Archive for the ‘Tracking & Analytics’ Category
Thursday, March 28th, 2013
More and more, search marketers are seeing “(not provided)” in their Google analytics associated with keywords they are tracking. This is frustrating for most of us, but it doesn’t have to be the end of your keyword tracking. There are other ways to mine the data that is important to you.
One way you can measure traffic by keyword – though admittedly it isn’t perfect – is to track the traffic for your landing pages instead.
Measuring Landing Page Traffic
If you’ve done a good job of optimizing your website’s landing pages, then each page should be optimized for one or two keywords. But no more than that. So how can you use those landing pages to measure traffic for your keywords?
Let’s say you have four landing pages optimized around five different keywords, like this:
- Landing page 1 is optimized for keywords A and B
- Landing page 2 is optimized for keyword C
- Landing page 3 is optimized for keywords D and E
- Landing page 4 is optimized for keywords C and E
Your task is to find out how much traffic you are getting for each of your keywords based on the traffic you are getting for the landing pages. So how do you do that?
Let’s say each of the landing pages received this much traffic last month: Landing page 1 = 1,000 visitors; landing page 2 = 1,500 visitors; landing page 3 = 500 visitors; landing page 4 = 2,500 visitors.
One way to break down your visitor count per keyword is to split the traffic count evenly for each keyword associated with a single landing page. So, on landing page 4, each keyword would receive half the traffic count – 1,250 visitors. You could then ascertain that keyword C received approximately 2,750 visits last month (1,500 + 1,250).
That’s probably not very accurate, but it could be close. The problem is, the more traffic you receive for each landing page, the less accurate this measurement is going to be.
Historical Keyword Traffic Measurements
Another way is to go back through your history. If you have access to historical records, then you could take a look at the last known traffic numbers for each keyword you are tracking. If you know, for instance, that keyword A had 800 visitors at last count and keyword B had 400 at last count, then the ratio is 2:1 in favor of keyword A. You could assume the same ratio holds and you could split the traffic numbers for landing page 1 accordingly.
Again, this likely isn’t an accurate measurement, but it can give you some idea as to how much traffic you could be getting for each of your keywords. It sure beats flying in the dark.
Friday, March 15th, 2013
There’s a company in Palo Alto, California that is calling itself “Google Analytics for the real world.” The company’s name is Euclid. I think this is a sublime idea.
Euclid collects data for retail businesses using smartphone technology to track consumer movement. Here’s how it works, in a nutshell:
Euclid installs sensors in retailers’ locations that can measure how many people walked by a store, how many walked in and for how long they stayed based on when their smartphones emit a kind of radar searching for wireless Internet signals.
The incredible thing about that is it allows retail store owners an opportunity to use actionable data that, before now, they haven’t had access to. If you know how many people pass by your store without entering, then you can work on ways to draw those people into your store – and track your results. An increase in store traffic after a few merchandising or promotional efforts and you’ll know your marketing campaign is working.
You can do the same thing with in-store traffic. Are people browsing and leaving without purchasing? Maybe you need to run a few store specials to encourage more buying. You can track your results to see if it is working.
This is powerful retail technology and a very clever use of smartphones for tracking purposes. Brick and mortar retail businesses now have access to some of the same tracking technology that online businesses have enjoyed for years.
Thursday, January 3rd, 2013
When it comes to business metrics, no metric is as important as the cost of acquiring a new customer. It is this metric that determines whether you are profitable or not.
For instance, if you spend $20 to acquire a new customer that buys a $12 product, then you are losing $8 on that customer – especially if you only have one product.
This is where the cost of customer acquisition gets sticky. Many small business owners don’t understand the high value of keeping a good customer. Service businesses are better at this than businesses that sell a product. You can take a loss on the acquisition of a new customer if you keep that customer coming back and buying more from you in the future.
Take the previous example: An $8 loss on a new customer can be turned into a gain if that customer buys your next product at $15. Now you’ve spent $20 to acquire the customer and turned that investment into a $22 win – a profit of $2.
There is a caveat. Your products and services have to be quality products and services to ensure that customers come back for more.
I’m not saying you should lose money on your first sale. If there is a way to control the cost of aqcquisition, and there usually is, then you should control that cost. If you can reduce the cost of acquiring a new customer from $20 to $15, then you might only lose $5 on the new customer. But you’ll still make up for it on subsequent orders.
Of course, if you can acquire a new customer for less than the cost of your first sell, that’s a better plan. But here’s the overall view: Keep your cost of acquisition low enough that your profit margin is worth keeping your customers. Manage that cost of acquisition. Lower isn’t necessarily better, but if it is too high you’ll never earn the profit you deserve.
Friday, December 28th, 2012
For the longest time Google Analytics was the only free open source analytics program. At the very least, it was the best of breed so to use any other was virtual suicide. Many online marketers opted to use a paid analytics program so they could keep their information private and away from Google’s eyes, or they had other reasons. But there is a new alternative to Google Analytics that is free and open source and is picking up speed.
