Every month, just after we publish the Seattle Startup Index, there is a few email of people congratulating on the list, on the blog or asking to add their company to the list. This month is not just a few. On the last 48 hours I received about 50 emails. All kinds of suggestions, ideas, congratulation messages, people that just want to say hi, and, of course companies requesting to appear on the list.
We received 10 requests and 9 were approved (we declined one because the company was 6 years old. Not a startup on our criteria). The companies added on the last 48 hours are: MyTypes, Fyreball, nPost, Postacrima, imaPodHead, PeepCode, Intelius and Zumende.
Of all of those, Intelius was our (err, mine) biggest oversight. They will debut next month in third place on the list. They are pretty big. It just seems they are not heard as much on the blogosphere because they are doing some, well, quite boring (yet useful) on the people background check space. It doesn't sound as exciting as Zillow and Redfin on real-estate, or iLike and BuddyTV with media, but hey they do get quite a bit of traffic with all the scare about people stealings other's identity, but I'm pretty sure much of their business comes from large corporations HR's.
Every month that we publish the Seattle Startup Index I get a few emails asking why do we use Alexa and not Compete, or not Comscore, why do we do this if the data is unreliable, that this list is worthless, yada, yada, yada.
Here is a deep philosophical question that I believe Plato asked 2,000 years ago: Which food is more healthy: McDonald's or KFC?
Ok, maybe Plato didn't ask that, and maybe McDonald's is better than KFC or vice-verse, the point is both are pretty bad to your health and people that go to those fast-food chains are not looking for healthy.
Alexa or Compete? Neither. This list is for fun purpose. It has no value for investors, for customers or for their business itself (excepting for bragging rights). Take a look at the list. We compare media, with real estate, with social media, with e-commerce -- Nonsense.
If each company on the list would provide me with their full monthly logs, we could create a serious and valuable result, but I don't think they are willing to do that (and I don't have the terabytes of space to store all that). Or they can give me their Google Analytics data.
Now, don't miss next month's list -- we'll be taking it seriously and we will start accepting bets.
Event: Ignite Seattle [link]
Date: Wed, Aug 8, 2007 (tonight!)
Time: 6:30 PM - 10:30 PM
Venue: Capitol Hill Arts Center - CHAC, Seattle
Price: Free
Presenters: Brady Forrest, Scotto Moore, Shawn Murhpy, Deepak Singh, Sarah Schacht, Maegan Ashworth, Rob Gruhl, HB Siegel, Dave McClure, Beth Goza, Dan Shapiro, Leo Dirac, Brian Dorsey, Scott Kennedy, Jesse Robbins, George Conrad, Elan Lee.
Event: Seattle Entrepreneur Networking Event [link]
Date: Tuesday, August 21, 2007
Time: 6PM - 8PM
Venue: TBD
During Lunch 2.0 at Wishpot, Kabir Shahani (from Appature) asked me to take a look at his website and give some feedback to him. I failed Kabir. I went there the same day, check it out, but didn’t know what to say. Soon I realized the problem was that Kabir has a company that sells one-of solutions, not a consultancy but almost that, and my expertise is with consumer websites. Today, I just received email from Krishnan Iyer (Expera) asking for feedback on his website as well. It’s a consultant website. Gosh I don’t know anything about that business.
Teach them to fish…
Ok, I don’t know anything about websites for consultants or enterprise software, but I’ve built enough websites and web-based services in my life to have an idea of a process I would take to create any kind of website. So instead of telling “you should use a font size of 36pt here”, or “this menu must be left aligned”, I’ll tell how I would approach such a project in “7-steps to website nirvana” approach:
Step 1: Identify the site customer
Pay attention to title of this step. It’s not “who is your customer”, but “who is the website customer”. Your customers are the people using your service or product. Your website customers are the people that need to get some information (or do some task) on the website. They can be existing customers, prospects, partners, employees, vendors, suppliers, etc. Go ahead and create a long list of those. Then, stack rank them in order of highest business impact. Ask yourself the following: “Is it more important for the site to serve well person X or Y?” Quickly you’ll have a rank of what is more important and you must know that to make tough decisions in the future because you can’t have everything for everybody.
