SearchLove Boston 2012 - Day One Recap

Didn’t have the chance to join us for SearchLove Boston 2012?

Too enamored with our charming and brilliant speakers to remember to write anything down?

Fret not, I spent day one of SearchLove Boston 2012 feverishly taking notes, wearing my fingers down to the bone and grinning appreciatively all the while. Just for you. This post recaps the first day of brilliant presentations.

Quick Jump Menu:

Intro from Duncan Morris back to top

Duncan, Distilled Founder and CEO, kicked the conference off famously with his unique brand of deadpan snark and a recap of the history of Google’s algorithm updates. Recently Panda/Penguin. MayDay before that. Florida even earlier. And so on.

And in the scramble to adjust strategies and tactics in response to these updates, what Danny Sullivan calls the Google Dance, Duncan holds up a quote from Ian Lurie (who’s speaking today):

Not one change in the last ten years has affected the way web marketing really works.
Duncan points out that while search volume for “Content Marketing” has spiked in the last year, it’s still nowhere near the volume of “SEO” - and we need to keep perspective about that.

Mobile is exploding and multi-device consumption is the new landscape for web marketing. Almost half of the traffic to online Olympics coverage (41%) was from a mobile device. Video on mobile is perhaps the biggest part of this.

With that, Duncan introduces our first session - the Breaking News Panel:

Breaking News Panel - Ian Lurie, Will Critchlow, Annie Cushing and Rob Ousbey back to top

Q from Duncan: Who thinks the disavow tool is a good thing?

Ian Lurie does. Harder to game than you think. His biggest concern with Penguin was that he’s seen several businesses that acquired businesses or had someone doing shoddy linkspam on their behalf, and the control was too much in the hands of spammers. He thinks the disavow tool puts power back in the hands of legitimate businesses.

Will/Duncan say there was an attempt of negative SEO on Distilled.net, and they have no intention to use the disavow tool as there was no negative impact. Will considers it an “input” - i.e., not a control mechanism, more of a “suggestion” tool and, perhaps more importantly, a data source for Google.

Ian: This is a tool of last resort.

Rob: Shouldn’t Google just figure it out for themselves, which links are ‘bad’ and remove that value.

Ian: Sure, they should. Ian thinks it’s perhaps the best solution out there.

Annie: Google has some complicity in the value of spammy links, because they worked. When the competition is doing it, and it’s working, in order to compete businesses have little choice.

Annie advocated for the disavow tool after seeing clients “put it all out there” in reinclusion requests in good faith with zero response or redemption from Google.

Will: The name is interesting (“Cuttsian”) - “how many people use the word ‘disavow’ in daily conversation?” It’s a data play - the tool is not just going to help businesses, it’s going to help Google.

Ian: I don’t mind that so much. I’m a classic white hat and have nothing to worry about in terms of “bad” links. I don’t build them.

Panel agrees, on a question from the audience, that proactively disavowing links isn’t recommendable. (Perhaps unless they’re really, really bad links.)

Annie: Heavy use of the disavow tool could put up a flag for Google - it puts you on the radar as an SEO.

Duncan: Wil Reynolds posted on SEOmoz about this - how Google allowing bad links to work makes white hat SEOs look like fools.

Annie: Penguin is making it easier to pitch creative, interesting marketing to clients.

Rob: Spam is still rampant, though.

Will: It’s made smart black hat more effective and removed some of the less clever black hat competitors.

Ian: A lot of my job as an Internet Marketer is risk management. He doesn’t mind being able to say “told you so” when it comes to linkspam.

Will: I believe Google would have rolled out Penguin even if it made the results worse - as a PR play. They were taking a beating in the tech press about linkspam. They had to respond. There is actually a high correlation between quality and paid links, because big, legitimate businesses could afford to buy links - and the incentive was there. Their competitors were doing it. Google created the link economy by allowing it to work.

Question from audience: “How much have these algo updates impacted Google’s international properties (not just US and UK)?“

Will: Western/European markets seems to be the focus. One of the reasons Penguin impacted US so much was that there was a bigger search market in the US. There isn’t enough of a market in Finland, for example, for a content farm to make money. So either the paid links signal was less significant and the algo update created less flux, or Google didn’t bother with those properties where the problem was less rampant. Google’s webspam team is also English-speaking in general, and tackling webspam in other languages may be a challenge for them.

Question from Duncan: “In two years, are we going to see Facebook/Yahoo showing up as sources of organic search referrals?”

Ian: Facebook? No way. They suck at search. Yahoo is unlikely as well. But Siri (Apple) is a search engine, he’s expecting that to play a part (and Siri-like services from companies other than Apple).

Will: Google will move more in that space as well. Don’t bet against them.

Ian: It’ll be a simplicity play. Google won on simplicity when they emerged. Nobody’s got this right yet for Mobile.

That concludes the Breaking News Panel. Great stuff from the panel. Up next, our first presenter, Dan Shapiro, who’s going to tell a few stories from his startup career around creating serendipity for your business.

Dan Shapiro - ”Creating Serendipity“ back to top

Dan has worked for big companies and tiny companies. He’s learned some interesting things and seen some trends.

At a big company, you think about strategy. You work on strategy and then on execution.

At small companies you have to get lucky. And you can manufacture serendipity.

Dan used to shirk it, or simply “wait for it to happen” - then he learned to go after it.

At his photo software startup, when they went to conferences, he and his co-founder would “end up” in the coffee line behind someone who worked for a company they could partner with.

But the people they really needed, investors, were hard to find.

In 2006, two years before the mortgage crisis, large investment banking companies threw the best parties connecting startups and investors. Lehman Brothers had a big one in particular. To pave the way for successful IPOs, Lehman invited investors and promising startups. Dan’s startup wasn’t one of them.

So they crashed the party. They went to Barcelona, where it was being held, and walked right in.

An investor from Seattle (their hometown), was there. Dan ”sauntered up,“ pretended to have no ulterior intentions, and ended up striking up a deal - Tom (the investor) joined Ontela (Dan’s startup).