It’s called Piwik.
Piwik has some pretty awesome features that should make any online marketer jump at the chance to demo the software. Features include:
- Full ownership of your data
- Ability to generate e-mail reports in HTML and PDF formats
- Unlimited users, unlimited websites
- Custom login panel
- Delete old data you don’t use
- URL and IP blocking
- First party and third party cookies available
- Export your data in a variety of formats, including Excel, PHP, Json, and XML
- Compatible with a variety of servers, including Windows, Apache, Linux, IIS, MacOS, Solaris, and more
Piwik is so flexible you can install it on your own server or host it on Piwik’s servers. They also have a Developer Zone and a community forum, so you can get technical help when you need it.
Am I saying Piwik is the solution you need? No. I am saying it’s a useful alternative to Google Analytics, if you are in the market for one.
Monday, July 23rd, 2012
There is more to analytics than simply looking at how many website visitors you have and where they are coming from. One often overlooked analytics measurement is which part of your pages are the most popular among your website visitors. To get that information you need to employ eye tracking studies so you can see where your visitors are spending the most time.
It’s important to note that eye tracking does not track which pages are most popular. It tracks which areas on a page are most popular. Theoretically, that could be different on every page of your website.
If all of your pages have the same web design, it is likely that the most popular parts of your web pages will be the same, but it’s not a guarantee. That’s why you should conduct an eye tracking study for every page of your website. It’s particularly helpful for landing pages.
If you know, for instance, that your Buy Now button is the least viewed part of your web page, then you can move your Buy Now button to a more viewed area of your landing page. That won’t necessarily mean that more site visitors will click the button. Maybe your call to action isn’t strong enough. But if no one is seeing the Buy Now button, then they for sure won’t click it.
You can also determine the best places to put your opt-in boxes, your advertising, call out boxes, and other page elements on the basis of your eye tracking studies.
Eye tracking is one analytics tool that you should not overlook if you want to increase your website’s conversions.
Tuesday, May 8th, 2012
The Panda and Penguin Google updates have certainly spooked a lot of people. The slightest move in website traffic volume is now being blamed on either one (or both) of these updates, and whilst significant drops in traffic are a worrying sign, a closer inspection sometimes shows the opposite to be true. I recently looked a website that had seen a 50% drop in traffic, and that drop coincided with the Google Penguin update.
Google Analytics is a great tool when used effectively. Using this tool, I could drill down into the various statistics for the both the day before the Penguin update, and for the day after when there was a significant drop in traffic. What I found to be interesting was that Google’s Webmaster Tools showed there was not a significant change in the average search rankings for the targeted keywords. So why the huge drop in traffic? Surely people didn’t suddenly stop using those search terms?
In this case, the drop in traffic came from two sectors, mobile, and Europe. For some reason this website had been ranking highly in search results in Europe. It was also ranking highly in mobile search. Neither of these two sectors were important to this business. They were not geared to provide a service to European customers, and being net based business, mobile traffic had tiny 0.01% conversion rate, so the loss of that traffic wasn’t harming the business.
Whilst there was a distinct drop in traffic numbers to this site, the number of sales were the same both pre and post Penguin. In fact, the post Penguin conversion rate as a number almost doubled – and so it should, half the traffic for the same volume of sales. When looking at traffic numbers from search results, it’s important to dig to find out exactly where you have lost traffic. Is it one page, is it site wide, is specific to a region or connection method?
If this business had kept a closer eye on their search analytics, the fact that almost a third of traffic was coming from Europe may have opened doors for new business. That window has closed now, however, they can look at much improved conversion rates.
Sunday, April 1st, 2012
There is an interesting post on Small Business Newz that discusses the much vaunted “Big Data Train.” The article makes sense up to a point, however, there are times when I wonder if small businesses are over analyzing their online presence. Being online does open up the prospect of measuring everything that happens. Whether it’s email views and responses, visitor activity on your site, or how readers respond to social media activity – there is a way of collecting data and measuring that data against a range of standards. Bit is it too much?
Small businesses have succeeded for centuries on very basic data – if they bothered to measure it at all. There is a real danger now that business owners are becoming slaves to this data, spending more time collating and analyzing than they do in actually running their business – then complain that their business is going down hill. Sometimes, the key is analyzing your analytics – do you really need all that data?
To succeed online, you need to know where to draw the line. You need to be able to filter out non essential data and to concentrate on the data that is really important. For many online small businesses, this is easy. Search rankings, traffic flow, and visitor activity. If you rank number one, receive 500 visitors each day with a 5% conversion – that is easy to analyze, and as you make improvements to your business activities, you can easily measure their effects. For most small businesses, the big data train and cloud computing all sound good – but are they actually good for your business? You know your business, don’t become snowed under by too much useless data.