most impressively built and reinforced barriers don't last forever. OpenOffice and Google Docs, coupled with a change in the way that offices work have meant that even giants like Microsoft are losing their heartlands of enterprise business software contracts.

We can no longer afford to be complacement with our customers.

The idea that it is impossible to support more than one version of a product presupposes that a) work is required to upgrade both versions simultaneously, and b) that the existing product isn't stable i.e. still many bugs being surfaced. We have many known solutions for the second malady (q.v. software crafting) but the first problem overlooks a simple strategy: Extensible Product Portfolios (EPP).

The idea of EPP is thus: when you have a product that works, and an existing customer base - freeze it. Instead of a major redesign because 'Material Design is so 2014' simply leave the product the way it is, bar minor BAU and bug-fix work. Instead devote effort into building a new, next-generation product that addresses (hopefully) a new customer segment, and allow existing customers to add this new product to their portfolio for a incremental fee. This allows existing customers to self-select into a new product, protects revenue and reduces the risk of existing product customers leaving due to badwill.

In this way a team/organisation builds up a protfolio of products, all of them profitable, all of them long-lived. After the vast majority of customers leave an old product for '2.0' then when only a small minority remain you can sunset the old product, perhaps offering customers a free upgrade path, or just leave it running indefinitely as it's marginal cost of maintenance is now essentially zero.

This treats your customers like adults, gives them the freedom of choice and empowers them to use that choice in order to best satisfy their own needs.

@athornton
Copy link

For most modern apps/websites, users are not the customers. Advertisers are. They're the ones paying the bills. Thus the designer's job is to allow advertisers to pack in as much of their product as possible without driving so many users away that overall ad revenues fall. Hate the game, not the player, and go long on guillotine manufacturers' stock.

@giohobbins
Copy link

Might regret wading in these waters but this post struck a chord for me.

Appreciate the post because it's indicative of a fairly common perspective I see around. I thought there were some good points around change in products being catalyzed by bad motives - specifically: over optimization for new users and designers just wanting to try a new system/color palette.

That said, there's a lot I disagree with which was interesting. Besides there being a lot of cynicism directed towards product managers, designers, and "brand consultants," there also appears to be a lot of bad assumptions around what people in those roles actually do.

Product managers don't just change the product to help new users - business model, architecture, and much much more comes into play. Designers don't just want to fiddle with colors and bike shed on details with "at best marginal value." These misconceptions don't really apply to the author's point but definitely serve to weaken his credibility.

But most importantly the big blindspot here is a myopic understanding of how software is used. The author assumes the way he is using the product is the way everyone else uses the product. When, in reality, there are other types of customers with different priorities, workflows, and requirements and an ambitious product must serve everyone to grow. Inevitably, every product decision (even the decision to not change anything) will hurt some user's experience. You might disagree with that a product should try to serve a lot of types of people but that's a wholly different conversation.

Also the idea of a freezing a feature once it works is laughably naive. Just because it "works" for you doesn't mean it works well for other users or that the current method is the best possible way for something to work in perpetuity. That's just shortsighted and stifling of innovation.

Look at the most successful products (yes FB, TW, etc but also products like Stripe and GH that devs love). They're always changing. Bad changes get a loud backlash, good ones go unnoticed. Meanwhile, the businesses grow, add users, and are more valuable generally because of all of the changes being made.

@bradleybowman
Copy link

Another thing I thought of too--you really have to look at whom your product is serving. If it is an application people use for leisure or to keep up with friends (social media the obvious example) then yes changes are met with a bit of backlash but ultimately become part of the flow. Especially in that type of market where the 'new thing' is always making some buzz, leaving things unchanged for long periods almost invites a sense of boredom with the product.

Enterprise-targeted product use is a lot different. Anything that requires employees to relearn a task or moves a feature, even if it's ultimately a better workflow in the end, requires time. And different users have different ways of adapting and will likely not be as productive as they usually are for an indefinite period. If, in the span of a day, your spreadsheet application introduces a 'ribbon' interface (you know what I mean), and a company has X amount of employees spending Y amount of time staring at a significant paradigm shift, there's a potentially massive 'hidden cost' there. Perhaps why enterprise is often reluctant to move away from programs that are virtually unusable at this point.

Might as well let their employees install the word processor with a ribbon at home and use it for 4 or 5 (or 15) years before taking the plunge, at least there's a chance that they know how to use it by then.

@ACoolmanBigHealth
Copy link

For me, the question of "should I change this", is a strategic decision based on the product. The language of respect and offence does not resonate for me personally. But regardless, when it comes down to it I assume we all agree: Sometimes it makes good sense to commit to old versions (especially distributed binaries, or products with high setup and low frequent use), while other times it makes sense to just sunset the old version (Saas, expensive maintenance costs).

@audreymgr
Copy link

100% agree with @giohobbins, it is a siloed view to think every user has the same exact use of a product. Which is exactly the reason why products need to evolve constantly, to try and capture the biggest mass of people who see a noticeable value in the product. I enjoyed the article, and I agree with the idea that users should be in charge at all times; but the concept of designers only changing colours is wrong, as our job in product is to interact with the user base and understand what their needs are. And figure out how to modify the product to respond to said needs.