Step 2: Identify what the customer wants
Yes, you’ve seen all your competitors’ website and you are drawn by some unknown magnetic force to do very similar to what they do, after all they are bigger and more successful than you are, so they must be right. Right? You want to have the same categories of menus, the same layout, a similar contact us form, etc. That is why so many websites on the same business domain look alike, because the people that created those can’t think for themselves.
The best way to figure out what your website customer wants is to ask him/her. You can guess, and you can get pretty close to it, but asking the customer, either 1:1 or through survey, beats any other method.
Step 3: Create personas
If you worked at pretty much any project at Microsoft you know what personas are. For those that don’t… Persona is going to be the fictional representation of a class of users. For example, if a typical user is somebody in their 40s that is the sales manager of a regional branch of a large retail chain, give him a name, an age, what he wants to accomplish on his career, what are his goals, his non-goals, his worries (personal and professional), what are his personal gains from using your site. If you end up with 7 personas you should start over and go to step 1. You should have 1-3 personas that represent the typical visitor of your site.
Step 4: Identify what you want them to do
You know why they are important, you know what they want, you know who they are, and the next step is for you to define what the ideal scenario is for you. Is it for them to purchase something on your site? Or call you? Or find the information that they need without calling you? What are the top 3 data points or tasks that each of your personas will find most useful for them? They might overlap between each persona, or they might be 100% different.
Step 5: Get a pen and paper and diagram your site
You can turn off the computer now. The next step is for you to create diagrams of the website. First, list all the content that will serve the personas from the list of the previous step. Then, using a pen and paper, write Homepage in a box, create arrows and new boxes representing new pages, like “Contact Form”, “About Us”, “Our Services”, “Sign Up”, from your list. You don’t have to list all pages, but the ones that are most important. Then, from the pages 1-click away from the Homepage, draw the second level pages, if any is needed. Now, with a pen of different color, create the typical user flow through those pages. For example: User comes to Homepage, clicks on “Product Information”, clicks on “Technical Specification”, clicks on “Contact Sales Team”.
Step 6: Start sketching the website
You don’t need an expensive designer for that. You don’t even need a cheaper design as matter of fact. You are not creating beautiful images, or perfect text. You are just defining from what you learned on step 5 how things are flowing on your site. That automatically defines the menus you must have, the action buttons that should be prominent on the screen, what are the three or four clear divisions on the homepage that will address each of your personas. The biggest mistake I see on websites with regards to the distribution of information is to make everything of the same style or making everything of different styles. What you want is to highlight the 3-4 things you want the user to do next, but if you highlight too much on the same screen, the value of highlighting just got diminished. I learned that lesson back in the early 90s while doing Desktop Publishing work: If every paragraph in a text like this has 1 word in bold, the value of having anything in bold is gone.
Step 7: Layout, Text and graphics
Now is the time to bring in the professionals. Hire a great web designer and great copywriter. Both are hard to find, but it’s absolutely worth having them. About designers, make sure he is a web designer, not a designer-designer. You want somebody that understand web, not somebody that is great at creating magazine ads, or posters, or videos. Designers are creative people. They understand beauty, symmetry, gestalt, etc. But the web is a tough medium with lots of limitations, accessibility and usability gotchas. Web designers have been through that before so they are less likely to give you a mockup that can’t be implemented. Copywriters are also great at making sure the right words are being used at the right places. If you can, hire a marketing firm to help with that, but if the price tag is too high you should hire a good copywriter. What you should never do is not get feedback on the text. Site owners tend to focus on the graphics and layout more than on the text while doing and asking for feedback.
Step 8: Go live, get feedback, repeat (free bonus step)
This is what most people already know. You can only improve on something that already exists. So, put your website out there and bug your website customers for feedback. Forget about asking friends, family or colleagues. They might give you feedback that is completely wrong, although with very good intentions, and distract you from what is right.
January, 2007
My Browser
Del.icio.us
Digg
Reddit
Furl
StumbleUpon