Social media’s effect on Browsera traffic
Dec1
A few of you have asked how much traffic the minor ‘press events’ have driven to the site. This is the analytics graph of traffic since our launch.
The interesting thing to note is that much of the traffic did not come directly as a result of a blog post/trackback. Most of it came from tweets or discussions after the editorial. From the referrers, I can see that while the initial posts on habrahabr and lifehacker drove a lot of traffic, there was more ‘direct’ traffic coming as a result of people clicking on tweet links.
This is the power of social media–click through rate seems to be much higher than content in anonymous yet highly reputable sources. I’m still working on correlating signup data with these events, so stay tuned for more detailed metrics.
Vote for the Best Design!
Sep9
My 99designs contest is over, and it’s now time to pick a winner. I was surprised at how many homepage design entries I received despite not having one of the top prizes in the category. Here are the top 8 5 finalist in my opinion, in no particular order, please vote for the best new homepage design in the comments below.
Design on a Shoestring Budget
Sep0
I’m very pleased with how my 99designs competition has been going. You can check out the latest entries here
My contest is for the redesign of 2 pages, and only a PSD file is required. You generally redesign your entire site, however, it is too difficult to hold a contest to redesign a lot of pages. The advice I received was to just have 2 different looking pages designed, and then work directly with the designer for any remaining pages that you cannot modify yourself.
When I first submitted the contest, I was nervous about potentially not getting any submissions. However, within 24 hours, I had already received 2 entries to my contest. These entries weren’t stellar, but they were far better than my current site:
It was important to me that the new homepage contain some specific elements. The logo, header, and navigation of course, but also a set of features, a prominent sign up button, and a screenshot. For this reason, I submitted a mockup I created using Balsamiq mockups. Balsamiq is an excellent tool for those of not artistically inclined, because it is really easy to use and fast to create application mockups using common UI idioms. Need a tabbed interface or a tree? Just drag it in and tweak the settings. This also allows you to resize, add, and remove elements without messy scratch-outs on whiteboards and paper.
Here is the mockup I provided for the homepage:
My inspiration for this mockup was derived from sites I liked that had similar business models. Namely, Serverdensity, Basecamp, and Squarespace.
Initially, I received many exact clones of this mockup, but some of them looked very good. Eventually, I took the mockup down to try and get more options. In all though, many of the designs do look quite different from one another.
After receiving about 15 design entries, I decided to ‘Guarantee’ the contest, meaning I will definitely award the prize to one of the designers. According to 99designs, it increases the # of designs entered by 100%! This was well worth it because I plan on definitely using one of the designs.
Everyone knows that providing feedback on the designs is key. This takes time. A lot of time. Figure about 10 minutes on average scrutinizing each unique screenshot submitted and providing a rating and feedback. It’s obviously better if you can provide very specific feedback, like “Make the navigation red instead of blue”, but for me, I only know good design when I see it and have a hard time articulating how to improve it. So, most of my comments instruct the designer to use cues from other sites or designs and integrate them. For instance, “Tie the left nav to the content, like in Gmail.” Or, “Make the sign up pop like on the Squarespace homepage.” Sometimes, things just didn’t look right, and I’d provide feedback that the design was dull or not the style I wanted.
Figure on constantly providing feedback 2 or 3 times per day–you’ll be surprised at how quickly the designers adapt their designs and resubmit. The total # of submissions has just crossed 100, though many entries contain small refinements. I’d estimate there have been at least 20 unique designers submitting to the competition.
I watched many contests on 99designs before holding my own, and often found lots of unrated designs. I think by providing a rating for EVERY design has helped, because the designers get a better sense of what you are looking for. While a few designers will outright copy one of the designs with a high rating, most won’t, and you end up with designs more in line with your style.
For rating designs, I used the following rating guidelines:
0* Still thinking about it
1* Concept needs a lot of work and does not appeal to me
2* Some parts of the page are good, but other areas need to be completely redone or the layout needs to change
3* Overall layout is good, may need refinement of sizes/fonts/colors/graphics. Please submit a design of second ‘Report’ page.
4* Finalist
5* Winner
I extended my contest a few days from the 7 day default in order to give some of the late entries time to refine, and get some more submissions for the second page design. I’ll post more once it’s over.
Redesigning a site in 7 days
Sep0
After touting the benefits of 99designs to others, I’ve decided to run a contest myself. The web design of Browsera has never been a reason to come back, or even stick around, and so it desperately needs a redesign.
At the same time, I thought it would be a good idea to improve the UX information architecture and workflow, so what started as a simple reskin turned into an entire site reworking. Since I wasn’t able to get a hold of the few designers I know, I decided to put up a contest on 99designs.
http://99designs.com/contests/28043
For those that arent familiar, 99designs is a marketplace allowing you to host a contest that where you reward the best design submitted with a prize. That means you get many submissions even though only 1 winner will be paid for his/her work.
It turns out to be pretty cheap both on an absolute scale (~$600 for web design versus $2-3k from a traditional web design shop) and on a per concept scale (25 design concepts at $24 each). I tend to like variety, and I noticed that in many of the contests, it was hard to pick a winner. The designs are just that good.
Competition is a good thing, and 99designs is capitalizing on the fact that every designer thinks they’re the best and wants to prove it. I’ll provide more details as the contest goes on. In the meantime, feel free to comment on the submissions.
Advice: Easy to give, easy to take, hard to make work
Sep2
As an entrepreneur in his first attempt at building a startup, I often solicit and receive advice. The topics range from the best library to use for a feature to how to brand my service. The problem is, advice is so easy to give, and pretty easy to take, but really, really difficult to make it work.
It reminds me of the days of school where things were “left as an exercise to the reader”. There are a ton of different theories and hypotheses about the best course of action that you’ll receive from people whose opinions you respect, but you get them because they are so easy to give.
It’s also pretty easy to accept the advice du jour, and dig in and attempt to execute on it. After all, they should know better than you, right? Well, it’s still too early to tell where my startup will end up, but I can say that it’s really hard to filter and consolidate all the different viewpoints you receive. Sometimes, you’ll get two opposite opinions, and then try them both.
I think that this is the reason why some startups have trouble focusing. You try to do the ‘right thing’ but 10 people you talk to have 10 different opinions, and you end up paralyzed by indecision or flip-flopping. Advice is sort of like advertising, half works, and half doesn’t, but you don’t know which half.
Who Needs Cofounders? Top 5 Advantages of Being a Solo-Founder
Jun2
In my last post, I wrote a post about some of the disadvantages of being a sole-founder. But, it seems like everyone already knows things are harder when you’re on your own. So, here is a list of the top 5 advantages being a sole-founder.
5. Zero Communication Overhead
The more people that work at a company, the more lines of communication that have to be formed. Mathematically, the number of possible lines of communication goes up factorially. So, even in a company with 3 or 4 people, there is a lot of communicating going on because everyone needs to be on the same page. When you’re on your own, everything sits right there in your head, so you don’t really have to worry about meeting to tell your cofounders what you’re doing at all times.
4. Fewer Distractions
Some people swear they are more productive when in a busy or noisy environment. I don’t buy it. There’s a reason libraries are quiet, and I truly believe that you need to shut out distractions to really focus deeply on something. Ever been debugging code with a mess of variables and state in your head only to have someone come into your office or start talking to you? By the time you look up and answer, you can’t remember what the value of ‘x’ was anymore.
3. Lower Burn Rate
The more people you have, the faster you burn cash. In a post about taking VC money, Hillel at Jackson Fish took a look at revenue per employee at some big tech companies. Now, obviously the more employees you have, the harder it is to have high revenue per employee. As a one-person shop, you are dividing by just 1. In fact, if you are moonlighting, you divide by less than 1, and your effective rate is even higher. Don’t get me wrong, I still do think that in many situations, having N cofounders can create something more than N times as big as one, but if you’re also going to go through cash N times as fast and may run out of runway.
2. You Cant Get Fired/Laid Off
Because, well, you’re the boss. Let’s face it, cofounding a company is a commitment not dissimilar to marriage. And even marriage has a 50% divorce rate. If you are in a group with multiple cofounders and you can’t work things out, you may get displaced. It’s a sad reality that many startups don’t talk about at parties, but if you dig deep enough, you’ll often hear about ‘that guy’ who was there in the beginning but was ousted after a year.
1. It’s Your Masterpiece
When you do everything yourself, from branding to marketing, design to implementation, and everything in between, what you’ve created is a reflection of you. I think this is why most artists work alone. They want their product to be the most pure expression of themselves. If someone asks, “Woah, how did you do that?”, you get to be a bit proud and excited because it was your work, and you know it. Keep in mind the converse is true as well, so when you’re asked “Uhm, wtf?”, you’re going to have to explain that.
Sole-Founder Bugs, Part 1
Jun3
As a current sole-founder, I thought it’d be interesting to write about some of the challenges I face that are worse when you are starting up on your own. I call these ‘Sole-Founder Bugs’.
1. Major Pipeline Stalls
If you’ve never built a company before, or never even done a startup before, you may eventually get to a point where you’re not sure what to do next. In my case, I spent a lot of time working on a prototype to reduce technical risk, theorizing that market risk could be put off almost indefinitely. Now that I have a reasonable beta prototype, I’m realizing that I don’t really know what to do next to mitigate market risk. For a business founder, this might be the other way around. Not knowing the specific set of steps to take to advance your business is very mentally challenging. When you have others in your company, a boss, or investors, there are plenty of people to tell you what you should do next.
2. Single-Threaded Performance
Ok, as a 1-person shop, you’re not going to be able to do multiple things at once. If you have partners or employees, even when you are stuck, others are making progress. If you are working on code, your partner might be talking with customers. If you’re doing IT, your partner can do marketing. This is not the case on your own. You’re doing one thing at a time and odds are, it’s not something you’re good at. Take some time off, and things are usually worse off than when you left them. Servers crash, you forget what you were working on, things get lost.
3. Can’t Solve The Halting Problem
When you work on your own, you generally set your own deadlines. In some cases, you might not even set a deadline thinking you’re “Almost Done” with whatever it is that you are doing. Hours and days can go by and nobody notices. When working for someone else, you’re almost certainly going to have deadlines put in place and it will at least bracket time spent on each task. I’ve spent days trying to make buttons round using only CSS and trying to trim a few megabytes off a memory footprint and lost track of time. It’s easier when someone else tells you to stop.
GetSatisfaction widget not finishing loading in Google Chrome?
Jun0
For a while, I’d been wondering why some pages on my site never ‘finished’ loading in Google Chrome. They would keep the busy spinner active and the mouse would be the busy cursor.
Turns out, the reason was the GetSatisfaction widget was doing something in the background. I found that upgrading to their new widget to version 2 fixed the issue. This was the old include code:
<style type='text/css'>@import url('http://s3.amazonaws.com/getsatisfaction.com/feedback/feedback.css');</style>
<script src='http://s3.amazonaws.com/getsatisfaction.com/feedback/feedback.js' type='text/javascript'></script>
new:
<script type="text/javascript" charset="utf-8">
var is_ssl = ("https:" == document.location.protocol);
var asset_host = is_ssl ? "https://s3.amazonaws.com/getsatisfaction.com/" : "http://s3.amazonaws.com/getsatisfaction.com/";
document.write(unescape("%3Cscript src='" + asset_host + "javascripts/feedback-v2.js' type='text/javascript'%3E%3C/script%3E"));
</script>
and updating the widget to:
<script type="text/javascript" charset="utf-8">
var feedback_widget_options = {};
feedback_widget_options.display = “overlay”;
feedback_widget_options.company = “<your company name>”;
feedback_widget_options.placement = “right”;
feedback_widget_options.color = “#black”;
feedback_widget_options.style = “question”;
var feedback_widget = new GSFN.feedback_widget(feedback_widget_options);
</script>
Fixes the problem. For some reason, I didn’t find any hits on Google about this, so hopefully this helps someone.