Dan’s 2nd startup: Sparkbuy (comparison shopping for consumer electronics - an affiliate model).

Dan has a rule: if he gets a first class upgrade, or is traveling to/from the valley, he forces himself (doesn’t enjoy it) to talk on an airplane.

He met Matt, from Google, on such a traveling occasion.

Dan told Matt about sparkbuy (just an idea at the time), and Matt asked him to email him if the idea moved forward. A little back and forth via email, then quiet for months.

Dan and his team at Sparkbuy soft-launched (beta) on TechCrunch. Matt emailed Dan with congrats and mentioned he was in the market for an ASUS laptop. Dan followed up with some funny bits (usability studies show 30% of Americans pronounce ASUS “asses”). Matt suggested lunch.

Long story short: Google acquired Dan’s company.

Now Bill believes in luck, and that you can create it. He has three principles around luck:

  1. Fish in deep water It’s impossible to predict what’s going to move your business forward, but you can often predict where and when they might (conferences like SearchLove, for example)
  2. Triage quickly When in business mode, gracefully/simply move on when a conversation is unlikely to pan out into any opportunities. Hard to do this (Dan doesn’t like to be judgmental), but it’s important to use time efficiently.
  3. Learn the one, magic phrase - “How can I help?” Note: this isn’t a commitment to help (it may not be plausible), but creating serendipity for other people has a funny way of coming back around

Q and A

Q from Rand Fishkin: “What can you share about the sparkbuy acquisition?”

A: “Not much. But he’s working on new products inside Google (cars and credit cards) that are based on the basic Sparkbuy model of comparison shopping.

Q from Will Critchlow: “What kind of serendipity are you trying to create now?”

Dan: “Inside a big company like Google, it’s no longer necessary.” He has access to resources and brilliant people. He can focus on building products and experiences he loves.

Q from audience: “Where is all of this [Google] shopping stuff going, in five years?”

Dan: I can’t speak to the overall plan, but I’m passionate about providing a fantastic experience to users. I’m focused on a small set of queries in that pursuit. “I’m not worried about corporate strategy or promotions - I’m worried about how I can build a brilliant experience for people who use our services every day.”

Q from Duncan Morris: “Does success come from serendipity or being ‘good?’”

Dan: Serendipity plays a huge role, but you can help it along. You can manufacture it. Some people stumble into it (see reality TV stars), but most people leverage it, whether they realize it or not, as something we can influence. “You have to keep your ears open for opportunity.”

Great stuff from Dan. Onto our next presenter, Distilled’s own COO, Rob Ousbey, who’s going to educate us on the characteristics of a successful outreach campaign.

Rob Ousbey – “Characteristics of a Successful Outreach Campaign” back to top

Rob’s presentation slides are available on SlideShare.

Rob’s going to talk uncharacteristically fast (thanks Rob!).

The conversion funnel for outreach:

  • Prospecting
  • Contacting Communicating
  • Conversion
Priorities for Outreach:
  • Build Relationships
  • Get Coverage
  • Acquire Links
Distilled has been doing outreach for years, but the team has grown quite a bit over the last year.

Rob wants to share the characteristics that Distilled has seen as highly correlative to successful outreach.

1) Get Buy In

Get buy in from your client’s internal stakeholders (or your bosses and their bosses if you’re in-house).

Rob wanted to get photos from a client of their data center, but it was incredibly difficult. Rob wondered why they couldn’t provide pictures, and the reason was because they had no buy in -

Flywheel of Personal Credibility. Keep pushing. You get results, you get more buy-in, get more access, better results, they like you more, so on.

Maybe even “hack” the flywheel, get some initial impacts (quick wins) however you have to, leverage that to get buy in for the work you really want to do, the big stuff.

2) Budget

Content Marketing is risky. Risk-averse people like PPC because you can scale up when something works. With big content pieces and outreach efforts, too small a budget can neuter the results. The risk of a big budget might be the only way to give it a chance for success.

Flywheel of Brand Credibility. As you publish/promote good content, you get credibility, an audience who amplifies outreach.

You can hack this one as well - use your website to demo credibility. Measure it and optimize it.

3) Access to Influential People

The notable people in your client’s companies, or your own, can carry more clout. Their Guest Posting efforts can be much more successful. (See Salesforce.com’s Marc Benioff, who guest posts to much success.)

4) Avoid “PageRank Magpies”

Silence the people who are ultra-focused on quantitative metrics (like PageRank). These are not telling to good brand and long-term value (and lateral benefits) to outreach.

5) Understanding of the Good Brand

Help your outreach team learn more about the brand and consumers. Find the target market, find the positive spins for the brand’s service/product.

Outreach is, essentially, online PR. Consider the real audience for the brand, connect there.

6) Gatekeepers

Understand the motivations of the bloggers/publishers in the niche and how you can align the brand with those motivations.

7) “Good Niche”

  • This audience needs to like the brand (or have values aligned with it)
  • A niche that might respond well to sampling the service/product
  • A good niche has good network effects (bloggers who know and like each other)

8) Product

Access to the product can be key in outreach - distributing it for free for reviews, for example. Distilled had a jewelry client. They identified bloggers who were a good fit for the jewelry and distributed it with a note.

Look for cake mix / add your own egg. Set things up so that the person feels they’ve done “just enough work” to feel accomplishment/ownership. Instead of just asking them to link, ask them to write a review, for example. Takes the transactional/awkward nature of the interaction away.

9) Great Content

Advertorial content usually fails.

Distilled had a client who created a self-promotional infographic. Failed miserably.

Competitive Aspect - leverage the natural competitive nature of tribes (vs other tribes) to get them interested in content.

“Apple users are more generous when giving to charity” (Razoo.com infographic) - generated a lot of coverage b/c it was controversial/competitive.

10) Why Now

Have a good answer to the question, “why I’m contacting you now.” “Because I was hired to do outreach” is never the right answer. Think:
  • new range
  • news worthy
  • special event
  • seasonality
Align Outreach with Content Planners and Creators Ensure scheduling lines up. Maintain agility (to repurpose/reform content on the fly for a given audience/placement).