@dhc02
Copy link

dhc02 commented Aug 2, 2019

Because examples are useful, it's worth noting that Basecamp (neé 37signals) actually does EPP (which I didn't realize had a name). I have never noticed another SaaS company taking this approach, and I always thought it was super interesting.

  • They still have users on Basecamp "Classic" (their 1.0) but only ship security patches.
  • You can still sign up for Basecamp 2 and it has an entirely different pricing structure, in addition to different features, design, etc.
  • New users by default get Basecamp 3.

Over the years, they've occasionally written about why they did things this way and why they feel committed to supporting Basecamp Classic and other sunsetted products for existing customers "until the end of the internet".

@DMcCunney
Copy link

I've been in IT for over 30 years, starting on IBM mainframes and working across and down, with recent work being systems, network, and telecom administration, and Windows, Unix and Linux servers as part of my domain. I've also been involved in corporate efforts to upgrade systems.

My experience has been consistent throughout that time. Users learn just enough to do what they need to do on the machines they use, and stop. Corporate upgrades are especially fraught, as user reaction will reduce to "I don't have time to get all my work done now, and you want me to learn whole new ways of doing it on top of that? Just when am I supposed to do this?"

Part of what I've found myself doing over the decades is whatever I can to simplify user's lives, like ripping out the menu system a Unix based application suite used and replacing it with with a setup based on userid, that dropped the user directly into the particular application they would use when they logged on, and logged them off the system when they exited the application. The users were happy about this. It was one less thing to think about.

I understand the need to upgrade applications, both to fix bugs and provide new functionality. But as much as possible, the UI should not change. The user should be able to use the programs the same way they are used to, and things like new functionality should be logical extensions to the UI they already use.

I've lost count of the number of major vendors who don't seem to grasp that. An old friend is a former Art Director and currently makes a living doing photo retouching. He uses a relatively ancient version of Photoshop, and won't upgrade. The one he has does what he requires and he's adept at using it. He grew mortally tired of the total UI revamp in every new Photoshop release that required him to relearn how to use the program to do what he was accustomed to doing every time he got a new release.

And Microsoft Windows is a worst offender. I was grimly amused when they dropped the Start Menu in Win 8.1, realized it was a bad move, and brought it back in Win10. But in the spirit of fixing what wasn't broken, they moved things around and forced user to again go through the process of discovering where a particular thing they needed to do was hidden this time. I just installed Classic Shell which reproduced the Win7 Start Menu by default, but users on corporate settings won't have that option. (If anyone reading this works at Microsoft on Windows development and can explain why MS fixed what wasn't broken in the Win10 UI, I'm all ears.)

But that stuff is prevalent in the open source community, too, and possibly worse. In proprietary software, people buy it, and developer salaries come from those sales. They have reason to try to understand what users do, how they do it, and what they will think worth paying for. Open source tends to be funded differently, and there is a disconnect between developers and the folks who use their code. I've been using Mozilla code since it was still the name of a Netscape internal project to write the next generation browser suite. I've lost count of the number of changes Mozilla made that drew storms of protest from the users, and got the impression Mozilla devs live in a echo chamber and only talk to each other. I don't recall ever seeing a Mozilla effort to reach out to the user base and say "This is something we are thinking of doing. What do you folks think?" (But they likely wonder why Firefox lost market share to Chrome...)

That way lies madness. Change what you need to change in the program, but as far as possible don't change what the user sees and how they deal with the program as a user. They will thank you and be more likely to actually buy new versions.

@mojimi
Copy link

mojimi commented Aug 5, 2019

This EPP you mention, now that I think about it many companies do this, freeze the old product and release a new one, sometimes with a new name or even under a new subsidiary.

I wanted to know more about EPP but it literally only links back to this article if you search "Extensible Product Portfolios" and "EPP" returns dozen of unrelated stuff. Got any links?

@HampusNyberg
Copy link

HampusNyberg commented Aug 5, 2019

Great counter point to major redesigns of B2B or workplace tools, excluding the condescending language. But I don't agree with a lot of the arguments if you look at general product development. Designers and PMs must prioritize (and supporting 2 versions doubles the complexity).

Who are more likely to generate and grown revenue for a B2C (social, streaming, media, news, car/home sharing, e-commerce, etc) business? Existing users, or existing and new users combined? The visual digital product design is overall objectively better now than it was 10yrs ago. In general: more useful, easier to use (except when bizz intentionally priorities dark patterns), better branded, better writing/copy, faster development/design, better interaction flows, more accessible than ever, etc. Sometimes designers and PMs doesn't succeed. But often it leads to improvement in the long term.

While I get that you/OP hate change, and I know for a fact that you are part of a user group that hate change, PM's and designers shouldn't avoid learning and improving a product because the small, but vocal, group of you want products to fill your specific need then forever stagnate. The learnings also follow the individual PM/designer to their next product, which collectively helps us from same/some misstakes over and over.

@TheSonicGod
Copy link

I logged in with an old account of mine just to say that I agree with OP, 100%. That was a delightful and educational read, and I appreciate the attention to detail when it comes to customer service.

I've been working customer service and IT myself for 25+ years, so, thank you. You've done justice today.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment