Internet Marketing Company Blog - SEO & Pay Per Click Management Services

Search

« Blog Home
Categories
- AdWords
- Analytics
- ASK
- Blogs
- Branding
- Design
- Entrepreneurship
- Google
- Google Relevancy Ranking
- Info Visualization
- Internet Yellow Pages
- Job Hunting
- Keywords
- Local
- Manufacturing
- PPC
- Pure Visibility
- Quality Score
- Search Engine Marketing
- SEO
- Social Media
- Teams
- Twitter
- Uncategorized
- Usability
- Web Standards
- Web Video
- WordPress
- Yahoo
Archives
- July 2008
- June 2008
- May 2008
- April 2008
- March 2008
- February 2008
- January 2008
- December 2007
- November 2007
- October 2007
- June 2007
- May 2007
- April 2007
- March 2007
- February 2007
- January 2007
- October 2006
- August 2006
- July 2006
- June 2006
- March 2006
Search Engine Blogs

High Bounce Rate May be Good

Posted by Mike Beasley at 9:45 am | Filed In Analytics, Usability

Jul

8

You may find that a high bounce rate on your website is no problem at all. Bounce rate measures how many people come to your website and then immediately leave. Jakob Nielsen, noted usability expert, has written about using bounce rate as a metric for measuring usability improvements in his June 30th, 2008 Alertbox column.

He first writes that unique visitors should not be used as a metric for measuring usability improvements on a website because it doesn’t matter how many people look at your content. Rather, you ought to be concerned with how many people look at your content and decide that it isn’t worth their time. Neilsen would have us believe that we must strive for a low bounce rate. His argument is actually a bit more nuanced and has some good points, but that is the gist of it.

However, bounce rate is not necessarily a sign of an unusable website. It may indicate that your site is working really well. Say, for example, you sell a very expensive service through your website. It could be the case that website is doing a good job of qualifying your sales leads - people are getting a good idea whether they even want to bother calling you. This is a good thing, because you don’t want just anybody calling up your sales team. You want people with the best chance of buying.

A far better way to measure the success of your usability efforts is to measure how many people are doing the thing or things that you want them to do on your website. If you’ve got a form on your website where people request a quote, you can measure how changes to your website affect how many people fill out that form. If you’ve got a phone number on your site, you can measure the change in call volume. If you’ve got a site that sells widgets directly to consumers, you can measure the change in sales.

Measuring bounce rate makes sense if your site just exists to provide information and nothing else. Otherwise, you measure success on how many people do what you want them to do. Of course, the point of usability was to try to make it easier for your customers to do the thing you want them to do.

What are your favorite web applications?

Posted by cjuon at 8:39 am | Filed In Entrepreneurship, Twitter

Jul

1

I am frequently asked by early stage entrepreneurs what software we use, so here’s a quick list of the five online applications I can’t live without.

Salesforce.com. I started using salesforce back in the demo days when you could get a 3 user license that did everything for $50/month. Ah, those were the days… These days, it’s amazing we don’t scare off new employees with the mantra “it’s in salesforce” as the answer to seemingly every question. Seriously though, salesforce is where I spend my day. Every person who fills out a contact form on this website winds up in there automagically, and every person I’m supposed to call or have ever called is in there. (It’s a founders job to sell, after all.) Essentially, Salesforce.com functions as a much needed extension of my brain as our sales pipeline grows, and it’s pretty darn handy for sharing customer contact info within the company.

Basecamp. Yes, salesforce has project management stuff you can tack on, but lets just say it wasn’t a huge hit when we tried it. Not that basecamp is absolutely perfect either, but you know they’re doing something right when you turn it on and people start adding data without any arm twisting. Basecamp also shares something in common with one of my favorite things about salesforce: the “wish for it and it will come” feature. I suppose it helps that we aren’t shy about sharing what features we need ;-), but we’ve happily seen things appear over time such as cross-project search and posting via email that make our lives SO much simpler.