Be prepared to reformat the creative Distilled switched monetary values in an infographic from dollars to pounds for a placement, “greased the wheels” for that placement.

Turning static infographics into video/interactive can work well.

11) Strategy

Don’t just create content. Create a strategy first. For a Distilled client who provides insurance for small businesses, Distilled created the following strategy:
  • Expert content that positions them as teh most knowledgeable in the business insurance niche.
  • Evergreen content for the small business that’s getting online.
  • Timely content for the business owner who needs up-to-date information.
Using the above strategy guidelines for content creation, Distilled created unique and interactive infographics that let to significant growth in the client’s search visibility and traffic as a result.

12) Plan & Calendar

Set a campaign plan over the year, around seasonal/special events, and around resources. Set launch dates. Q4 for example, is a tough time for outreach (Dec is totally quiet).

13) Pre-outreach

Begin kindling relationships, or even getting feedback for content ideas, with placements/bloggers. Makes the promotional aspect later much more successful.

This can also help you get a sense of whether there’s actual interest in the content. If there isn’t you can adjust or scrap the strategy (and save yourself and your client resources/heartache).

Micro Conversions Getting some initial feedback or a “yeah that sounds interesting” makes it more likely someone will follow-through later with a link. Get a small “yes” first, the big one later.

14) Little Black Book

Treat people like people. Keep in touch. Do things when you don’t need something from them. It’ll pay back. Conversion to Contact Use Facebook demographic targeting to advertise to bloggers (Rob tried this with fashion bloggers). Low CTR, Poor conversion rate, campaign moved slowly. Spent $70 - got two useful contacts out of it. Conversion to Links Carson Ward writes a philosophy blog. Spent some money on ads (AdWords, StumbleUpon, Reddit). $180. 15k visits. 192 linking domains (154 were followed). 154 new relationships.

Q and A

Q: How can we help visually report to clients on the results of our outreach efforts? A: Recommends not getting too focused on ROI b/c there is a chasm between the outreach effort and the measurable outputs we’re looking at in analytics. A lot of the value is missed. Better to show relationships built, direct traffic/links built, and then show the SEO metrics on the other end (since outreach is generally done with SEO in mind, if not as the central goal). Q: How can we scale outreach efforts? A: Doesn’t scale easily. But building up the little black book helps (contacts are everything in PR, after all, and outreach is essentially online PR). More people adding relationships to the bank creates of a flywheel of its own (everyone on the team has more opportunity). Q: Outreach on behalf of client, or agency? A: On behalf of client. Used to pose as in-house, not anymore. Totally transparent. Main reason is scale. They get to use that relationships again for other clients.

That’s all from Rob - go forth and leverage this new outreach knowledge! We’re breaking for refreshments, then back with Ian Lurie, who’s going to take us to school on creating better proposals.

Ian Lurie – “Making your Pitch: Writing Clear, Compelling Proposals” back to top

Ian was kind enough to compile his slides and resources on the Portent Interctive site here: portent.co/ian-proposals (thanks Ian!).

Kids: Nature’s way of teaching us our parents were right. Because they ask, “Why?” Kids are also nature’s way of teaching us to write proposals. “Because I said so” doesn’t work. “Because I f**king said so” also doesn’t work.

The instinct with proposals is to write down everything you do. That massive description is basically a “because I said so” because it isn’t designed to be read. It’s designed for the reader to give up and say, “OK I get it, it’s complicated and you do it all.”

Most proposals answer What and How.

You must answer Why.

Read Simon Sinek’s Start With Why (here’s Simon’s TED talk on the topic).

“Why do I get up in the morning and go to work.” - You need to be able to answer this or you end up with crap mission statements.

“To make money” is a crappy Why. “To help all people live healthy lives” is awesome.

1. Getting the Why

“We are in business because we believe that _________”

Ian like to structure it this way: “if you _____________, then we’re your agency.”

Outcomes are the What not the Why.  Money/ROI/Profit/Revenue - all Whats.

Whys don’t apply to everyone.

“We are a full-service digital agency providing PPC, SEO, Analytics...” - Ian used to do this

“We are in business because we believe that great marketing can save the world” - this is what he does now.

“If you think what you do matters, and want to work with folks who know that means it all matters, then we’re your agency.” WHY = VALUE

2. Imputed Value

The sum total of the details around an experience are what give the customer reinforcement of value. HOW||WHY (parallel). This is where the little details come in. It all matters, it all reinforces the Why.

Tools Ian uses:

  • Powerpoint/Keynote for all proposals (don’t use MSFT Word)
  • Picks a few great fonts (Ian buys them - they’re worth it)
  • iStockPhoto for animal photos (Ian is known for his animal photos)
  • Skitch for markup
  • Excel
Research Ian likes (on the client):
  • Look at how they promote themselves
  • Look at who works there (LinkedIn) - seem like people you’d like to work with every day? It’s important.
Use “You” & “Us” & “We” - talk directly to the client. Make “eye contact.”

Have a personality. Don’t be afraid to be quirky in your proposals. When clients say “we felt you really got it” when they hire you, that’s personality. Don’t be afraid to use words like “suckage.”

Get to the point. Avoid words like “majority of” or “clearly” or “in order to” etc - use “most” or “to.” Use active voice. Read Writing with Style by John Trimble.

Illustrate. Mack Web does a good job of illustrating process. Visualize dry/boring process discussions, it gives people a break.

Don’t assault the client. Break it up. Don’t try to overwhelm them with bulk. Use whitespace.

Avoid stock photo abuse. No snails having sex (Ian I’m surprised at you). Generic crap is obvious. The most attractive/diverse group of people I’ve ever seen? Right. Be genuine.

Use data sparingly (and well). A ton of data is a “because I said so.” It fails. Make it really clear. Use spare data points that get right at the why in an obvious, direct manner. Line spacing / formatting. Break things up. Use bullets and lists.

Be consistent. Please. Keep formatting between slides/pages consistent so the user’s reading/viewing experience is fluid and not jarring.

Use PDFs. The person reading it will have different fonts, or read it on a cell phone, or be running a new tablet. People will find a way to ruin the format. PDFs lock it in.

3. The Wrap Up

Know the Why. Otherwise you’re phoning it in. It’s uninspiring and clients will be uninspired to hire you.

Build imputed value. Care - a lot. All the little details matter. Pay attention.

WHY||HOW. Make sure there are parallels. This is about more than just proposals, it’s about marketing overall.

(Special thanks to Ian for speaking slowly and clearly and saving my fingers for the last 40 minutes.)

Q and A

Q: This is great for new prospects - what about getting buy in from existing clients for different efforts? A: I do it every day. It can be difficult, but keep at it. Communicating with clients around their own Why will get through to eventually.

Q: Do you integrate competitive analysis at all with your proposals? A: Depends. We’ll talk about opportunity gap, but again I always tie it back to the client’s Why. You don’t close a competitive gap by undercutting on price, developing a slight advantage tactically. Tie it back to “here’s why you’re different and here’s how you make the competition irrelevant.”

Q: What happens when you go down a road (with a Why at the core of the strategy) you thought was good, but it ends up not working well with the client? How do you address the tension? A: I’m not the only person at Portent, but I’m not in it for the money. If it’s not working you need to deal with it. I’ve had discussion with clients about the grief/tension in the relationship and getting at what can be done. Communicate - silence from the client does not mean things are going well.

Q: How do you narrow down / cherry pick what should go in vs what should be left out of a proposal? A: Character as a company has to be there (it’s the Why). Also talk about what they’re going to get (what will be the outcomes) - this isn’t the centerpiece, but it’s got to be in there. Focus on opportunity gap. Thirdly, the How - processes (visualized), how things will progress, etc.

Q: How about RFPs with specific format - how do we insert the Why? A: When you get an RFP you know you’re not talking to the right person, it’s someone down the chain. Little chance to differentiate. RFPs suck. They kill the Why Look for opportunities to attach additional documents - it’s a chance to differentiate and deal with the Why.

Seriously useful tips from Ian (as always). Up next is Jen Lopez, SEOmoz’s Community Manager, who’s going to show us how to leverage social media for customer acquisition and retention.

Jen Lopez – “Utilizing Social Media for Customer Acquisition & Retention” back to top

Grab Jen’s presentation slides on SlideShare.

Why is the community girl talking about acquisition & retention?

We generally think of customer acquisition in terms of $$ - ads/SEO/content marketing.

When we think of social, we think of broadcast. “Tell all the people!!”

Jen wants to talk about a way to earn customers through social without spending a ton.

In-App Sharing

FollowerWonk (SEOmoz tool), when users sign up, enter their email, they can check a box and tweet about it (and they do - and it generates retweets and favorites).

Shareable URLs

Simply Measured creates short, shareable links that Jen uses to tweet with - makes it easy for her to promote their reports. Note: the content has to kick ass.

Show some data

When users share your stuff, show them the results. Show them how many people clicked the links they shared. This gives them some incentive/motivation to keep sharing.

“Pay” with a Tweet/Share

Oli Gardner, with The Noob Guide to Online Marketing, used a “pay with a tweet” mechanism to aid virality. Low-friction conversion for a high-value report. Results were through the roof.

Social Proof

Show people what their friends are up to on your site. Jen logged in to TripAdvisor and saw David Mihm had reviewed a B&B in Ireland. Not only aids the products/listings, but helps reinforce the user’s engagement and familiarity with your site.

Connect G+ to AdWords

Bring social proof into your advertising. Improves clickthrough rates.

Facebook EdgeRank

Leverage this. Jen doesn’t like that Facebook makes us pay for people who already “like” our stuff to see more of our stuff, but it’s their playground. Promote posts, but make sure they’re valuable, not utterly self-promotional.

Uhm...Ask?

Barack Obama DMed Jen. She clicked, then tweeted what they asked her to tweet. If you have enough clout, simply asking people to promote can create a “street team” effect - they’ll help just to be associated with you (or feel associated with you).

Community

Use the info you gather from users to make it easy. Tell people, for example, when people’s birthdays are coming up. Minted does this (they allow you to create/send custom gift cards for friends).

Can people find you?

If someone wants to reach out via social channels, how easy is it? (Seriously, go check if your companies Twitter/LinkedIn/Google+ profiles are clearly visible and linked to from your website.)

Take it Offine

Send cupcakes (for example). Take it offline. They’ll take it back online to thank you.

Customer Service

Use social for it. It’s quick. Be responsive.

Empower the Manager

Make people’s lives easier. They will love you for it.

Test it all

Don’t just assume the “best time to tweet” tips out there, while measured and interesting, fit for your audience. Run your own experiments and find out for yourself. Yours will be different.

Q and A

Q: I feel spammy posting the same article more than once per day. Advice? A: Go ahead and post multiple times, but make it unique. “In case you missed it earlier” helps.

Q: G+ for industries outside SEO? Use it? A: Definitely. Not as much viral potential, but when you mention someone by name (with + symbol), they’re more likely to see it and respond.

Q: Use cases for LinkedIn? A: Many sites get the majority of their traffic from LinkedIn. It can be a very rich channel. SEOmoz spends more time on LinkedIn than Google+. There are a ton of people asking questions on LinkedIn about SEO, and it’s a great place for us to do some outreach.

Big thanks to Jen for some seriously actionable tips on leveraging the social channel to attract and keep customers. Up next...lunch! Then we’re back with the ever-ignited Wil Reynolds, who’s going to scold us for chasing algorithms and show us a better way.

Wil Reynolds – “Chasing Algorithms – Think Deviously, Act Angelical, and Never Get Hit by a Penalty” back to top

Get Wil’s presentation slides here.

Wil’s glad we haven’t been talking algos so far, because we shouldn’t be.

Wil likes maximizing his time, not wasting it.

Algo view from a non-techie

Wil is not technical. He finds it’s a good thing when working with clients (they don’t care about the tech details).

Google has us outnumbered on the PhD/Tech thing. We’re not going to outsmart them.

“Marketing = Formula?”

Persuasion (people) vs Manipulation (spiders). We’ve trended toward the latter, mistakenly. We became like mad scientists with zero social skills.

“Too Transparent” - Wil says there is no such thing. So he’s opening his kimono:

  • Wil cloaked (gasp). He was told it worked, so he did it.
  • He did white text / white background. He thought he was helping (being angelic). Then he saw the kind of garbage the was ranking on those same tactics.
You may be using a tactic for good, but if others are abusing it, it’ll be devalued eventually.

In 13 years of SEO, Wil was surprised one time by a site that was hit (a site he wasn’t expecting to get hit). His own. SEERInteractive.com was banned in Google for 12 hours.

Wil’s take on directories: Never saw them as a “bad” thing. The web can never be that simple. Nobody here has a PhD, but we all have an opinion on the algo.

Wil did 80% directory link building for a bit (2005), and it worked. He thought, “too easy.” In 2007, he was moving to 50% directory link building, 50% other tactics. 2009, he did only 30% directories. About the same time Wil 301’d the old domain to the new SeerInteractive.com. Then they got hit.

Even with links from Google, Inc., Rackspace, Microsoft, CNN, CNBC, Basecamp, etc, they got hit.

“The interesting thing about building a real business is that at some point you no longer care about ranking for anything other than your brand name.”

It was the links to ThinkSEER.com (their old domain) that got them penaized. They were heavy on the anchor text for “SEO company.”

RCS won’t save you. SEER was legit, got hit.

Let’s walk through what happened:

  • SEER had a client in a highly-competitive niche. Client was ranking well. They built a crappy microsite (didn’t tell SEER), built links to the microsite with link networks. That didn’t work so well, so the client linked to the crap microsite from their home site.
  • Microsite got hit, client site got hit - then SEER, for linking out. (Mike’s note: Really? This seems unlikely to me.)
  • Wil was told ThinkSEER.com was toxic (by the “powers that be”). He didn’t believe it.
Why Wil didn’t try to crack the algo:
  1. Time is limited and better spent elsewhere
  2. You never really know. You just don’t. It’s a black box.

4 Realities of Google Patents:

  1. They may never implement it
  2. If they do it could be 8 years from now
  3. If it doesn’t work they’ll yank it
  4. Surfacing Quality Content - this is what they’re always trying to do. Every algorithmic change is aiming at this goal.
Algo = rewarding quality. So spend your time on quality.

How many algo updates purposefully reward thin low quality content?

Quality content is how you protect yourself. It’s a lot easier to get out of trouble (that’s why SEER was back in 12 hours). A manual review by Google proved SEER was legitimate and deserved to be included in the index.

Wil noticed the 3rd largest referer to their “how X makes money” infographic was m.[siteyouknow].com. Responsive design FTW. Seer was able to provide custom-dimension versions of the infographic on the fly because it was designed to be responsive to the viewport.

Wil tweeted the infographic once. @avinash retweeted, then @theverge, then @gizmodo - and commence viral liftoff.

1 good piece of content. 1 tweet. All it took.

7700 links in 38 days. 500 LRDs.

The infographic has a life of its own. Wil checked Topsy recently, weeks after publishing/tweeting the piece - 5 people had tweeted about it in the last hour.

One good reason to follow algos: Protect yourself from affiliate/partner risks. Partners/affiliates are important, but they might engage in shady stuff - and get you hurt as a result.

Pro tip: Buy an affiliate’s site when they’re hurting, turn it around (with RCS).

Directories and articles can’t kill a quality site. Reciprocal links same. If search overall would break if the tactic was penalized, it’s not worth worrying about. Footer links for example - it’s happening everywhere. So how can it be a quality filter for Google? Widgets are the same, as well as Infographics.

None of these tactics is “bad” by nature. The algorithms are simply not that simple.

Don’t put authorship on non-authored pages. A sales landing page is not an article. “CEO as author” of the site home page?

Algos are becoming like a “swinging axe” - you might get through one pass, but the next time they might catch you. Iterative. Gets smarter every time.

Chase DISPLAY algo changes, not ranking. Think KnowledgeGraph. Universal results. Google scraping book/movie/etc info and giving the answer directly (cuts the business out). 7 results instead of 10. Etc.

Q and A

Q: What do you think is going on with on-page, visual/CSS issues, now that Google is rendering the full page? A: The minute you’re talking about “SEO content,” or content that’s there for SEO reasons, you’ve lost the fight. And think of a sub-nav, it’s like a panel of hidden site-wide links. No way that’s going to get someone penalized.

Thanks to Wil for his usual energetic and engaged approach to a compelling and important topic. Next up is Distilled’s own Mike Pantoliano, who’s going to highlight the importance of combining SEO and CRO for better results, happier clients and building a more complete skill set.

Mike Pantoliano - “CRO & SEO – Better Together” back to top

Get Mike’s presentation slides right here.

Mike tries to be a complete marketer - not an SEO. A given SEO tactic has a shelf life, marketing does not.

In 2009 we weren’t talking about inbound, RCS, etc. We were talking SEO.

Meet Mark - an in-house SEO in 2009. He was doing good, white hat SEO.

But the ROI Mark was getting was hitting a point of diminishing returns. Same effort, fewer and fewer results. The low hanging fruit were exhausted.

CRO is how Mark can get passed this.

Reason 1: Make your client more money. This is the obvious reason, but important. SEOmoz made $1 million extra hiring Conversion Rate Experts to do CRO for them. Rand later said this extra revenue allowed them to build Linkscape (now their flagship product).

Reason 2: We like to talk art & science in SEO. We do analysis, but we also brainstorm/create unique, interesting content. This is the same with CRO - there’s science, but there is qualitative work involved.

Reason 3: (Not Provided) - there could be a point where our keyword reports (organic) are worthless. So let’s focus on content reports. Where are people who come from Google organic landing? Where do they go from there?

