Posts Tagged ‘Search Analytics’

More on Social Reporting in Enquisite

February 10th, 2010 by Richard Zwicky

Yesterday, I posted about Enquisite adding a Social Tracking function to our reporting suite. I got quite a bit of email expressing interest in knowing more. In re-reading my post, I also felt a little egg on my face for making the beginner mistake of leaving a [LINK] comment in the post, and not making it live – sorry. Stupid mistakes are easy to make; when you mess up, clean it up!

Anyhow, most of the email & DM’s via twitter (@rzwicky) asked for more info, and screenshots. Beginner mistake #2 – always include screenshots whenever possible so readers know what you are describing. I do realize that none of you can look inside my head and see what I do, so why the heck would I assume that you could visualize what I’m writing about without a picture? When I used to do SEO full-time, I always explained to clients that a picture’s worth a thousand words to human visitors, but zero to the search engines. Why would I ignore such a basic tenet of providing meaningful information? Text is great, but a picture seals the deal. If a picture is worth a thousand words, this is one lengthy document!

So, this post is about correcting yesterday’s errors, and providing more insights into Enquisite Social Reports.

We created Enquisite Optimizer’s Social reports to help provide online marketers with perspective. Last year I spent a fair bit of time talking to all sorts of online marketers: search marketers, social search marketers, social marketers, video optimization specialists, etc., and also VP’s of marketing and CMO’s.

What struck me about social was everyone talked about social, but no one seemed to understand how it related to any of their other marketing activities, nor what conversions it really drove. We already were collecting all the user referral data, but didn’t display the social information. Looking at the information, I realized that while there are some apps which help you understand how many referrals you get from any one channel, there’s no holistic views. For example, when you do a post using bit.ly’s service, you can track how many click through you’re getting from the U.S., but you do have not context as to real geography. We already were reporting on search referrals down to the zip code, so why not do the same with Social? Why not make it possible to track a social push, show where the traffic spiked geographically, and then look at the search referral traffic to see how it compared? Did it follow a similar pattern? Did conversion rates go up? How does one benefit the other? This information was missing from the discussion.

Social marketing offers a huge potential of opportunities in branding, driving visitors, raising awareness, and delivering valued customers to a business. But it’s not a stand-alone channel; it’s a piece of the puzzle. We’ve added social reporting to offer businesses some perspective, so that you can start to understand how they really fit together. For example, take this series of screenshots from Enquisite Social Reports, and then compare it against our Longtail search analytics report.

First off, we have a Longtail type view into social referrals, using categories as a definition. Instead of just reporting on all the referring sites, we added a category layer, so that you can understand the traffic types at a higher layer. In this screenshot you can see that for the particular website being looked at, shopping and consumer review types of social networks deliver the best conversion rates relative to overall social traffic.

In the upper left corner of the screenshot, you should also be able to see the site-wide bounce rate, page view rate and average time on site for referrals from social marketing. Watch how this number changes, and compares against search referral traffic.

Next up, we segmented out just the shopping and consumer review types of social sites – note the segmentation panel. I’ve also dropped in Twitter, as I wanted to see how it related. Note how the traffic quality improves as shown by the increased time on site / pages viewed, and lowered bounce rate.

Now, in this next step, we’ve segmented out social referrals to just ones that came from within the U.S., and are showing this information on the map. Look at the distribution pattern of visitors from social search, and keep this in mind for the comparison to search referrals to come later.

Drilling in to the map view, we have two important perspectives: 1) where do the referrals come from, city by city, and 2) then the following screenshot shows us where visitors were really located when they purchased a product as a result of a referral from a social network or social marketing initiative.

Now let’s compare this against search referrals in Enquisite Optimizer’s Longtail reports. First off, the bounce rate is much lower from search, and the pages viewed and time on site are much higher. So a more engaged visitor from search.

At the present time, search is sending almost 100x the referral traffic that social is to this site. However, to be fair, the business in question hasn’t really engaged a full-on social campaign. More like dipping their toes so far. But, all of a sudden they are recognizing value where they couldn’t before, in that they can understand the conversion rates better, and also they can compare and understand how the two traffic sites overlap.

