top | item 7489870

We lost a customer. This is how we found out

373 points| wickedcoolmatt | 12 years ago |blog.ramen.is

121 comments

order
[+] RyanZAG|12 years ago|reply
It's pretty common on a lot of recent startup landing pages. Great looking design, nice little touches like animation - but very little detail on how the product actually works. The assumption is that someone curious is going to sign up to find out. Lots of people won't do this and will just bounce. I think every landing page should try to answer these questions:

  What is your product?
  Why would I use your product?
  How does your product work?
  What does your product cost? 
    ("Still working on it" is fine, but say so)
  What countries are you available in?
[+] ggchappell|12 years ago|reply
You've made a good list, but I think we need to begin with even more basic things.

(1) Make sure people know that the page deals with a product that is sold by a company. A web page can be anything. Maybe it's a comic strip. Or a page of essays ranting about various topics. Or an open-source project. Or a demonstration of some CSS technique.

(2) Tell what kind of thing the product is. Some companies, when asked about their product, will say something like, "It helps people discover and share ... [whatever]." But given that, I don't know whether it's an iOS app or a Windows binary or web site or an e-mail list or a metal detector or a buy-some-food-and-send-half-to-my-friend shopping service.

[+] danielweber|12 years ago|reply
Many times I see a link on HN and click on it, and they act like everyone knows exactly what is going on.

It gives me a nagging feeling that I am completely out of touch with HN, too.

The way moderators edit headlines doesn't help this at all.

[+] cube13|12 years ago|reply
Yeah, that's the problem here. The main product page doesn't answer any of this at all, and there doesn't appear to be an FAQ or any actual information about how the process works.

At best, I infer that this is a kickstarter clone for startups, but my major questions as a potential backer, like how does this work legally don't seem to be answered at all. The signup for the email list doesn't actually explain what "curated" means here, and the signup is a 1 field form for my email.

And if I were a potential project creator, there's even less information. They mention "support", but there isn't any info about what that support is.

Clicking on the "About" just brings up pictures of the team and investors, which is pretty much worthless information. I want to know about what the company does, not just who's in it. Their "product" page is a kickstarter-style pitch page, but it's not clear that they're actually talking about the product(which they apparently are)... and it appears that they're using it to sell things to raise money for the startup.

[+] clogston|12 years ago|reply
When you've spent hours upon hours deep in the engine room of your product it's SO easy to forget that no one else knows what you've built. Doubly-so if you've come from the corporate world working on a product that already has immediate brand recognition.

This is a good checklist.

[+] fishtoaster|12 years ago|reply
Honestly, at my last company we decided against adding a lot of this information. After measuring, it turned out that a simple, bare-bones landing page converted to signup best. After some investigation, it turned out that, for our business, the overwhelming majority of users showed up already knowing what they were about, rather than by searching "how do I x" or landing there via hacker news. What converted best was a site with our logo to confirm they were in the right place and a big, obvious way to sign up/in.
[+] krapp|12 years ago|reply
And they'll want your email address, and global permissions for your twitter and/or facebook account, just for the chance to maybe find out in the future what they should be telling you up front. "signing up" has, unfortunately, become something of a user-hostile act of late.
[+] icedog|12 years ago|reply
Indeed, this helps solve a bigger problem than usability: what the hell are they selling.

I'm often frustrated by the poor messaging written on landing pages or even the 'About [Product]' pages. Either the page is 100% marketing fluff or 100% technical info. With marketing fluff, I can read a thousand words and still not get any substance. With just technical info, they often leave out answering the key marketing answers: what are the competitive advantages?

[+] DanBC|12 years ago|reply
It's not just startup pages. Often open source software homepage is a news page, which means recent release change lists are what the user sees first.

I guess someone good at writing and wrangling the change system could submit change requests, but it's a shame that's needed.

[+] MasterScrat|12 years ago|reply
Also if it's a blog post provide a big, fat, highly visible link to your actual site.

It's incredible how many product blogs will just bring you to the blog's main page if you click the title instead of the actual product website.

[+] bobbles|12 years ago|reply
Also, When answering the question "what is your product" DO NOT SAY "we are like yyy for xxx"

Half the sites I see that use this I have absolutely no idea what 'yyy' is or does, so you may as well have put nothing at all.

[+] angilly|12 years ago|reply
Yeah I hear you. We put ton of long form philosophical waxing onto our blog in the early days. We're working now on taking all that and boiling it down to homepage-speak. Homepages are hard ;)
[+] jagath|12 years ago|reply
> Lots of people won't do this and will just bounce

That's a huge assumption. The only way to know for real is to A/B test. You may be surprised by what testing will tell you. We constantly A/B test landing page designs at Ordoro, and have often found results that are counter-intuitive.

[+] pyrocat|12 years ago|reply
Or better yet

Here is a problem This is how our product solves that problem

followed by additional features that make your product stand out.

[+] groovy2shoes|12 years ago|reply
A big one for me:

    What does your product look like?
[+] rowyourboat|12 years ago|reply
"I failed to explain the benefits Ramen can provide."

No, I think those came across quite clearly. He wants to know how it works, i.e. what is required of him, and what will the process of getting funded look like. That's a difference.

[+] wickedcoolmatt|12 years ago|reply
Our wording is a little off there. Appreciate the feedback, and you're totally right.
[+] jaredandrews|12 years ago|reply
Interesting comment on the bottom of the article:

  This random user is a prick.

  1. Why didn’t he scroll to the bottom of the page? I noticed an “about us” link in the footer.

  2. When I’m curious about companies, I look at their blogs. Sometimes they use a blog CMS system. Why didn’t this guy check the links in the header, like the link to your blog?

  3. This guy has a baby babbling in the background. Maybe he was distracted?

We do a lot of user testing at work and it has really opened my eyes. This sort of attitude really bums me out though. You can sit around all day and complain about customer incompetence. Meanwhile they are out using a different app/product and you still aren't making any money/conversions.
[+] angilly|12 years ago|reply
FWIW, I thought for a while about whether or not to approve that comment due to it's flamewar-factor, but I approved it so I could write this response:

"Ehhh I wouldn’t go so far as to call him a “prick”. Yeah he certainly could have gone through our blog and read up on Ramen. He could have clicked on our project page and watched our video. I think the point of Peek, though, is that you can’t tell a visitor that their first impression (and/or their first confusion) is _wrong_. I think he brings up a lot of completely valid points about our homepage. We haven’t put enough energy into making our homepage something that succinctly describes what we do. He called us out on it. I give him kudos."

[+] 6cxs2hd6|12 years ago|reply
As a founder I loved to get feedback like this. Was I proud of some feature or design that I'd worked hard on? That's nice, but if it's not working for customers, it's not working. Swallow pride, move on.

Although that's a healthy attitude, it gets tricky when the company grows and the customer is criticizing work done by your employees. Your instinct is still that the customer is automatically right, and that's still true, but you have to be less blunt with employees than you'd be with yourself. You need to acknowledge their work was good, we just need to try something different. Regardless, they need to hear and act on it. And there is no way you call a customer a "prick", for offering polite feedback like in that video. You thank them.

Sometimes I'd remind folks that it wasn't me, it was our customers who were paying for our salaries, the office around us, the chairs we're sitting on. Ultimately the customers are the boss, not me.

Edit: If you have investors, now you have two bosses.

[+] vacri|12 years ago|reply
It's pretty clear that that commenter doesn't have any serious experience with user testing. A design needs to work with the user - you don't abuse the user because they don't efficiently use your design. Well, not if you're professional.
[+] adrianhoward|12 years ago|reply
Everybody seems to be missing the biggest lesson from this. No matter how obvious you think the problems with the OPs site was you should...

DO USABILITY TESTING

Often.

Because if those problems were obvious to the OP - they would have fixed them. By definition. I guarantee that everybody here has there own blind spots with there own application or service.

I've been doing usability tests for nearly twenty years now - and the number of times we've found nothing that can be improved can be counted on approximately no hands.

Use online services like peek (there are many, many others too). Do them yourselves. Do them regularly. If I could pick only one thing to help folk improve their product - usability testing would be it. Even above customer interviewing. Nothing beats watching your customers try and fail to use your product.

Here's three books to get you started:

* Rocket Surgery Made Easy: The Do-it-yourself Guide to Finding and Fixing Usability Problems, Steve Krug - Does exactly what it says on the tin. Short sharp guide to getting you started.

* Handbook of Usability Testing: Howto Plan, Design, and Conduct Effective Tests, Jeffrey Rubin & Dana Chisnell - The last book but in much more depth. The first edition of this was my bible when I started doing usability testing.

* Remote Research, Nate Bolt & Tony Tulathimutte - A great guide to how to approach getting the most out of remote usability testing services like peek. The tools are a few years out of date now. The advice isn't.

Seriously. If you're not already doing usability tests go spend an hour or two reading 'Rocket Surgery Made Easy' and then go test your product with some actual human beings. You'll thank me.

[+] userium|12 years ago|reply
I agree with the importance of doing usability testing. I made this usability checklist (http://userium.com/), which can be used to catch common usability problems before user testing. By fixing obvious usability problems you get more meaningful feedback from users.
[+] dredmorbius|12 years ago|reply
So much this.

A few bits I've noticed over 25+ years in the industry:

• Tell me what your product is. What it does, where it works, how it does it, what it requires. Is it a physical product (or is it shipped in one), an interactive application, a Web service, a programming language / tool?

• Tell me what the fuck it is EVERY GODDAMNED TIME YOU COMMUNICATE ABOUT THE PRODUCT. It doesn't have to be long or detailed, you can link to your detailed description in the communication. But your press releases, emails, Tweets, blog posts, marketing collateral, etc., are going to get passed around, word-of-mouthed, and/or pulled out of drawers (or browser history / searches) for weeks, months, and years to come. Make them work for you.

• The Economist's practice of briefly introducing any individual, no matter how famous or obscure, is a wonderful practice of microcontent contextualization. "Using the Economist house style offers an elegant alternative, wherein virtually all people and organizations are identified explicitly, no matter how prominent. For example, you might see 'Google, a search giant', 'GE, an American conglomerate', or 'Tim Cook, boss of Apple'." http://redd.it/1x8yky

• Tell me how to try it out. Preferably for 60-90 days (a 30 day cycle can go far too fast. I've been very, very impressed with New Relic's "use it for free, convert upmarket for additional features" model, and it's apparently worked well for them. For small accounts, their cost of sales is effectively nil (and for large accounts, COS is always a PITA). But for those large accounts, you've got a proven track record with the prospect, and they really know what they're getting.

• Put your tech docs front and center. As a technical lead / director, my questions are "how the fuck do I make this thing work", and if you can't tell me

• It's been observed many times that those who have the best appreciation for how a product works are those who use it directly, and secondly, those who either service it or support those using it. John Sealy Brown's The Social Life of Information addresses this with both Xerox copier repairmen and support staff. Use this to your advantage two ways: let these people share and collaborate, even if informally For the repairmen, this was a morning coffee break turned out to be a hugely valuable cross-training and troubleshooting feature. For phone support, after an "expert system" and changes in technology separating phone reps from technicians, researchers noted two reps who consistently provided good advice: one was a veteran from the earlier stage, the other a recent hire who sat across from the other and learned from her. Similarly, user support groups (mailing lists, Web forums, Usenet groups), in which users interact and share knowledge with one another directly (Hacker News would be an instance) are often (though not always) far more useful than direct tech support.

• Provide clear pricing information. This has been noted from Jacob Nielsen on forward as the information people are most interested in.

• Make damned sure that whatever process or workflow you've created online works, and for as many possible end-user environments as possible. Keeping interfaces as simple and legible as possible is a huge bonus.

• Remove distractions from your transactional webpages. Once someone's homed in on a product, focus on that, though you may mention alternatives or (truly useful) related products. Every additional piece of information on the screen is an opportunity to confuse and lose the sale. I've been restyling many websites simply for my own use (1000+), and simply removing distracting elements produces a far more productive environment.

• Ensure your pages are legible. Backgrounds should be light, foregrounds light (and where, with extreme reluctance, you invert these, separation should be clear). DO NOT SCALE FONTS IN PX. On far, far too many devices this renders as unreadable, particularly from older (e.g., more senior w/in the organization) readers. Grey-on-grey is just cause to fire whomever suggested or required it. See ContrastRebellion: http://www.contrastrebellion.com/

• Don't organize your website according to internal corporate structures. Your website is an outward facing tool, and should address the needs of users, not of internal departments. Lenovo's laptop site organization would be highly typical of this: I want a Linux-capable, large-display, full-keyboard, trackpoint device. The rest I generally don't give a shit about, and its product line confuses me every fucking goddamned time I try to buy something there (usually every 2-3 years). I'm not a sufficiently frequent customer that I keep up with every last change, but I've spent thousands of dollars on IBM/Lenovo products, as an individual (hundreds of thousands to millions as an enterprise customer).

And of course: test all of this, don't simply take my word for it. But yes, I've walked from far, far, far too many product pages, from free software projects to Fortune 10 companies to edgy app devs.

Life's too fucking short for that shit.

[+] Malarkey73|12 years ago|reply
Jeez. That is like sooooo many modern websites.

What is this site? What is thingumajig.io? Its a webby thingumajig? Sign up? Sign up for what? Oh it's a website for web something.

nice font.

[+] webwright|12 years ago|reply
If you do this, it's best to do it in clusters of 5-10 testers in your target demographic and try to identify patterns. It's dangerous to say "UserX is confused, so all of our users must be confused." No matter how perfect your design, it will be confusing/frustrating for someone.
[+] jrochkind1|12 years ago|reply
I had not known about the Peek service, that's pretty awesome.

Does anyone know of anything like that, but that's _not_ random, where I could actually send volunteers from my current users to my sight, and have their clicks and voice recorded and sent to me? Is there such a business that works well at a reasonable price?

[+] robby1066|12 years ago|reply
usertesting.com has this in it's main product (which Peek seems to be just a scaled back, lead-generation version of), but I think supplying your own testers is one of their enterprise features, which is more expensive than the normal service.
[+] mauricio-OH|12 years ago|reply
Great post but it sounds a lot like trying to sell that peek service. Would've been great to see what actions were taken to improve the site rather than wait to the new design launch.
[+] harrystone|12 years ago|reply
I don't do web design but that Peek thing just looks brutal to me. It's probably a good tool though, I'd just hate to have it pointed at my work.
[+] wickedcoolmatt|12 years ago|reply
Hey y'all thanks so much for the support!

UserTesting (the folks behind Peek) shot us over a promo code to get the first 100 of you to the front of the line if you want to give it a whirl: ramenreader

[+] carrotleads|12 years ago|reply
not sure where I should enter the promocode.

Have submitted my site carrotleads.com for a test though. It free anyway, so what was the promo code for???

[+] rurban|12 years ago|reply
Come on, this guy is not everything. The first link I clicked was at the bottom left: Project which brought me to https://ramen.is/projects/ramen which explains everything in detail this guy did not find and searched for. Overall, looks a fancy new kickstarter site. I care about projects not kickstarter per se, so I like the idea how the projects are presented in ramen.
[+] edmccard|12 years ago|reply
>The first link I clicked was at the bottom left: Project

Well, if I was looking for information on how Ramen worked, a generic link in the footer called "Project Page", under the equally generic heading "Product" is not the first thing I would click. And if I eventually got there, and saw what looked to be a page intended for backers (it has a blurb and options to back at $2.00, $10.00, etc.), I probably wouldn't notice the "How it Works" section buried three screenfuls down the page.

[+] farresito|12 years ago|reply
You said it all: the link was at the bottom left. I rarely go to the bottom links when surfing a webpage, unless I'm really sure what I'm looking for could be there.
[+] vacri|12 years ago|reply
Primary information should not be put in a low-contrast footer, the place where you usually put things like the privacy declaration, or which company made the theme, or which CMS is 'powering' the site. The footer is a blind spot which plenty of people mentally block out because it's usually full of meaningless information.
[+] joshvm|12 years ago|reply
The information may be there, but if the user has to scroll to see it then that's poor UX design. I did wonder why he didn't scroll to the bottom, but still "About Us" doesn't scream "how the service works" to me. And indeed, if you click on the link you get taken to a page of smiling founders.

I got the impression that the designers were trying to be clever by integrating the tutorial into an actual project page - and going to the Ramen project explains a lot about the process. However, that's intuition from using the web a lot, not a logical step. It's the same kind of intuition that gamers have when crawling a dungeon and you know that taking the short route will almost certainly be a Bad Idea.

This backfires in another way: I also wonder if there are only four projects on the site? Can I search for more? It makes me think that the projects there are just dummy pages to demonstrate how the site works.

[+] htk|12 years ago|reply
Great post, the video is as simple as is enlightening.

What I like the most about it is that the user is genuinely interested in the service. But he acts natural and realizes he doesn't see an easy way to get more info on it.

He could try to read the blog, or search for small print, but that's not what the average user is going to do.

I'm going to try Peek soon!

[+] carrotleads|12 years ago|reply
Well their main problem was they had a 2 sided market and catered to both half heartedly.

Dropbox caters to a single market and the message is more simple for them.

A "How it works" with subsection for both target market would help.

multiple sided products always have trouble selling effective messages and would like to see examples if you guys know of any.

I am doing a redesign of my site http://carrotleads.com on this topic. Was targeted at companies earlier and now I will have trouble converting network'rs. I can see the problem, but solutions need more deep thought. Working on it.

Submitted site to Peek. Want to see how it turns out.

[+] fallinghawks|12 years ago|reply
There's this one link at the very bottom of the page, which the Peek user kept missing, and would be the first one I would go to if I couldn't find what I was looking for: Product/Project Page. That appears to be the page that would have answered many of his questions and kept him as a customer. That should totally be at the top of the page.

However, even that's a little confusing because it is itself a project -- so a user might wonder what the heck they are looking at, site info or some other project?

[+] bpodgursky|12 years ago|reply
It's also odd that there's no "browse all projects" or "find projects to back" option, only a few hand-curated options. No discovery options at all.
[+] angilly|12 years ago|reply
One of the Ramen founders here. We just launched a few weeks ago, so that's actually all we got right now. Feel free to submit a project if you'd like :)
[+] tdicola|12 years ago|reply
Are user videos from Peek supposed to be public? I would be a little creeped out if I was the random user from this study and suddenly found my video plastered in a blog post.
[+] arnklint|12 years ago|reply
Another solution to _this_ problem might be to get your site reviewed by a conversion specialist or using some sort of heuristic review. In many cases my experience is that a combination of more qualitative methods such as user testing combined and heuristic evaluations and heatmaps and with Google Analytics solves _most_ of the issues you have with your site.
[+] nedwin|12 years ago|reply
While you're getting all this traffic you should definitely put a couple of calls to action to check out your product.

Even just linking the word "Ramen" in the blog post would be a great start to get more conversions happening.