Reason 4: Panda. Panda is an algorithmic attempt at understanding user behavior on your site. If we’re doing CRO, we’re looking at the user experience and improving it. Algo updates like Panda, as a result, will be less likely to negatively impact us.

Reason 5: All the traffic in the world from organic search means nothing if it doesn’t convert.

Reason 6: By doing CRO and testing, we’re developing an innate understanding of customer behavior. This makes us valuable. It’s market research. SEO : Market Research : CRO.

See the Highrise (from 37signals) write-up on their headline updates (for CRO) on their home page.

Reason 7: You’re watching your presentation. You can’t claim ignorance anymore. But you need to test things yourself.

Stephen Pavlovich is a CRO expert, not because he reads blogs. Because he tests and has experience.

Read The Lean Startup. Eric Ries talks about the Build > Measure > Learn feedback loop. Most of our processes as SEOs misses that Learn piece.

See Stephen’s presentation, “What a CRO Expert Bookmarks” (from SearchLove 2011, available for purchase in the Distilled store).

Make friends with designers

SEOs don’t have much of a design sense (overall). Bringing designers to bear on SEO and CRO efforts is valuable. Often there will be assets that weren’t used, or were used in the past, based on some previous research/planning - and these can be brought to bear on new CRO efforts.

Ask the right questions

Gathering feedback (with tools like Qualaroo4QSurveyMonkey, Sales/Support teams etc) generates essential insights.

Mockups/Wireframes are powerful

Balsamiq/Notable/Mockingbird are three great tools for this process - get feedback and, more importantly, buy-in for what you want to test/launch

Solidify allows you to create interactive flows of mockups - for checkout flow, etc. Great for multi-step conversion prototypes

Tools of the CRO Trade

Get snippets integrated early (so you’re not delayed by the IT bottleneck when you want to launch your tests)

Best Practices* (*Ya gotta test)

Testimonials - careful not to use stock-looking photos
  • Twitter testimonials are great (they’re verifiable, you can click and see the Tweet)
  • Align testimonials with personas - so potential customers can self-identify with one of them
Holy Cats - Headlines are important!
  • Simplicity
  • Specificity
  • Emphasize “you” not “we”
Optimizing page loads is doing CRO
  • Jonathon Colman - for every 1 second of load time, conversion drops 7%
  • (Extra from Mike T: 37signals also reported that a half-second improvement on page load for the Basecamp home page yielded a 5% conversion boost)
Be a Human
  • Users know stock photos when they see them. Even real “professional” photos with too much polish get ignored.
  • See Jakob Neilsen on use of photos in web content (including some interesting details on user engagement with stock photos)
Sweat the Small Stuff, Test the Big Stuff
  • Make users lives easier - if it’s obviously an improvement, don’t “test” just do it the first time
  • Don’t test “button colors” if there are big things that could drive bigger improvements

Why should SEOs do CRO?

  • ROI (and happy clients)
  • Career level up
  • (not provided)
  • Panda
  • Traffic that doesn’t convert doesn’t much matter
  • Customer behavior intuition
How:
  • Strive for validated learning
  • Be authentic
  • Simplicity
  • “You” not “we”
  • Be a human
  • Be inspired, don’t just copy
  • Sweat the small stuff, test the big stuff

Q and A

Q: We (audience member) do a lot of testing. We think we got conclusive results. Then we go back and test again, and the original result was wrong. Why? A: Will Critchlow is working on this now - you can be tricked by significance. Don’t jump the gun. If it looks like an obvious win, let it ride. Run another test if you’re unsure. It can be a tricky thing, even when your CRO software is telling you you’ve got statistical significance.

Don’t know about you, but I’m going to up my game on CRO after this excellent talk from Mike. We’re breaking for afternoon tea (how delightfully British), then we’re back with Annie Cushing, who’ll be dropping some serious knowledge on a framework for SEO audits.

Annie Cushing – “Establishing a Framework for SEO Audits” back to top

Annie has put together some great Google Doc resources for her presentation. Available, along with her presentation slides, right here.

Too much data isn’t necessarily a good thing. Annie loves data, but every time she gets an audit project, she gets the sinking feeling of not knowing where to start.

She had a client who came to SEER for help, they took a big hit in Feb 2012. Wasn’t links (profile was clean). Wasn’t technical. A ton to look through. Without the checklist (she will share) she wouldn’t have figured it out.

The most expensive mistakes we see regularly (in audits):

1. Subdomain reporting errors in GA

Annie recommends adding the “show hostname” filter for GA (so it’ll show full URL, not just URI). Instructions are in Annie’s audit doc.

Add an annotation when you do this. 1st day of the month preferred. Otherwise sorting out what happened will be confusing.

2. Index bloat

Way more pages being crawled/indexed than are helping your business (using up crawl budget)

Causes of bloat:

  • Borked internal linking
  • Search results pages (internal)
  • eCommerce parameters (sort variables)
Check indexed pages. Then look in GA landing pages report for organic traffic. The difference between the two indicates how much index bloat you’re dealing with. (Be wary of 0-visit pages showing up, a filter will help remove those).

Let’s do some math 5.8 million pages indexed Only 37k receive organic traffic Only 645 drove revenue .6% of indexed pages got traffic .01% of indexed pages generated revenue

This is a huge chasm, not the 1:1 ratio you want.

3. PPC/Display traffic showing up as organic

Take a careful look at organic landing pages - if PPC tracking tags are showing up in URLs, there’s a problem. At a minimum, filter this from GA data. But addressing the root problem is better long-term.

4. Dirty sitemaps

Duane Forrester said that after 1% of sitemap pages returning other than 200 status, Bing starts to lose trust. That’s a bit girly of Bing, says Annie, but it’s something to look at.

5. Sitemap hasn’t been submitted

This is basic. If you aren’t using a sitemap.xml file to help Google crawl the right pages on your site, you’re missing SEO 101 stuff.