Finally, let’s look at the map of search referral traffic. First up are referral rates.

Obviously, unlike the social referrals, the search traffic to this business is very broadly dispersed. Looking at conversion rates however, a different trend emerges:

Interesting how the Pacific Northwest is over-represented for conversions, relative to search referrals.

Finally, let’s compare that against conversions that were generated from social marketing, and we can see similar patterns emerge, with certain locations better represented proportionally.

These screenshots were built using the same time range throughout. When tracking specific campaigns, you can get much more granular to understand time lag.

Additionally, for marketers and business operators who want to understand the financial contribution of any channel to the bottom line, we offer Enquisite Campaign, which was designed from the ground up to report on, and provide predictive analysis of opportunities across all online marketing channels, and let online marketers, VP’s, CMO’s and CFO’s understand how the various channels interact, and combine together to contribute to revenue.

Marketing via Social networks is still in the early days, and the impact is usually difficult to understand. But a combined perspective on Search and Social will continue to become ever more important to any online marketer. We recognize that marketers are having a hard time measuring the impact of both channels, independently and together, so we’re bringing some perspective to the marketplace.

Businesses need perspective to properly invest in worthwhile initiatives. We provide insights to act.


Google Chrome, Safari and Firefox Continue to Grow; Don't Ignore Mobile

October 6th, 2009 by Richard Zwicky

Around this time last year, I looked at browser market share in the context of Google’s Chrome browser release. The market has definitely changed quite a bit since last year. Most surprisingly for me is how Chrome rebounded from a soft launch. Firefox, Chrome and Safari now account for nearly 31% of search queries worldwide, up 50% over September, 2008. Interestingly, Firefox and Safari have both grown by almost 45% in their combined market share vs. November, 2008.

At first glance this particular data set doesn’t appear to help much with SEO. But usability is a major concern of any SEO, and browser compatibility is key to usability. So SEOs need to understand these trends and plan accordingly to render properly within the leading browsers.

Separately, my friend Cindy Krum is always asking me about mobile browser data. Total mobile browser market share now surpasses 1% of all search engine referral browser views. Not surprisingly, the iPhone leads the pack at is at almost 0.6% marketshare, which is significant, and double the share vs. April of this year. RIM, Opera and Android and a few others make up the remainder. One year ago, Android was a non-issue. Now, it’s 0.04%. Still tiny, but noticeable. SEOs need to pay attention to this emerging trend. The real estate in the organic listings on the mobile browsers is much smaller, and the likelihood of people on mobile browsers going to page 2 in the results is also lower, so being at the top of the page one listings in mobile really is all that matters.

July 2009 August 2009 September 2009
MSIE (all) 67.59% 66.75% 66.61%
Firefox 19.09% 21.99% 21.13%
Safari 5.12% 6.86% 7.08%
Chrome 1.83% 2.27% 2.38%
iPhone 0.45% 0.59% 0.58%
Android 0.04% 0.04% 0.04%

About the data. Enquisite works with thousands of sites worldwide and captures a trove of relevant search-related data every day. The browser shares reported here are based on data from a selection of Enquisite-tagged sites that cumulatively represent over 350 million page views/month, across most major industry sectors - a very significant sample size.

Lastly, yesterday marked the kick-off for SMX East. Sadly, I’m not going to be there this year, as I had to bow out at the last minute for personal reasons. I know a few people, Jessica Bowman among them, had commented that were looking forward to some data, so in that vein, I hope this provides some insight into what’s going on in the search world. More data on blog posts to come.

postscript - for those who don’t like having 2 windows open… here’s the numbers from last year…

Date Chrome Firefox Safari MS IE (All)
09/30/08 0.501% 15.007% 4.321% 79.832%
10/15/08 0.433% 15.387% 4.178% 79.592%
10/20/08 0.462% 15.643% 4.296% 79.183%

New Enquisite Feature - Opportunity Analysis Report

January 12th, 2009 by Richard Zwicky

Ever wonder if you’re missing out on fantastic opportunities that are lurking within your own web site? Now you can find out. The page 2 optimization strategy I’ve written about in the past is a great way to discover potentially lucrative opportunities, but that strategy focuses on identifying existing opportunities in the search rankings-your pages are out there and recognized by the search engines-you just haven’t made it onto page 1 yet.

The strategy I’m going to talk about today is different. The page two strategy works by identifying the low-hanging fruit that, with a little optimization work, can move new pages onto page 1, an action that typically results in a 4500% increase in traffic. The Opportunity Analysis Report delivers you another way to improve your search referral traffic, and conversions!

Enquisite’s Opportunity Analysis Report is found within the Search Engine Comparison report section. It helps you identify which phrase are driving referrals, actions, and conversions from one or many search engines, but not all. Let me explain. Imagine that you have a keyword phrase that’s driving conversions from both MSN and Yahoo, but not Google. Wouldn’t that be nice to know, at a glance, in 10 seconds or less? Would that have an impact on the search phrases you bid for on Google AdWords? Exactly. The Opportunity Report saves you hours of analysis and decision making by highlighting those phrases that have opportunities on specific search engines, and exporting those phrases to a list that you can easily drop right into your bid management system.

To access this report, simply log in to your Enquisite reports, go to the Comparison tab, and select the “Opportunities” option. Then, choose the search engine (or engines) you want to use as a source, as well as the target search engine-the target will be the engine that isn’t referring traffic (or conversions) for terms that are performing well on other search sources.

And that’s it. We think it’s the most innovative way to do real keyword research on your site. Try it out, and let us know what you think!


New Enquisite Feature - Potential Analysis

January 9th, 2009 by Richard Zwicky

As you probably remember, we released the first full version of Enquisite back in August. After our initial release, we didn’t miss a beat-we got right to work on tweaking existing features and adding new ones. I’m happy to tell you today about the first of a series of new features, all centered around the concept of turning data into action.

The first new feature we’re going to discuss is the Potential Analysis. I’d like to thank Andrew Shotland for providing an original version of the formula upon which we ended up basing the math behind this analysis feature on.

The Potential Analysis is found within the Longtail report in Enquisite. It looks at search phrases and calculates a value based on the following variables;

- Referral rates - The number of referrals for a given term

- Conversion rates - The number of conversions a particular term generates

- Page views - Average page views/visits for the phrase

- Time - In seconds, the average time/visit for the phrase

- Bounce ratio for the phrase - The number of sessions driven by the phrase where the visitor left a page or site without visiting any other pages before a session timeout occurred or the total number of sessions driven by the phrase during an interval.

- Activity - in relation to the other terms, and outcomes

Two things worth pointing out:

- A bounce occurs when a website visitor leaves a page or a site without visiting any other pages before a session timeout occurs (we set this to 15 minutes).

- There’s a known issue when it comes to sample size. The larger the sample size, the higher the accuracy of the analysis. In other words, the more referrals, the more accurate the analysis.

Potential Analysis runs automatically and is displayed next to the appropriate column in your Longtail reports. By default, it’s turned on to show you the potential of your keywords. The way it works is simple. There are four levels of potential: High; Medium; Low; and None. A keyword with high potential has a good chance of keeping visitors on your site longer, looking at more pages. Keywords with no potential have virtually no chance of keeping a visitor’s interest.

But why does this matter? Well, for one thing, it helps you make better choices. If you’re interested in selling advertising on a page and are wanting to optimize that page to gain visitors, isn’t it good to know which terms are worth optimizing for? Think about it from a longtail perspective: If you have multiple variations of a phrase driving all traffic, the Potential Analysis will show you the phrase that will have the best ROI for you-in one simple box.

Now think of potential in relation to actions or conversions. Using Potential Analysis, you’ll be able to decide which phrases to optimize (or ignore) even if you have multiple converting phrases arriving on a particular web page from search referrals. This kind of analysis will help you take the guesswork out of the process and make quicker, more effective decisions.


New Year, New Features

January 8th, 2009 by Richard Zwicky

Welcome to 2009… where, like last year, I’m tempted to use the first line of this space to repeat my refrain of “I’m sorry for not writing more often.” I really am. And while I won’t go so far as to swear that I’ll write more blog posts in 2009, I do think that my schedule will make it a bit easier to get some blogging done.

I think that this year will be different, for a variety of reasons-the most interesting of which I should be able to share with you sometime after mid-January. Keep your eyes on this space, as a lot of last year’s hard work is soon to be unveiled.

Exciting things are happening around Enquisite, not the least of which is the recent addition of several new major features being added to Enquisite Pro. Over the past several months, we’ve completed 3 updates: first came the Opportunities Report, followed by the Potential Analysis functionality, and just last week, Map Overlays. Rather than describe all three right here, I’m going to let the suspense build a bit-I’ll give you the scoop on each feature individually in upcoming blog posts.

But first, a sneak peek at the next major update we’re tackling: our new Links report. Rand Fishkin has been eagerly (and somewhat patiently) waiting a seemingly infinite amount of time for this one-and it’s almost here. Though it’s still in production, we’ve shown it to a few folks and the general reaction to it is that they’re blown away. The Links Report will be delivered over the course of several iterations, with additional functionality being added as programming on it is completed. The first piece you’ll receive will be reporting, and the second starts into analysis. Right now, everything is on track for this to start coming out during January.

Anyhow, Happy New Year to everyone. You’d better strap on your seatbelts-we’re all in for one heck of a ride in 2009!


Search Engine Referral Rates by Page in SERPs

May 22nd, 2008 by Richard Zwicky

I’ve been asked a lot of questions over the last 10 years about how deep in the search results do people actually go before they clicked through on a result. In the past I’ve run a few reports on this information, but using only a month or two worth of data.

I just ran another report, but instead of two months, I used a sampling of our data representing ~300MM search referrals pulled from a much longer time period. What I found was the percentage of traffic from page one is actually increasing over time

I didn’t segregate out PPC or image searches, so this data does represent referrals in the aggregate. When we look at the hard numbers behind the data, the growing gap between page 1 and the rest is stunning.

2007-04 2007-05 2007-06 2007-07 2007-08 2007-09
Page 1
85.50% 86.03% 87.18% 87.79% 88.07% 88.40%
Page 2
7.61% 7.52% 6.90% 6.52% 6.47% 6.44%
Page 3
2.84% 2.71% 2.48% 2.35% 2.28% 2.21%
Page 4
1.30% 1.19% 1.09% 1.04% 1.00% 0.92%
Page 5
0.82% 0.75% 0.69% 0.66% 0.64% 0.58%
2007-10 2007-11 2007-12 2008-01 2008-02 2008-03
Page 1
88.42% 88.47% 88.81% 88.90% 88.78% 89.71%
Page 2
6.47% 6.44% 6.23% 6.19% 6.39% 5.93%
Page 3
2.20% 2.16% 2.05% 2.06% 2.04% 1.85%
Page 4
0.92% 0.91% 0.89% 0.88% 0.87% 0.78%
Page 5
0.57% 0.57% 0.55% 0.55% 0.54% 0.46%

It’s stunningly obvious that Page 1 generates the vast majority of traffic. Everyone knows this intuitively, but this data provides the facts to substantiate it. Page 2 still gets some traffic, but it’s negligible by comparison. While not appearing to hold much value, these placements are not entirely worthless.

Although a Web page which is found on Page 2 or lower on search engine result pages, (SERPs) may not get much traffic, you want to make these pages some of the prime targets in your SEO campaign. Although people aren’t finding these pages as often, they have incredibly high value simply because the search engines are finding and placing them, just a few small steps away from the success of page one.

Consider it from the opposite perspective: 90 percent of search engine users never venture beyond the first page of results. Listings found on page 2 of the SERPs are incredibly valuable, just not quite valuable enough to make it to page 1. These pages are your gems in the rough, and should be thought of as home-runs in waiting. With a little work, they can easily place on the first page, and you can hit it out of the park on an SEO campaign, just by concentrating your efforts in the right places.

Find the pages where you’re achieving page 2 or 3 placements, and focus on optimizing and improving the pages found there. Small adjustments can bump you up onto page 1, and will make your traffic soar. Get more pages moving up in the listings, and the effect on other pages in your Website is cumulative.