Socialtext. I wasn’t exactly sure we needed a wiki as yet-another-place-to-store-data. But once again, it’s hard to argue with a technology that is easily adopted by the organization and has taken on a life of its own. In a way, it’s like a giant shared filing cabinet for everything about anything internal. (Whereas basecamp shares project-specific data, and salesforce captures the sales process. So it’s not as complicated to have 3 systems as it might sound.)

blist. I’m adding this to the list a bit early, as it’s still got a ways to go to get through all the bugs. However, I would will it out of beta, if I could. It’s not uncommon to need to compare a list of stuff and figure out which entries match criteria a, b, and c. Blist does this with “lenses” over an online spreadsheet with simple and/or choices; no programming lingo (ie SQL) required. I honestly haven’t seen anything similar, but let me know if you have!

twitter. I have been known to say we’re in the business of creating serendipity, and I think that’s what I love most about Twitter. This micro-blogging tool encourages you to share what you’re experiencing or thinking in 140 character bursts. And because people can follow either your persona OR create feeds on any word of interest, you wind up serendipitously meeting people with similar interests. People who often become instant friends, pitching in and answering questions or cheerleading when you have a breakthrough moment at 2 am. Twitter is a truly social network that I’m grateful for as an entrepreneur.
So that’s my top five web applications - what are yours?

The Structure of the Web

Posted by jhullman at 15:48 pm | Filed In Uncategorized

Jun

27

webgraph

The web remains the largest text corpora to date, although in recent years corporate archives are beginning to catch up. But the web is more than just a massive text database. The above picture is one I enjoy geeking out over. It shows the structure of the web. Here’s a description of its parts:

SCC: This is the Strongest Connected Component, a group of pages connected by links where, from any page, you can get to any other page via a series of links and then back to the original page via another series of links. This is believed to be the largest component, though this can not be proven at present.

In Component: Those pages from which you can get to the SCC, but cannot get back to from the SCC. Think someone who links their personal page to all their friends pages, but gets no links in return.

Out Component: Those pages that you can get to from the SCC, but once there, cannot get back to the SCC.

Tube: A page in the In Component that links directly to a page in the Out component without going through the SCC.

Tendril: This is a difficult thing to explain. My best attempt is as a subset of pages that are all strongly connected, that may be linked to from either the In, Out, or SCC, but once you are there, there are no links out of the Tendril (like a company internal page or set of pages).

Mike Beasley vs. Michael Beasley

Posted by jyoung at 15:01 pm | Filed In Pure Visibility, Teams

Jun

26

Overall, we are very pleased to have acquired Mike Beasley. His experience and training in usability is indispensable; his input can really improve the user experience on websites. But one wonders what it would be like to have Michael Beasley around the office. I decided to do a quick comparison.

Name	Mike Beasley	Michael Beasley
Visual Impact
Height	5′10″	6′8″
Education	B.A. & M.S.I. University of Michigan	1 yr. Kansas State
Organizations	Pure Visibility, Compuware, Thomson Gale, MiUPA, ACM SIGCHI	Kansas State, Chicago Bulls? Miami Heat??!
Income	As of April 2008, Mike was earning more than Michael	It is generally believed that whichever franchise picks Michael in the 2008 NBA Draft will pay him more than Pure Visibility pays Mike
Strengths	User experience, public speaking	Ambidextrous, scores like 26 pts/game
Stuff in common	Likes sneakers, likes hip hop music, has been to Manhattan Kansas	Likes sneakers, likes hip hop music, has been to Manhattan Kansas

Yahoo!’s Outsourcing to Google

Posted by steve loszewski at 8:21 am | Filed In Yahoo

Jun

23

Yahoo! recently announced that they are going to display Google AdWords ads for some of their queries. If you read a lot about internet marketing, you’ll see that the popular angle towards explaining the extra revenue Yahoo! will generate from such a move is AdWords has greater relevance in ad distribution. I don’t think that’s the primary reason why Yahoo! will make more, and if it was, it’s bad news for Yahoo!. I’d count that as a vote of no confidence for their new Panama system, and that fact could be more damaging to their reputation than an extra $250 million in revenue is worth.

It’s true that Google provides more match types, better negative matching, more location targeting, and apparently less spammy search partners. When I manage ads on AdWords and Yahoo! Search Marketing for the same company, I typically expect better conversion rates on Google than on Yahoo!. But I also allow for a higher cost per conversion on AdWords. Getting more traffic, more exposure on first page results, itself has value to the companies I manage. These companies are less concerned with YSM. This pushes more budget into Google AdWords, regardless of the cost to generate direct business. The reasons I think why Yahoo! will generate more revenue by using AdWords’ distribution are:

Google Analytics Bummers

Posted by jhullman at 12:17 pm | Filed In Analytics

Jun

20

I’ll be the first to admit that Google Analytics is, like many Google tools, well-designed, free, and puts a great deal of functionality in the hands of users who might otherwise have no access to such data. But while some great posts have been written about Google Analytic’s strengths–(check out this comprehensive list of high points)–it’s much rarer to find web analysts complaining about some of its weaknesses. The few posts I have read on this subject often cite two in particular: the problems with data accuracy, and the lack of real time data.

These weaknesses are actually less of a bother to me, in comparison to a couple other “Analytics bummers” that have always bugged me a bit. So I’m finally calling them out:

1. The Danger of Averages
With the exception of a few reports, most of the numbers Analytics provides are averages over the Date range. This poses a risk whenever a user is trying to get an accurate picture of what a user’s experience on the site is like. Some sites may have a minority of dedicated visitors (a blog, for example, think the bloggers friends and family) as well as a majority of very dissatisfied users with high abandonment rates. Google Analytics in its current state obscures situations like this by pushing averages and not providing the same easy access to indicators like median and mode. Often in analysis its the high-end and low-end performers that should be zeroed in on for real results.

Also, some reports can’t be segmented (Visitor Loyalty, Depth of Visit, etc) so you’re given only averages over all visitor segments. What if the average is lumping one segment, with a very high depth of visit for example, with another segment with a very low depth? The picture you’re given via the number hides this completely.

2. No Emphasis on Significance
My next gripe has to do with what seems to me a complete lack of effort on Google’s part, in designing the tool, to add functionality that addresses the importance of calculating statistical significance before getting excited about a trend. Think of any market with seasonal fluctuations. If my pageviews and length of visits, etc are way up in the last month for a certain online channel, should I necessarily assume that I did something right? A split test is an easy way to find out if a trend is really worth watching, but I wouldn’t count on the fact that most users of Analytics are bothering to do it before they start acting in response. How hard would it have been to make a little applet for users to enter their numbers into, along with an alpha rate, in order to find out the confidence level?

3. My Data = Your Data?
My final concern has to do with the fact that Google stands to profit quite a bit from having access to extensive data on a large portion of the indexed web, especially given that they do profit considerably by pitting website owners against one another via the paid search auction. Between their index and the pile of data Analytic’s users are handing over, they have every opportunity to continue monopolizing the web and steering it in whichever direction they please. Do you trust Google? I do. But I can’t help but be aware that in the end, Analytics equals a data for data exchange, within a very unbalanced relationship.

Making Data Meaningful: Our Favorite AdWords and Analytics Reports!

Posted by jyoung at 18:02 pm | Filed In Uncategorized

Jun

19

Admittedly, the Pure Visibility team often geeks out about tech tools and trends. But one of the really satisfying things that we do is to emerge from the depths of geekdom and share information with clients that shows what’s happening with their online presence, stuff that can help inform decisions about web marketing efforts. Inspired by Google Analytics Evangelist Avinash Kaushik’s “best web analytics report”, we’d like to share a couple of reports that we like, ones that can make your Google Analytics and AdWords talk to you.

Jessica Hullman:
Google Analytics Site Overlay Report
Jessica picked the site analytics report because it’s beautifully simple. It’s an intuitive visual representation of where visitors navigate on a site. See how many times visitors have clicked each link on a page for a period of time you specify. Look at conversion data for the goals you’ve set for your site.

Google Analytics - Content - Site Overlay

Steve Loszewski:
Google AdWords Search Query Report
Steve went with a report that helps him save advertisers money. Steve looks through the list of Search Queries that cause his ads to display. Then he picks out terms he definitely doesn’t want to show up for (you know the sort I mean). He adds these keywords to his list of negative keywords and his account is on the way to more relevant placement and visits, higher quality scores and lower bids!

Google AdWords Search Query Report

Mark Williams Picks:
Google AdWords Keyword Report

Mark is the guy on the left, not the guy grimacing at taste of funky water samples. That's Daniel.

AKA the victory lap report. We <3 AdWords because it is measurable, accountable; we can quantify the performance of our online advertising campaigns. Mark uses this this report to show what’s working in his campaigns, which keywords are generating clicks and leading to conversions. Take a look at the settings in the screenshot below; Mark likes his reports to include only the columns that help explain the information he wants to highlight.

Google AdWords Keyword Report

Easy Social Media Metrics

Posted by doneil at 11:26 am | Filed In Google Relevancy Ranking, SEO, Social Media, Usability

Jun

17

One of the biggest obstacles to Social Media being perceived as a legitimate marketing activity is its measurability, especially for companies with an annual revenue of $1 million or less. Paid Search and SEO have fairly specific metrics of effectiveness that can be easily related to the bottom line. Social Media, on the other hand, is less focused on end-result outcomes and is therefore harder to measure.

Yet in a lot of ways Social Media is a very familiar kind of marketing. Social Media can be broken up into two broad categories:

its creation and dissemination, and
its consumption.

The majority of participants in Social Media are not involved in its creation or dissemination, but instead simply consume it, in much the same way as people consume news or broadcast media: through a trusted channel that provides new and novel content.

This insight suggests a simple way to measure social media: essentially in the same ways as you would measure traditional marketing efforts.

This strategy doesn’t capture many key benefits of Social Media, and also doesn’t take into account the emergent and interactive effects of combined marketing strategies. However, with this strategy you can start your company down a path where the simplest benefits of Social Media can be implemented and measured.

Charlene Li at the excellent Forrester Research group has explored this topic extensively, both in research articles and in a blog post outlining some of the key benefits of Social Media and its value. She argues, for example, that raw blog traffic can be compared to the effort of print media to get a similar number of visitors to the site. Li argues that companies can start to do analysis of Social Media immediately with this approach.

How would this work for your company? There are five steps:

Characterize your current marketing efforts. Make a list of all the marketing or client-focused activities your company is involved in and their costs, including resource costs. This list would typically include things like trade shows, focus groups, PR efforts, and print advertising.
Define your efforts’ value. For each effort, identify an item of value that is an outcome for that effort. This could be leads, page views, user feedback on a product, etc. Try to keep these values as close as possible to traditional marketing measures of success, such as “views”.
Develop or estimate a “cost per value” metric. This should be based on the costs and values identified in (1) and (2).
Contextualize your Social Media effort based on the values of your marketing. Categorize the expected outcomes of a Social Media effort in the context of the values defined in (3) so that you can compare the costs of the Social Media effort with existing efforts.
Test and Review. Run your Social Media campaign for a quarter and review the results. Use this time to identify both new “values” and to tweak your cost estimates.

Strong supporters of Social Media and marketing have objected to this model as being too simplistic and atomic. We accept those limitations because in order to be able to measure the emergent values of Social Media, we have to be able to measure the ways it matches up with older marketing efforts, particularly in environments where its essential value must be measured in order to justify further effort.

Social Media Analytics Software

Posted by jhullman at 10:54 am | Filed In Analytics, Social Media

Jun

16

As social media continues to gain popularity as an advertising channel, some companies are advertising software aimed at mapping and make analyzing online social communities. I’ve been considering some of the pros and cons of tools designed for online social networks.

Analysis packages like Radian6 and BuzzLogic consist of dashboards (not unlike Google Analytics) that present and perform further analysis of data that their crawler programs harvest from sources like blogs, image sites, and video sites. A big selling point is the ability to monitor both the response to your posts, and other activity relevant to a given topic, continually. The ultimate objective is to help companies zero in on what social network to join, who the major influencers are, and what they are talking about. At least one puts an emphasis on segregating key influencers on a given topic from key influencers overall. This is of course important, given the evolving nature of networks. The topics of interest, influencers, and participants are constantly changing, and analysis needs to account for this.

Pricing of these tools is often on a monthly basis. This is where the hard questions come up for an SEM company, like, do I have enough clients using social media to warrant the spend? Do I feel confident investing the time in learning a new tool, which may or may not give me output I agree is useful? Should I hack together my own tool instead?

The biggest consideration should be the proportion of your client contracts revolving around social media. The contracts would need to be fairly large-scale to make the investment worth it. If you don’t clients that fit this description, then its probably not the best choice (at least not yet!)

If you do, it comes down to your confidence in your own knowledge of social media networks. If you have a real need for analysis but don’t have ideas on how to measure things yourself, then a package is for you.

But based on the conversations between SEMs online, there are lots of us who have some ideas of our own about how to measure social media. So assuming your company falls in the group of those that would rather design their own tool, here’s a rough scope of the process of creating an analysis tool that can output enough metrics and analysis to fill, say, a monthly report on social media performance for big clients:

Phase 1: Obtaining data
Ideally, a single time investment in resources (time) to create a suite of crawlers that can withstand spam and abnormal sites. You’d want a crawler for all the big social networking sites, and you’d need to be able to work around the steps in place on some social networking sites to prevent crawlers. How hard is this? Depends on who you talk to. But don’t expect it to be easy.

Phase 2. Making sense of data.
Requirements: Depend on desired goal. To identify patterns in a network over time like key influencers (as the products on the market do), a couple algorithms should do the trick, as long as they’re used correctly (check out research in physics and information science departments on what they use to analyze evolving networks).
Keep in mind that there are tools that are designed for network analysis already, but without the data aggregation component, and without some of the social-media-targeted splash of tools mentioned above in terms of visualization. Pajek, Netlogo, and Guess are a few to check out.
Keep in mind, even tools like Radian6 and BuzzLogic require some manual analysis, such as in separating positive and negative feedback. From a text processing standpoint, this is a difficult problem, and assuming some manual analysis process every time is probably the safest bet.

Phase 3. Maintaining Analytics
Once a detailed enough output report was achieved from last steps, this should be easy. But what equals enough of course depends on the status quo for social media analytics. Will it be skyrocketing? I’m guessing more flashy tools are on the way, but as far as the number of companies doing really intricate analysis (versus lots of aggregation, and a bit of analysis), I don’t know that that will really come into popularity. Think about the measurement of other marketing channels. The popularity of Google Analytics, for example. What tends to be most popular are a few key metrics, with only the minority really crunching numbers via complicated algorithms and statistical models.

What Really Drives Social Media Buzz?

Posted by jhullman at 11:23 am | Filed In Analytics, Social Media

Jun

12

After a recent discussion with a non-marketer who studies citation networks, I’ve been asking musing on influence in social networks. What is it, exactly? Does it depend on the network? Is it as difficult to define as charisma? Or is it an (albeit inexact) something that an analyst can feasibly zero in on?

The question may be too big. It may simply depend on the network. But its worth asking, because it would be fantastic information for any company performing social media analysis to know.

Blog and other online social networks are not unlike academics’ citation networks. Research on these can inform this investigation. Citations, like comments on a blog, can not be removed or added after the fact to a post, meaning that despite continual evolution at the “leading edge” of a network, the structure is mostly static.

In traditional network theory, there are cut-and-dry definitions for metrics that begin to pinpoint the influence that SEMs talk about. What’s interesting is that research of specific networks often determines key influencers to be the same, regardless of metric used. In citation networks, it is commonly the fact that these key influencers are those that have been around a relatively long time.

I have a feeling (which I would certainly like to try and back up with data) that a similar situation probably exists in many blogging communities.

If you’re analyzing a blog community, and can find a correlation between time a participant joined and their influence, a question immediately arises as to why this is the case. Do people tend to link to or reference these key influencers because they are experts on a given topic, because they are tuned in to the driving forces/topics in a given field, or because they have been around the longest, and everyone else cites them?

It suddenly becomes a question of measuring influence in multiple directions at once - the key player on those reading, and the opinions of other participants on each other. Not one I can answer at the moment, but one that’s unlikely to go away.