6. Loss of valuable links

  • Home page that 302s essentially kills your inbound links
  • Pages with links that 404
  • Pages that 500

Annie’s Google Doc

  • Multiple tabs (for different areas/categories of factors) with a list of checkpoints in each.
  • Many checkpoints include instructions (often with links out to more in-depth walkthroughs)
Also has a Tools G Doc with 300 tasks, categorized, with useful tools for each task. Here’s the link again to her post on the SEER blog with all G Docs mentioned herein.

Thanks Annie! Next up we have CRO expert Stephen Pavlovich, who’s breaking from his normal topic of CRO and talking about his startup, Wish.co.uk, and how he and his partner leverage some interesting, agile online PR techniques to drive massive traffic and sales.

Stephen Pavlovich – “How to get Attention, Links and Sales for a Scrappy Start-Up” back to top

Stephen normally speaks about CRO. Not today.

Stephen is talking about Wish.co.uk. He wants to walk through taking it from the idea inception to a $1 million business.

They’re an experience company. They invented the Zombie Shopping Mall, which drove a ton of PR.

In the next 45 minutes Stephen will cover:

  • How a scrappy start-up can get attention (without paying for it)
  • How to multiply the impact of PR ( and keep costs down)
  • How to get interviewed about threesomes on Columbia national radio (ha)

Intro: Why we set up Wish.co.uk

One of the first rules of CRO is to “become a customer” - engage the business as a customer would and learn from the experience.

They wanted to apply that same approach to their own company (Conversion Factory). Wish.co.uk was that experiment.

Punch above your weight

Two people running the business, but they wanted to look bigger. The domain name was the best opportunity for that. They spent $10k on Wish.co.uk, and Stephen says it was by far the best investment they made. It has influenced every single marketing/PR effort they’ve done.

“Even though the company never existed, people thought they’d heard of it before because of the name.”

Ways to do this with PR/stories

  • Topical - create stories based around news in the moment
  • Sexual
  • Controversial
  • Celebrity

Creating Valentine’s Day PR on the cheap

They created Romantic Break for Three - a product that drove PR in itself. It drove coverage/links/etc. Got on the radio (BBC/Columbia) and some US tv.

But they had to pull the plug. The “owner of the hotel” wasn’t the owner of the hotel. He was the (soon to be ex) husband of the owner. Stephen: “it didn’t work out well.”

It was free, but it didn’t generate any sales.

“Eighty percent of success is showing up.” Woody Allen, favorite quote at Wish.

So they tried something else: can we get a tweet from @stephenfry (the British actor/comedian)

They created an (expensive) experience (it wasn’t real) offering brunch with a British celebrity, and Stephen Fry tweeted it. They got 10k visits, but no sales.

Start with the end in mind

“What would you do if you weren’t afraid?”

Wish used this mantra (from Facebook) to arrive at their next strategy. With their Zombie Shopping Mall product (which they’d just created), they thought @simonpegg would be a perfect target for tweeting the product.

@simonpegg tweeted. WIsh.co.uk’s servers went down. Several times.

They ran with it. They did a Reddit “Ask Me Anything” around the product (people were curious about health insurance, etc).

They leveraged Facebook sponsored posts.

The traffic went exponentially up and to the right. Stephen says the graph is embarrassing because it makes them look like they forgot to installed Google Analytics properly - when actually, they were getting steady traffic beforehand.

They then leveraged this initial seed coverage to drive more PR.

Revenue followed the traffic graph (way up, to the right).

Waves of coverage have occurredon their own since then. @jamieoliver participated and tweeted about it. The product’s viral quality drove its own PR after the initial pitch.

They started doing zombie videos with a YouTube producer (the guy who created the World’s Largest Rope Swing video).

They found the journalists who covered this guy’s previous videos, then reached out about the new zombie video. They were able to generate more considerable coverage, traffic and sales this way.

Checklists

Prepare:
  • Domain name (optional)
  • Server (must have for big traffic spikes)
  • Social mechanisms (for sharing)
Create:
  • Start with the goal, work backwards
  • 1+ of topical, sex, controversy, celebrity
  • Create a new product, create a fake product, spin an existing product
  • Gather quotes, media and “proof”
Launch:
  • Who’s the ‘ideal” personal to seed it?
  • As soon as you get seed credibility, repitch to other celebrities
  • Follow-up with a press release (include images and quotes; write the article for them)
  • Push on social media (advertise) and scale with services like BuzzStream

Q and A

Q: Did you consider providing any incentives to celebrities to encourage their promotion? A: We tried it, but it didn’t seem to work well. It seemed they were frequently pitched that way and were resistant to it.

Q: Did you do much CRO on the site throughout this process? A: Not much during those events - the type of traffic you get during those periods is tough to evaluate. Not likely to convert. We prefer to wait until after those periods to do testing.

Big thanks to Stephen for a great presentation (one of my favorites of the conference). Next up: Eytan Seidman, founder of Oyser.com, on content strategies that work.

Eytan Seidman – “Content Strategies that Work” back to top

Eytan founded Oyster.com after a travel experience in Alaska with two different accommodations - one good, the other...not so good.

The travel space is dominated by big players - Travelocity, Orbitz, Hotels.com, Expedia, Priceline.com, etc. Tough nut to crack. ~$50 billion of market capitalization. Just ultra-competitive.

Oyster goes to hotels in person, take extensive photos, document everything and post the review. Photos are sectioned into albums for user experience.

They categorize hotels around popular themes - “Best Party Hotels” “Best Luxury Hotels” etc.

Highly-visual product, rich content, adding hotels as quickly as they can.

The Challenges:

Massive Incumbent Marketing Spend (in the millions per year for most big competitors).

Big players have massive marketing budgets - Oyster can’t compete on budget.

Travel is highly episodic

Most people only take 1-2 “big” trips a year, and users aren’t necessarily looking for new solutions (they go to the brands they know).

So how did Oyster break into the market?

They started doing Content Marketing (before it was called “Content Marketing”). They learned fast that the content has got to be really, really (really) good. Honestly ask the question: Will someone take time out of their day to share this

Not only does this benefit via SEO signals, but it creates a positive brand experience (exactly what Oyster needed to start to be in customer’s minds when they were booking travel).

