PubCon Vegas 2010 – Day 3

by Elmer Boutin on November 11, 2010

PubCon Day 3 started off for me with a great conversation on the bus and then over breakfast with Mike Marshall (@michaelmarshall). He’s one of the speakers on advanced SEO techniques. I’m going to put him on my list of people to catch speaking next time around.

Here are summaries of the sessions I attended:

Keynote: Future of Search: Mobile, Social & Vertical – Tim Meyer (@timmeyer)

  • Suggestion, prediction and serendipitous search is going to take off more than the “traditional” way we have been searching.
  • PC Search is only growing in single digits, mobile search is growing in double digits. Look for local to take off very soon.
  • Five Key trend to look for
    • Mobile Queries will be 20% of search by 2012. Much of the growth will be informational, local, and product lookup.
    • There will be lots of growth in apps and not in browsers. Apps are taking more and more face time from browsers on mobile devices
    • Input is currently difficult. There will be better ways to input search data.
    • More Context: c
    • “Verticalizing” SERPs. Google and Bing will need to continue to improve their SERPs to keep up with niche search services.
    • App discovery is the key to customer satisfaction in the Smart Phone Era.  There are opportunities for app developers to create the “all in one” search tool which will be able to anticipate what the user really wants.
    • There’s going to be a lot more verticalization in search. Think local now, later think TV & video.
    • “If the past is search, the future is suggestion” – Roger Ehrenburg

Convergence of Online Marketing and Analytics – Moderated by Joe Laratro

Analytics: Then, Now & Tomorrow  -or- Bringing Sex Appeal/Analytics Makeover – Alan K’necht (@aknecht)

  • We can’t move forward in our industry unless we understand where we came from.
  • Birth of Audience Analytics or Audience Sampling. Analytics really started with newspaper ads, then radio ads with A.C. Neilsen.
  • The first banner ad was on HotWired.com in 1994. Easy to measure, but it blew up because analytics were first set up for server managers. Webmasters started counting unique IP Addresses and page views.
  • SEO and SEM helped get a marketing spin on numbers.
  • Web Analytics, Social Media numbers, Off-line, and Reputation numbers are all needed to be measured.
  • Where’s the sex appeal? When we can bring together the Why, Purpose/Intent, and all conversions – simplify it down to how money was made – that will be great.
  • Show management where the money is using funnel diagrams and intention flow in easy-to-understand ways.  Show people simple graphics so they can get it faster.
  • Use Mongoose Metrics voice analytics and combine with web analytics to get the larger picture.
  • Don’t just show the graphic data, offer action-based solutions.

5 Ways of Getting Actionable Marketing Insights from Analytics – Tom Critchlow (@tomcritchlow)

  • Don’t use stacked graphs. They’re hard to read. Pie charts are stacked graphs in pie shape.
  • Do segment traffic. Use separate smaller graphs to show individual data.
  • Don’t break down social media into individual pieces. People want to know how many people came from “Social Media” but not necessarily from Reddit, Digg, Flickr, etc.
  • Check out http://analytics.blogspot.com/2010/07/using-wrong-tracking-code-can-cost-you.html – about using wrong Google Analytics code on your site and why you should fix it.
  • We should be focusing on the money. That’s what people really care about.
  • Look into using Custom Variables in Google Analytics to track individual pieces of information. You can segment the data by different criteria (like content author or number of reviews on a page).
  • Check out First Touch Traffic in Google Analytics: http://dis.tl/1touch
  • 80/20 – target influencers. Don’t just spam them, call and use the personal touch. You’ll get a lot more value from it.
  • Data is marketing.
  • Numbers != Engagement. You have to find out the why behind the numbers. Heatmaps are great to help understand intent.

Convergence of Online Marketing and Analytics – Chris Zaharias (@SearchQuant)

  • Check out soldiersangels.org – hook up with soldiers who need support.
  • Online marketing is real time and fluid. Analytics must automate decisioning.
  • Analytics must automate decisioning to be SCALABLE.
  • People are more and more using search as a way to direct-navigate to brands they already know and want to go to. Google and Microsoft is making tons of money off searches of trademarked terms.
  • Leverage search campaigns. Example: someone in Chicago looks for flights to Phoenix on a certain date. You can post an ad on their page showing a well-converting deal on flights from O’Hare to Phoenix. Automated behavioral ad impressions. Goes beyond simple search-based banner ad placement.

