Sunday, August 24, 2008

Using Direct Traffic to Measure Site Growth


Here's a way to squeeze your Google Analytics to get a meaningful report of your comic's progress.

The idea is to focus on direct traffic and see how it's trending.

Remember, total traffic is easily affected by how many ads you have running, or whether you get a lot of traffic from places like Stumbled Upon or EntreCard.  We can call that traffic "visitor traffic" because many of those people will not linger on your site, or return.

Direct traffic, which you can locate in numerous ways on your analytics reports, is traffic coming from browsers instead of web sites. That traffic is more likely to be made up of your regular readers, so good numbers and growth in this category are useful.

When I checked my own comics, Scratchin Post and Li'l Nyet, my gut told me to expect a 100% monthly increase for SP and 150% for Nyet. I wasn't really sure, because of a lot of unique events, like one site being featured on a major site and dumping 10,000 hits on me, and my phasing out of sites that I feel generate too much phony traffic with high pass-through rates.

Let me tell you how to do it, and then I'll report how my sites did. Go to your Google Analytics home page. Click "Traffic Sources" in the left side column. Then click "Direct Traffic," which is well to the left of the pie chart. It will announce in prominent text something like "500 visitors came directly to this site." Write down that number. Next, we'll do the previous thirty days.

At top right but still under the orange stripe is the date period you are in. It will read something like "May 20 - June 19." On the right side, almost to the scroll bar, is an arrow for a drop down menu. Click that arrow and a box will open, with a calendar array. Above the calendars is a link called "Timeline." Clicking that will give you a tool for navigating your time period -- a box that you can drag leftwards, into the past. As you drag it, watch the date range change in the boxes to the right. Drag until you have gone back 30 days. Finally, click "Apply," which is a button to the right of the scroll bar that the box rides on.

Now you are looking at data for the previous thirty day period. Collect your direct traffic number and write it down, making note of the time period.

Repeat as often as you like. 

The math for calculating percentage change is simple, but I'll repeat it for those who are rusty. First a talk-through, then an example. You can skip this part and just eyeball your data if you prefer -- you'll at least now whether you are rocketing, creeping, flat or sinking. If you skip it, jump to the paragraph beginning with "Obviously."

Number of visitors from the most recent 30 days x 100 divided by visitors from selected earlier period.

500 direct visitors in May x 100 = 50,000
20 direct visitors in January

50,000/20 = 2500% direct traffic increase from January to May.

Obviously, a very nice result. My sites have been around less time, so I could only use several months.  It turned out that when measuring what I'll call "loyal traffic," my sites grew by well over 400% and 600% respectively over the last several months. 

I'm not aware of any public device which reports direct traffic to strangers, which means that if someone like me comes along and says your site seems to be doing poorly, and I show a graph of total traffic, you can say, hey wait. I stopped a major ad campaign about when the slide began, and my direct traffic has continued to rise by 150%/month. So in fact, you may be doing better or worse than public data indicates. There may be times when you want to offer a rebuttal, or when you want to remain silent.

Remember:  Anyone can get traffic, but whether it sticks is what determines your prospects.