A client contacted us recently concerned that his traffic (according to Urchin) had dropped significantly in November of 2009. Since the drop, the Urchin has been reporting no more than 1 or 2 visitors per day. This 0-visitor week span has been pretty typical over the past 8 months:

Our first step? Install Google Analytics. I talk a lot about Google Analytics and how intuitive it is; plus I’m much more comfortable using it than Urchin. Lo and behold, Google Analytics had a very different traffic report than Urchin:

An interesting difference, n’est pas? It seems Urchin just stopped tracking data. I attempted to do some research on the occurence with no luck. Has anyone had a similar experience?

Nicki Hicks
Stumped by Urchin

Rand’s going to talk about how to make standard metrics actionable. [woot]

To make Analytics actionable, always ask:

Why am I measuring this?

What would I do if results were different?

Search Referral Analytics

# of visits per search engine over time – You want to see this number increase over time, in relationship to the search engine’s market share/overall global growth [Action: find out if it's a ranking/indexing issue]

# pages getting search referrals over time – Measure for each search engine and often. [Action: discover if indexation is an issue worth effort, read more about that here]

# of keywords sending traffic from a search engine over time – Look for increase in number of keywords sending traffic to your site – especially if you create a lot of content [Action: might be a rankings/demand issue; determine if content additions are accretive and what drives growth/shrinkage in search traffic]

Keyword Referral Analytics

# visits per keyword – Compare week by week, or even month by month [Action: Analyze top traffic drivers from a value perspective, check rankings for potential easy wins & get answers if traffic dips]

First time vs. returning visits per keyword – Business decision: which is better for you? [Action: Determine value of reaching new visitors vs. converting branded users (focus efforts on the more valuable one)]

Keyword rankings – They can be valuable! [Action: know if traffic spikes/dropoffs are from rankings, indexation or search demand shifts by matching traffic with rankings, SEOmoz has their own ranking tracker and you can track rankings using Analytics!]

Engagement Analytics

Time on site – Take it to the next level [Action: compare ROI metrics; if they correlate, improve on keywords/landing pages with low time on site]

# of Pageviews – Again, take it further than “sweet, it went up this month” [Action: Depending on your metrics, a "sweet spot" of pages browsed often dictates a conversion event - optimize towards it!]

Repeat visit ratio – [Action: Find what content/activities/referrers send engaged (read: returning) traffic and copy those while improving subpar pages]

Sharing/linking activity – Start tracking these actions, just like a conversion! [Action: Find patterns/sources that predict sharing activities (both content and CTAs) and make them testable conversion events]

Latent Conversion Tracking

Removing last-click attribution – Do you know what happens before their last click when they become a conversion? Look at full path analysis within Navigation Summary in your Analytics. This post will show you how to do this.

ROI Analytics

Lifetime Customer Value = CLTV (Customer Lifetime Value): how much money they spend, how many referrals

Cost of Acquisition = CAC (Customer Acquisition Cost): how much spent on sales, SEO, marketing, advertising

ROI = CLTV – CAC

Always be asking “What’s the ROI?”

Q&A

Q: In your opinion what’s the best Analytics software?
A: Depends on the person using it. Omniture experts will have difficulty with WebTrends. You can get a lot of the same information from all of the packages.

Q: Why the symmetric dip in SEOmoz’s traffic?
A: B2B websites will usually see dips on the weekends.

Q: How can I see the number of unique visitors for a particular section of my website?
A: At the bottom of Google Analytics, you can filter  landing pages based on page, like:

Q: Is there a big difference between different Analytics packages?
A:  Eric Enge did an extensive study with this at StoneTemple.

Q: Is there a way to automatically see rankings in Google Analytics?
A: Yes, here’s the Yoast link again.

Q: What’s the best tool for measuring ROI?
A: Salesforce.com, SEOmoz uses Infusionsoft.

Q: In your opinion, what’s your favorite part about Google Analytics?
A: Simplicity; the greatest thing is it’s easy to use, fairly fast.

Q: How do you compare bounce rates with other websites in your industry?
A: Google Analytics will let you view your site stats compared to similar sites in your industry using Benchmarking. You can find these stats under Visitors > Benchmarking. [You can set your industry at the very top of the page.] It’s up to you whether you actually believe these stats or not, obviously.

Q: Tell us something about bounce rates.
A: I only care about them when they happen in my conversion funnel. Always tie it to conversion and ROI.

Q: Why is rank tracker different than actual search?
A: Make sure personalized search is off (use &pws=0 after your query); local search could also change your results.

Q: When looking at social media sites, what are the metrics to track?
A: It depends on what you want them to do. Branding? Time on site. Go to your site and look at ads? Maybe page visits.

You’ll be able to find the archive of this webinar here shortly. And here is all of the Twitter chatter about the #mozinar.

Rand Fishkin, SEOmoz
@randfish

Since there was too much to say from last time, I’m continuing thoughts on drawing Analytics conclusions today.

Bounce Rate

Crap. Bounce rate went up.

• Did it increase drastically? For the most part, bounce rates don’t change more than a few percent unless something major happened. Did you get a powerful link that brought in ton of traffic? (In which case, maybe the call-to-action on that page wasn’t strong enough.)
• Does your website have a blog, or is it a blog? Blogs tend to have excruciatingly high bounce rates, since individual blogposts answer a question and once the question is answered, the visitor will move on. The key for a blog: is your high bounce rate coupled with a high new visits rate and/or a low subscription number? If visitors never return (your new visits rate is high), then you need to work to bring it down. Likewise with subscription rates – what can you do to make people subscribe? Sometimes it’s as simple as making a call-to-action to subscribe a little more “in your face”.

Yes! Bounce rate went down!

• Like pageviews, a decrease in bounce rate can sometimes be a negative thing. Is it difficult for people to find what they want so they have to visit multiple pages?
• Did you start displaying your phone number prominently? We’ve become a web-based world, but that doesn’t mean people don’t pick up the phone anymore. Many companies actually prefer to be contacted by phone, by burying any web-based means of contact and displaying phone number in the forefront. In this case, the call-to-action is to call so a low bounce rate (hopefully) means they took that action. By creating a means of measurement for this (by using a service for it), you’ll be able to successfully track this assumption.

A note on bounce rates: good vs. bad bounce rates depend heavily on your industry, as well as website vs. blogs. The flyte website typically stays around a 75% bounce rate, this blog about an 80% bounce rate. I’ve seen them get as low as 20-30%, typically in the hotel/inn industry.

Average Time on Site

Crap. Average time on site went down.

• Are your conversions going down too? Then you might have an issue on your hands. If people are spending less time on the site, and not doing what you want them to do while they’re there, take a look at your top landing pages and optimize them in order of popularity. Take time to make your call-to-action bold and attractive.
• Did pageviews also go down? That might not be a bad thing. Did you add a better call-to-action? Check your conversions. They might be up!

Yes! Average time on site went up!

• Has bounce rate also gone up? Then there might be a problem. Luckily, people are still taking the time to look for what they need; but then they’re leaving. Make it easy for your visitors with easy-to-use navigation and, can’t say it enough, your call-to-action.
• Add a really popular new page or blog post? Or perhaps you have a great new incoming link? In that case, people are actually reading what you write – which is fantastic! (Find out which page/post it is by sorting your top content by time on site.)

New Visits

Crap. New visits went down.

• First decide: do you want new visitors? If the answer is “yes”, then have you been writing articles? Blog posts? New content? That’s where you should start. Then, try social media and social bookmarking. You’ll reach new audiences (nearly) every day there).
• If you want to keep your old visitors coming back, give them a reason to. Add a blog, or a section that changes often. Maybe an email newsletter is the answer: incorporating links to keep your visitors hungry for more.

Yes! New visits went up!

• Have you been active on social media or social bookmarking sites? Make sure you’re well aware of any pages that go viral (and thank folks who might have ReTweeted you .)
• Check your traffic sources for where these new visitors are coming from. That should share even more insight.

In conclusion

Google Analytics isn’t perfect and you can’t see why every visitor does what they do on your website. But, you can get a great picture and even gain a lot of insight.

Nicki Hicks
Mind reading through Analytics

Hopefully you have Google Analytics installed on your website. (If not, I’ll wait for you to go and do that now.)

Done? OK, let’s continue.

As you probably know, there’s a vast amount of data on Google Analytics – nearly to the point of information overload. It might cause us to gasp and say Crap! My bounce rate went up! or Yes! I have 10% more visits this month than last month!

All of these seemingly deep, thoughtful conclusions may seem well and good, but what do they really mean? For some businesses, an increase in one stat might be great while for another, it might be negative.

I think the first and most important step in analyzing analytics data is ask yourself why something happened. So we’re going to go through a series of questions to ask yourself for the six major stats in GA: visits, pagesviews, pages/visit, bounce rate, average time on site, and new visits. (We’ll leave some of the deeper sections for another time.)

Visits

Crap. Visits went down.

• Holiday – or seasonal – downturn? It sounds ridiculous, but I noticed a drastic decrease in visits in many clients’ statistics in December (those not centered around holiday business, that is).
• Do you have a blog you’ve been ignoring? If you have a blog on your domain, or if you’re looking at GA stats for your blog, traffic will more often than not decrease if you haven’t been keeping up with it.
• Have you decreased your marketing efforts? Web and/or traditional? There’s an obvious reason for a decrease in traffic.

Yes! Visits went up!

• Did you send out an e-newsletter? I often see a drastic boost in traffic if an e-zine is sent out – and guess what? The more subscribers to an email newsletter, the more visits you’ll see! (Hence the need to always include a link back the website.)
• Did something go viral? An article you wrote? A YouTube video? A tweet? Track back through your analytics to see where the traffic came from.

Pageviews

Crap. Pageviews went down.

• Did you recently do a redesign? Pageviews going down may not be a bad thing. It may very well mean that people can find what they were looking for more quickly in only a few clicks!
• Did you launch a new landing page? A simple landing page with an easy call-to-action would certainly decrease clicks. Again…GOOD thing.
• Do people get frustrated and leave? Are people leaving without doing what you want them to? Double check by looking through the Content section: top exit pages.

Yes! Pageviews went up!

• Did you recently do a redesign or add a landing page that is a dud? Pageviews going up might be a bad thing. Making more clicks might frustrate your visitors.
• Did you add a scintillating new post, page, or copy? In this case, say if your pageviews are currently no much more than 1, increased pageviews is fantastic. Look through your content section to verify what did the trick – then keep doing it!

Pages/visit

All of the same issues as pageviews, as pages/visit is simply an average of pageviews/visit.

That’s going to do it for today. We’ll save the other half (bounce rate, average time on site, and new visits) for tomorrow.

Nicki Hicks
Long-winded

About a year ago, a client requested to receive weekly Google Analytics reports; a request I was more than happy to oblige. In order to stay on top of the client’s statistics, I added my own email to the list.

Now, months later, our contract has been completed and I’m still receiving those pesky emails. While I’m ecstatic my client’s website continues to perform well, it’s become annoying that Google Analytics emails me every Monday morning..like clock work.

This morning I decided to do something about it.

After some searching, I found the current Google Analytics Help section on the topic is somewhat out of date, having been written before one of the minor redesigns.

In an effort to help others pestered with the same annoyance I was having, the solution is simple.

Go into “View Reports” of the account in question. On the left hand menu, you’ll see hiding under “My Customizations”, an Email button:

Once here, you can change, manage, or update any of your scheduled reports!

Nicki Hicks
Happily no longer receiving GA reports