They created content showing the reality of promotional hotel photographs - “Photo Fakeouts.” They show what the experience actually looks like (not what the hotel wants you to think it’s like).

They grab users when they’re not thinking about traveling with interesting content pieces. This keeps their brand in mind when those people do end up looking for travel solutions.

Zillow has a similar strategy. People don’t think much about real estate until the 1-2 times in their lives when they need to buy or sell a home. Zillow finds angles around the real estate topic (celebrity homes, trends, etc), to entertain users and create a positive brand experience.

Forbes created the Forbes 400 to achieve much of the same. An otherwise dry/specialist topic (finance) is transformed into something with broad appeal (everyone’s intrigued to see who the richest people are).

One of the reasons Oyster’s “Photo Fakeouts” work so well is that it’s not easy for anyone else to copy. Oyster’s business model results in their owning many hotel photos. That’s a unique asset they can leverage that their competitors would have to invest heavily to copy. It’s defensible.

Your content marketing should reinforce what is unique about you

Examples:
  • Oyster was picked up by Lifehacker Australia for their Bundle Breakdowns (showing the true costs of bundles and whether or not they’re good deals)
  • 10 Charts About Sex (OKCupid.com)
  • HowAboutWe.com “The Date Report” - “8 Reasons It’s Amazing Anyone Gets Together + 14 Sex Tips from a 1960s Japanese Sex Guide”- what’s unique here to HowAboutWe.com? (Eytan is suggesting they could have done a better job, but he still loves the team.)

Focus your content marketing on vertical niches, then go broad

Don’t pitch USA Today - get some smaller successes first.

Oyster produced content (image galleries) for Shape Magazine (online) and others (such as the Huffington Post). Oyster always gets the attribution, mainly via a watermark on each image.

Similar to how Google provided AOL with search results (in exchange for a “enhanced by Google” credit). It put Google on the map.

Example of this strategy done wrong: Microsoft paid a ton of money to power Yahoo! search. They didn’t negotiate for a nice, big “powered by Bing” mention (their current mention is tiny, nobody sees it). They missed the opportunity. This is a big mistake - if you’re trading a business asset for exposure, maximize that exposure.

Oyster’s content pieces have been everywhere (Eytan is sharing a slide now with many recognizable big publication logos). Oyster gets content on these sites daily.

Oyster traffic by quarter (slide chart) shows fast growth over the last few years.

Q and A

Q from Rand Fishkin: What percentage of traffic is social vs content pieces vs organic? A: It’s peaky so we don’t get too granular. We haven’t done great with social (not a meaningful driver of our traffic).

Q from Rand: How do you find the people who produce content for you? A: The core hotel content on Oyster are contract employees, trained by Oyster. We find people in the areas themselves (rather than fly people around). They’re part-time contracted workers. The content marketing effort, though, is driven 100% by full-time in-house teams.

Q from Rand: Do you do influencer marketing (i.e. targeting notable people with audiences for coverage)? A: Not a lot. We generally focus on big audiences, HuffPo and similar.

Q: Have you guys seen any turning points (month/year) that you noticed a shift in Google organic traffic? Has your authority as a photo provider in the travel space pushed down local results (i.e. have photo results taken up SERP real estate that has hurt your traffic) A: We’ve seen big shifts in organic traffic, but overall we’ve been happy with recent Google updates. On the 2nd question, we’ve taken some hits in local Google traffic lately. We’re going to talk to Google about fixing that.

Q: How did you reach out to HuffPo and develop that relationship? A: One of the key things is starting small - don’t go right at the big guys. In some cases those relationships are outbound (targeting ~50 prospective places we’d like to be featured). As we get placements we find it easier to make inroads with new publishers (it’s a “rich get richer” effect).

Q: How do you track traffic sources? What metrics for success are you using to evaluate your content marketing efforts (KPIs)? A: It’s somewhat straightforward, we just look at referral data in analytics. As for KPIs, we like to look at returning/recurring visitors, when evaluating sources of traffic. We’d rather get fewer initial visitors with a higher rate of return.

Q: What tools do you use to attack the copyright problem? A: Queries on Google images generally. We’ve caught people cutting off our watermarks. But this is something we probably need to do a better job of. We’re not so worried about the little players scraping/copying our content, because the only people outranking us for key terms now are the big players.

Q: What kind of tools do you use for keyword research to help determine targets? A: We focus on hotels that are perceived as “higher risk” for consumers. Jamaica for example. This helps us determine hotels that consumers are interested to research before they book.

Thanks to Eytan for a nice presentation and some great tips for leveraging content not just for links but for building brand awareness. Up next is the Let’s Get Real session, where all speakers are invited to share 1-2 juicy tips the audience can put in action right away.

BUT you’d have to be here to get the tips (them’s the berries folks). Join us next year, don’t miss out on what Rand just called “the best marketing content I’ve seen at an event, period, including Mozcon.”

P.s. The official photographer for the conference was David Zwickerhill. You can view more photos from the conference here.

Mike Tekula

Mike Tekula

Mike joined the Distilled NYC team after managing a small business consultancy and heading up digital strategy at a New York ad agency. He specializes in organic search but has worked on a wide range of web projects across channels. Outside of work...   read more

Get blog posts via email

4 Comments

  1. Holy moly and I thought the ~8,000 word blog post I put together covering the event was huge!

    reply >
    • Mike Tekula

      Yeah I knew this was a bit of a monster when I hit 'publish' - hopefully it's readable though!

  2. The link to Annie Cushing's slides is wrong, the right is http://www.slideshare.net/anniecushing/establishing-an-audit-framework.

    reply >
    • Mike Tekula

      Thanks George, I meant to link to her post over at the SEER blog where she'd compiled her pres slides as well as the additional G Docs resources. Updated the link, thanks for catching and alerting me to that.

Leave a Reply

Your email address will not be published. Required fields are marked *

*
*

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>