Instant Analytics Vendors and Package Implementation – Moderated by Joanna Lord (@JoannaLord)

Extracting Real Value, and no BS – Richard Zwicky (@rzwicky)

  • Analytics packages are mostly wrappers. The raw data and the analyst are the real valuable stuff.
  • Business leaders want good valuable information to help them make good business decisions.
  • Marketing is no longer about just speaking. It’s about speaking when and where your customers are there. (Very true information today).
  • Conversions are all that matters. Everything else is just fluff.
  • Analytics is about listening to your customers. You need to be where your customer want to find you, when they want to find you. If you’re not listening, your losing your customers.
  • Use your analytics data to know where you can extract value. It helps you find your money tree.
  • Do a conversion analysis. Learn how marketing can optimize the conversion rate without driving in more visitors. Identify pain points in the funnel, find out why customers are leaving and fix the problem.
  • Segment your data out, especially by geography and time/day analysis. Don’t be afraid to turn PPC campaigns off during low-performing times of the day.

Google Analytics – Trevor Claiborne (@tclaiborne)

  • Web Analytics Measurement Model
    • Instead of starting with the data, start with the business objectives. In the end, income will come into play. (This is the way we learned to do analysis when I was in the Army. Please check out my post “Why Intelligence Analysts Make Good Marketers.”)
    • Use DUMB Goals: Doable, Understandable, Manageable, Beneficial.
    • After you know what the business objectives are, then you can start looking at Key Performance Indicators.
    • GA Intelligence function looks for anomalies you might want to check out.
    • GA also has email and SMS alerts on metrics you set. Handy to get an alert when things change.
    • Use the Intelligence portion to help you learn the “whys” of things changing.

Adobe Analytics – Driving Value – Brent Dykes (@pptninja)

  • Adobe Site Catalyst has hooks into Facebook and other social sites to get data from for analytics.
  • Analytics and the path to value:
    • Data
    • Reporting – this is more about set up and looking at the data in the next step. This is the friendly version of your data.
    • Analysis – You still need to do other reports.
    • Action
    • Value

I missed part of the next session because I was getting a demo of Raven Tools at their exhibitor booth. That was time well spent. We’re looking for a tool like that and I’m looking forward to using the free trial subscription.

I did catch the end of Dan Zarrella’s numbers piece about analytics and measurement. I can listen to Dan talk numbers all day. I’ve mentioned many times that you should be following him if you’re into Social Media Marketing. He mentioned a new tool, http://www.tweetpsych.com, which he wrote to help you find influencers in your social marketing.

Website and Content Protection: Scraper Prevention – Moderated by Brett Tabke (@btabke)

Taking Control of Your Site, Transforming From a Free-For-All ‘bot abuse to tightly controlled site access – Bill Atchison

  • http://www.crawlwall.com
  • Bill had a problem that scrapers overloaded his servers and brought them to their knees.
  • Good bots are desirable. Google and Bing give you value for their crawling.
  • Bots that check for plagiarism often mask themselves as Internet Explorer or spoof legitimate search engine bots and usually don’t honor robots.txt
  • Bad bots want to get something for nothing. The build web sites using your content, mine information using your content, get traffic using your content and make money using your content.
  • Webmasters have responsibility to protect their customers’ data.
  • Bots download enough data that they can badly skew your analytics numbers.
  • The bots will spoof their user agents, but sometimes not their IP addresses.
  • They take the legit content off your site and show it to Google, but actually show people junk ads and other bad content.
  • Robots.txt exclusions alone don’t work because bad bots don’t follow it; most don’t even download it.
  • Blacklist filters don’t work because you have to constantly update it.
  • White list filters or “Opt-In” methods work better. “Only these bots can come in and no others.” Let in browsers and “goo” bots only.
  • Use CAPTCHA when a single user accesses too many pages within a certain time to allow them to prove they’re really human and not a bot.
  • Most bots don’t download CSS. You can use this for a possible check.
  • Bill sets up spider traps to stop bad bots from getting past the first page or two on his site.
  • It’s your data, it’s your content – control it because you own it.
  • In the end, controlling bots keep your site running better, gets you higher in SERPs and keeps your content yours.
  • Consider contacting the scaper’s web host to see about getting them shut down.

Rogue SEO Protection – Brett Tabke (@btabke)

  • Self-referring quite a bit in text will keep copying and scraping down.
  • The higher you rank on certain key words, the more likely you are to get scraped. The scrapers just want to make money, so they want to rank for anything that will get them eyeballs.
  • PPC Sites – Pills, Porn and Casinos.
  • Don’t allow user-generated content on a PPC landing page.
  • Google Quality Rater Guidelines: http://blog.searchenginewatch.com/080314-101545
  • Monitor your trademarks.

So, another PubCon comes to a close. It was a great time, I met some great people and learned quite a bit. I have a lot of actionable things to do when I get back to work.

What about you? If you were at PubCon, what did you like about it?

  • Twitter
  • Facebook
  • Digg
  • Reddit
  • StumbleUpon
  • Google
  • LinkedIn
  • Technorati
  • RSS Feed

Previous post:

Next post: