Central Perk

miercuri, 23 ianuarie 2013

A Fireside Hangout with Vice President Biden

Your Daily Snapshot for
Wednesday, January 23, 2013

A Fireside Hangout with Vice President Biden

This week, the White House will continue a series of conversations with top administration officials on Google+. On Thursday, January 24 at 1:45 p.m. EST, Vice President Biden will join the latest "Fireside Hangout" – a 21st century take on FDR’s famous radio addresses – to talk about reducing gun violence.

Find out how to watch the Vice President's Google+ Hangout.

Photo of the Day

President Barack Obama and First Lady Michelle Obama greet visitors during an inaugural open house in the Blue Room of the White House, Jan. 22, 2013. (Official White House Photo by Pete Souza)

Announcing Moz's 2012 Metrics, Acquisition of AudienceWise, & Opening of Our Portland Office

Posted: 22 Jan 2013 07:29 PM PST

Posted by randfish

2012 was an amazing year at SEOmoz. We grew tremendously in members, revenue, and employees; finally raised a second round of funding; added some cool features to PRO; and acquired three companies (though, until today, we've only publicly talked about two of those). In this blog post, I'm going to cover our 2012 numbers in some detail and tell you about our very exciting, final acquisition of the year, AudienceWise.

The blog post is broken into several sections for those who'd like to jump around:

Moz's Acquisition of AudienceWise
The Opening of Our Portland Office
2012 Moz Financials
2012 Employee & Customer Growth Data

Moz's Acquisition of AudienceWise

I first met Matthew Brown along with his partner-in-crime at Define, Marshall Simmonds, in Xiamen, China. I was with my grandfather, Si, speaking at SES Xiamen, and the NYTimes' SEO team was hosting us for dinner. I still remember my Grandfather commenting that he'd been a subscriber to the Times for 50 years, and it was about time they took him out to dinner :-)

Since then, Matthew's gone on to start his own consultancy, AudienceWise, with Tim Resnik, former CEO of a gaming startup and a tremendously accomplished marketer & data junkie. Both have been frequently involved in the Moz community - Matt spoke at Mozcon two years ago and will again. Tim and he are both quiet lurkers on the blog, but with this transaction, I expect that to change a bit. Over the past couple years, we've talked extensively about recruiting them to the SEOmoz team. They were on our list of potential acquisitions for our failed funding round in summer 2011, and part of the "use of funds" we spoke about with Brad Feld in our April 2012 round.

Matt and Tim are really here at Moz to help us scale our in-house marketing and product expertise. Both have built software products in the past (Matt worked with Marshall on SearchCLU, Tim on an online poker subscription service) and have tremendous depth-of-knowledge in the fields of both inbound and paid marketing. We have a lot of phenomenal talent at SEOmoz, but only a few of us are deep into the fields of SEO, social media, content marketing, email, CRO, etc. Matt and Tim are here to help serve as mentors and as internal-consultant experts to our entire team, a role that I've been far too busy to fill effectively the last 18 months.

Matt and Tim Visiting the Mozplex

The AudienceWisers have been by the Mozplex several times, but in their first official visit as employees, they impressed a lot of folks on our team and have already jumped into a ton of projects. For example, Tim is working on visiting our rankings data & how we'll build reports for rankings going forward, and reviewing a big secret project that I'm not allowed to talk about on the blog. Meanwhile, Matt's working with Erica to head up our search for great Mozcon speakers, co-piloting the 2013 ranking factors work with Dr. Matt Peters (as an aside, doesn't Matt & Dr. Matt sound like a good sitcom title?), helping with the new version of the Mozbar, working with the product and engineering teams on the web classification system, and much more.

I particularly loved the email Matt sent on the allstaff thread welcoming him and Tim:

Matt's email to team

Some notes on the acquisition:

The total acquisition price (including salary, stock, and deferred payments) is in the low seven figures.
SEOmoz is acquiring AudienceWise's process & products (including some research work and software Matt & Tim have built), the team itself, but NOT the consulting business. Matt & Tim will continue to do a small amount of consulting outside of SEOmoz, and our business will continue to remain free from services revenue.
The AudienceWise Portland offices only hold three people, so we're getting some new space (more on that below).
Technically, the deal closed in mid-December, but we wanted to wait to announce until Matt & Tim had wrapped up their other obligations and started at Moz full time (which happened last Monday, Jan. 14).
Matt will be reporting to me with the title "Head of Special Projects," while Tim will be on Adam Feldstein's product team as "Principal Product Strategist." We're doing more with titles in the next couple months at Moz, so these may change. Neither will have any direct reports, but both will be contributing as consultants/advisors/project leads on a number of teams.

I'm sure that Matt & Tim would love to hear from you and are happy to take questions in the comments of this post, so feel free to leave them, and please join me in welcoming them to the Moz team!

The Opening of Our Portland Office (aka Mozlandia)

We Mozzers have long loved Portland from our perch in the Emerald city. We visit on weekends to sample their insanely weird and tasty food carts and restaurants. We stay extra nights after conferences to tour their far-too-cool-for-Seattle clothing stores. We rant jealously about their much lower cost-of-living and their lack of a state sales tax (which adds to the retail goodness). And, of course, we poke fun at their hipsterdom.

In fact, after watching three seasons of Portlandia, and experiencing the magic that city has to offer, we could no longer resist its pull. Starting in April of this year (probably, maybe May or June depending on lease details), SEOmoz will be opening only its second office ever in Portland, Oregon, nicknamed "Mozlandia."

We've already created a poster of our own:

Mozlandia

Pictured from left to right: Peter Bray (FollowerWonk), Matthew Brown and Tim Resnik (AudienceWise), Galen Huntington (FollowerWonk), and David Mihm (GetListed).

I'm pretty sure this picture alone means our Portland office is going to be an amazing place to work (honestly, Galen looks WAY more "Portland" than his counterpart in the IFC photo). We'll start recruiting more formally soon, but in the meantime, feel free to check out any of the open positions at Moz, many of which teams may be open to staffing in Portland. We will continue to offer our $12,000 referral and signing bonus for software engineering positions in both cities.

2012 Moz Financials

It was a good year for the company financially, despite our focus being on a lot of other issues. We ended the year at $21.9mm in revenue - nearly doubling from 2011's $11.4mm.

I think 2012 and 2013 are going to go down in our history as "investments in the foundation" years. After our funding round closed in April, we spent the vast majority of the year building products that have yet to launch (stay tuned), building up recruiting and onboarding processes, bolstering our product and team with acquisitions, experimenting with how to handle a much larger big data product (Mozscape - sadly most of our efforts to dramatically grow size & increase freshness in 2012 failed, but we believe we now know enough to have success in 2013), and managing culture at a mid-size company (which went pretty well and led to some nice kudos like Seattle's Best Place to Work).

Below is a look at overall product revenue growth from 2007-2012:

SEOmoz Revenue 2007-2012

More than 90% of total revenue comes from SEOmoz PRO subscriptions, with additional contributions from the SEOmoz API (these are combined in the "product revenue" chart above). Mozcon tickets sales and DVD sales are not included in this graph, nor is consulting revenue, which ended in 2009.

I did, however, want to show our expenses for 2012, compare them to 2011, and break them down by category so you can get a better sense of what's in our costs (and see how we're spending that fancy VC money!). I didn't have a great way to show this as a visual graph (pie charts over time are funky - I guess I could have done the stacked graph, but they're also funky), yet the chart conveys the data pretty well:

SEOmoz Company Expenses 2011-2012

There are a few interesting takeaways from the above:

Personnel as a percentage remains the same, and I'd guess it will go up a small amount in 2013.
Hosting is where most of our COGS (Cost of Goods Sold) and pain comes from. It should be going down as we reach larger scales, but in 2012, we chose to invest in building faster rather than going slow and finding solutions to our declining margins. That will change in 2013, and while this year as a whole will probably still be high, we're predicting that our total hosting costs will be ~50% of what they are today (~650K/month) by Q4.
Contractors are a resource we've leaned on heavily in the past, particularly on the development front. That number will probably remain similar in 2013, though eventually we plan to bring the vast majority of production in-house, and rely on contractors only for specialized needs.
Facilities is one that will, hopefully, take a huge leap up in 2013. We need a new office here in Seattle, and while we've been a bit stymied on our first six months exploring spaces (we've had two offices we wanted fall through on us), we want to be moving to a much larger headquarters as soon as possible. In the meantime, Mozlandia will help us grow a bit in Portland.
Not a believer in inbound marketing? If this data doesn't convince you, nothing will. The incredibly low percent of costs that go to attracting traffic and acquiring customers (on the "marketing" line - remember that SEOmoz has no sales team or costs) are a testament to SEOmoz drinking our own Kool-Aid, and investing in sources like content, organic search, social media, email, CRO, and word-of-mouth to spread our brand. It means we can invest much more in research, product, and data. Check out traffic from the last 6 quarters:

SEOmoz & OSE traffic 2011-2012

Some spikiness from viral content skews the trendline a bit, but in general, we're seeing healthy growth from every channel.

If you have questions about this stuff, feel free to ask in the comments and myself or Sarah Bird (our COO) can answer.

2012 Employee & Customer Growth Data

The financials tell part of the story, but a few other data points felt interesting to me and may be to you as well. First up is our growth in employee count from our first year as a software company to today:

SEOmoz Employee Headcount 2007-2012

The chart shows headcount of full time employees at the end of each year. We've obviously had a ton of growth here in 2012, and we're budgeting to add another 66 team members in 2013 (though a lack of new office space may slow that down). What amazes me the most is how well our culture has managed to handle this growth. I feel better about the persistence of TAGFEE and the other cultural aspects at SEOmoz today than I did when we were at 50 people, 25, or 5. To be honest, that's not what I expected. I thought things would get invariably harder and worse at this scale, but given the trend, I'm incredibly optimistic about 150, 250, even 500! Though, I know all of those will take incredible effort to succeed.

Next is our customer growth:

SEOmoz PRO Subscribers

18,731 was the final count of paid PRO subscribers on Jan. 1st, 2013 (our historical numbers for prior years were less precise, hence the rounding).

It's pretty remarkable and truly humbling to have nearly 20,000 paid customers using our product. But we know that we've got a long way to go. In 2012, we had four pretty severe incidents and several smaller ones where critical customer data like rankings, crawl info, or Mozscape index updates were missing or late. We launched a few cool features at the end of 2011 and very beginning of 2012 (social analytics, historical link analysis, universal SERPs tracking, and custom reports) but with the exception of Followerwonk (which is a huge addition to PRO, and continues to develop new features itself), it was a very quiet year for features.

2013 is going to be very different. Our first major launch since Wonk is only a few weeks away, and spring should see the start of many more. We also have an entire team of five engineers, under the leadership of Shawn Edwards, focused on uptime and reliability. The levels of unreliability we've had in the past are unacceptable, and the speed of product improvement is, too. In our reviews for each other this month, Sarah and I were chatting about a large release we've been working on since late 2011, and Sarah told me, "If we haven't launched by June, we should both fire each other." I couldn't put it better myself. This year, we need to kick ass for our customers and be more deserving of the incredible support and growth you've enabled for our team.

January 2013 marks my 11th anniversary working in this job (prior to 2004, I worked with my Mom, Gillian, at the web design/marketing company that would become SEOmoz). I've never been more amazed by what the company's accomplished than I am today, but I know every day from now forward presents the challenge to all of us at Moz - to prove we're worthy of the fantastic things we have (customers, revenue, investors, supporters) and to not be trapped by the mistakes of the past, nor fall prey to the pitfalls of the future.

Matt & Tim will be a huge help, as the teams from FollowerWonk and Getlisted have been already. Mozlandia is going to be an exciting new experiment for us. And 2012 was a great year, but honestly, I can't wait for 2013 to get going, and for us Mozzers to be able to show all of you what the remarkable team we've built can do.

Sign up for The Moz Top 10, a semimonthly mailer updating you on the top ten hottest pieces of SEO news, tips, and rad links uncovered by the Moz team. Think of it as your exclusive digest of stuff you don't have time to hunt down but want to read!

Designing for SEO

Posted: 22 Jan 2013 04:27 AM PST

Posted by Justin Taylor

This post was originally in YouMoz, and was promoted to the main blog because it provides great value and interest to our community. The author's views are entirely his or her own and may not reflect the views of SEOmoz, Inc.

Disclaimer: This is a supplemental post to the Mozinar “Designing for SEO: Improving Site Visibility and Enhancing UX.” Justin Taylor answers Q&A from the Mozinar and expands on the question that was asked most frequently by listeners.

"Why do websites that look great nearly always have SEO that sucks?"

This is the question I set out to answer during my recent Mozinar about designing for SEO.

To be a true king of the SERPs, you are, of course, going to need more than a well-designed, search-optimised website. You are probably going to need a ton of social mentions, quality links, citations and co-citations, etc. The problem with these ranking factors is that they are difficult to accomplish in volume and generally require a lot of work to achieve, and, subsequently, the vast majority of websites don’t have them.

You can, however still achieve big wins with great design and on-page optimisation

Whats the problem?

The problem is simple. Websites that look amazing typically offer little opportunity for on-page optimisation and conversely pages that are well optimised will often compromise the design and user experience.

This creates a chicken and egg scenario - what is the point in having a website that looks great if it can’t be found? And is there any point of being easy to find if the website isn’t engaging?

How can we build sites that look amazing and are engaging, yet still maintain SEO performance?

Enter the webfont

Webfonts from the likes of Google, Font Deck, Typekit, and Fonts.com have been around for a couple of years and offer a great way to give a website style without compromising crawlability. They form the fundamental structure and underpinning of any well-designed, well-optimised site.

To add visual impact, designers will add graphical elements to websites such as banners and calls to action. These elements are usually created as images so the designer can use gorgeous fonts, add type effects such as drop shadows, gradients, and a whole host of other treatments that form part of the designers toolkit. Websites need these kind of graphics, as they make websites engaging, they improve the UX, and they make the user much less likely to bounce.

Take the graphic below, as great as it looks there is too much information to include within an Alt-tag. Also it is difficult to emphasise and prioritise the information within an alt-tag as it is just plain text.

By using a combination of webfonts, HTML, and CSS, it is possible to retain the beauty and achieve good SEO by creating all of the text elements within this banner as “live text.”

Not only can the live text banner now look great, but they can also be marked up with H1’s, body type, bold text, and updated dynamically. Search engines will just see this as standard HTML. Best of all, these banners or graphics can be even be marked up as rich text using schema or microdata.

Design for optimisation

The biggest hurdle in building great looking websites that also have great performing SEO attributes is uniting these two disciplines. Designers focus on sites that look great and create a good user experience whilst being engaging, whereas an SEO typically wants a site which is very crawlable and one which ranks well.

If the design and SEO teams gain an appreciation of each others' requirements, the results can be innovative and outstanding. Take the example below: these panels are for a fashion retailer, the one on the left was visualised by the designer, in terms of UX this panel is great, it shows a model wearing the product, explains through the use of well positioned type exactly what a user can expect to see on click through. The trouble is, from an SEO perspective, this panel does not cut it.

The problems with traditional panels

An SEO is going to need something more like the panel on the right hand side. It has a clear, defined header, possibly an <h1>, followed by some great long tail text. Its clear that from a UX perspective this panel falls well short of the mark, the panel on the left will get a lot more click throughs than the panel on the right.

One potential solution to this problem is a mouse-over. Initially when viewed, the panel will look as it does on the left hand side (exactly as the designer want it), yet when a user rolls over the image the panel changes into what you see on the right hand side (exactly what the SEO wants).

Panels using webfonts

The beauty of this solution is the user experience and click thru are maintained and as all of this text is live text, it is crawlable and very accessible to robots, giving the search engines everything they need to index the site.

The expandable div

Another great way of incorporating indexable content into minimalist page design is the expandable div. It can deliver big SEO and UX wins by making relevant (and crawlable) text visible on mouse click.

Take the example below: frequently, these kind of product panels are represented as images, and they do a great job of engaging the user by offering a visually rich single click method of navigation.

A standard subpanel

With the exception of some alt-text, these kind of panels offer very little for search engines to crawl.

By adding an expandable div to these panels, it is possible to present a host of SEO opportunities. Clicking on one of the items above can now provide a compelling description of this product category, include additional imagery to aid the purchase process and as a result increase conversions and user engagement. But perhaps the biggest bonus of the expandable div is that we can provide search engines with so much additional long tail text to index.

The expandable div offers great opportunities to improve UX and crawlability

The inclusion of expandable divs within web pages are not only great for the user, but also offer incredible opportunities for indexable content.

The concealed weapons!

Calls to action, trust signals, billboards, and all other page elements will often contain the kind of messages we want google to crawl, yet in most cases, these will be represented as images. As a result, discounts, free delivery, next-day delivery, and other offers are not being crawled.

Create all of these elements with webfonts, CSS and HTML so that spiders and bots can crawl them. Do we really want messages such as “Free Delivery,” “10% discount,” or “SALE” to be hidden from search engines?

The following items contain great sales messages, and all of them have been built using webfonts, CSS, and HTML so are all fully crawlable.

Bringing it all together

The techniques outlined all sound great in theory, but in reality can they actually be implemented?

The simple answer is yes, and to prove it, we have built a site using these techniques for a fictitious retailer “random boutique.” The site showcases all of these techniques and uses webfonts, HTML, and CSS to build an experience which is not only pleasing to the eye, but provides a great user experience and provides plenty of data for search engines to crawl. The demo site uses rollups, expandable divs, and live text billboards, plus many other techniques to deliver a site which has a great user experience. However, none are at the expense of the on-page SEO. The site gives search engines a ton of great indexible content without compromising the UX.

The use of live text also inherently delivers some other big wins for mobile, accessibility and multi-language sites, but most notably with the use of webfonts A/B testing becomes super easy. To demonstrate this we have created A/B versions of the demo site which are served from the same URL, share the same code base and look identical to search bots, the only difference between the two sites is achieved purely by the use of webfonts, CSS, and a couple of different background images.

The demo site is available on the following link:
http://www.graphitas.co.uk/randomboutique

Anatomy of a webpage

Armed with webfonts, HTML, and CSS, you have the tools to create amazing websites that contain all of the elements an SEO would also require.

To fully capitalise on these assets, it is also important to understand page structure. Naturally, this will vary depending on your customer, market sector, and the messages they want to deliver.

One of the first things designers are taught to understand is the importance of position of information on a page, calls to action, and trust signals.

A very good parallel for homepage design is a magazine cover. It is no coincidence that most magazine covers are very similar in layout, this is because magazine designers understand the parameters that are most likely to engage people at point of purchase (i.e. on a magazine rack).

They will know that often magazines are stacked on tiered shelves, and therefore they have to have a clear mast head identifying the magazine.They will also tend to use this top quarter of the page to communicate key features/offers. These parameters for magazine design have natural parallels with websites, web designers have to contend with page fold and have to focus more than ever on page position with Google's recent top-heavy algo update.

On and offline, the requirement is to grab your attention. In the case of the magazine, the mast head area lures you with familiarity and offers that are designed to make you pick the magazine up, and once you pick it up, you will see seductive photography and more key offers. The areas these key offers occupy are known as the "hot spots" - essentially tactical positions on the page that magazine designers know will have the best chance of gaining your attention.

These exact same parameters apply to web page design, so with the team at Graphitas, we have put together an infographic which outlines the key positions and hotspots for any homepage and indicates their likely impact.

Download our infographic here

More Mozinar Q&A

During my Mozinar, there were a few questions which I either didn’t get time to answer, or that have required more in depth answers. Here are the fully answered questions:

Q: How does browser compatibility affect the design?

A: Very little. All recent versions of browsers (including IE 6) support web fonts, so design is only compromised by browser CSS compatibility. For example, some versions of IE won’t allow text to be displayed on an angle. In these scenarios, the browser will ignore the CSS entry, and then results in layout issues that we would have to fix by creating a separate CSS for the offending browser. In most circumstances, it is possible to make design exceptions for incompatible browsers without compromising the design.

Q: Are there any good tools for creating “live text” graphics?

A: In truth, I haven’t come across any great tools yet. Tools like Adobe Edge do show some potential, although at the moment the file sizes of the elements it generates are too large, and it is heavily tied into its own webfont library Typekit. We tend to create our own elements as in their basic form these are just classic HTML and CSS which incorporate webfonts and these can be created with any standard web development tools.

Q: Is there a good resource for using “live text” within web pages?

A: Again, there isn’t an all-encompassing resource that caters for webfonts, HTML, CSS, and SEO. There is, however, a lot of resources that deal with these items individually. The best advice I can give is to utilise the work that other people have created and borrow that. The example site shown above shows some great examples of common elements, so please feel free to use the code from there. I am planning to create code snippets for the various elements of the demo site so they can be more easily used. Please drop me a tweet or an email if you would like me to notify you as they become available.

Q: Do webfonts slow down page load?

A: Each webfont is around 50k in size, so overuse of webfonts can affect page load. As webfonts are often used to replace graphics, the end result can be that webfonts could easily decrease page size. My advice is to evaluate the file size during the development process and keep to your existing page size thresholds. Obviously it makes no difference if bandwidth is used downloading a graphic or a webfont.

Q: What about roll-overs and mobile viewing? Do they work on touch screens?

A: Roll overs (or mouse overs) don’t work on touch devices, there are two solutions to this, either remove the mouse over event for touch devices so that the panel just acts a button, or change the mouse-over event to an on-click event to show the text behind the rollover. Either scenario is fine as the text hidden behind the mouse over event will still be visible to search bots.

Thanks for reading, and please leave any thoughts or other tips in the comments below!

Seth's Blog : On behalf of yes

On behalf of yes

Yes, it's okay to ship your work.

Yes, you're capable of making a difference.

Yes, it's important.

Yes, you can ignore that critic.

Yes, your bravery is worth it.

Yes, we believe in you.

Yes, you can do even better.

Yes.

Yes is an opportunity and yes is an obligation. The closer we get to people who are confronting the resistance on their way to making a ruckus, the more they let us in, the greater our obligation is to focus on the yes.

There will always be a surplus of people eager to criticize, nitpick or recommend caution. Your job, at least right now, is to reinforce the power of the yes.