This is a post I wrote available HERE. I am posting on this site to make it more widely available.
For starters, Tealeaf tracks things that are beyond the scope of your average web analytics tool. I spent many years at a web analytics company, so I can highlight the important distinction with a real-world example of a successful campaign.
Before I came to Tealeaf, I had a client with an interesting issue. The company delivering this client’s campaigns reported a large number of click-throughs. And their click-throughs to impressions ratio was stellar. So this was a successful campaign, right? The problems were that the analytics tool showed only fraction of the reported click-throughs and conversions were actually very low. After some phone calls and discussions with their IT department, it turned out that their web servers could not handle the traffic. They had lost money on a “successful” campaign and had given their users (most being new to the site) horrible who’s who list of poor customer expience—slow-loading pages, status-code-500 errors, and the like. Now, if they had been combining the click-through data with their IT data in real time, this campaign may have had a better outcome. An alert would have warned them of the issue, they would have paused the campaign and worked through the hardware issues. Tracking your campaigns and site performance ensures that new customers, who are less forgiving, have a great experience.
Here are some tips on how you can ensure that your campaigns are tracked to ensure the best user experience and, therefore, greater campaign success:
Setup Tealeaf with performance metrics to measure your campaign’s user experience. If you are not measuring these metrics, put them in place right away. Most of these events come built in with newer releases of Tealeaf.
- 500 Level Errors – Track how often the server returns internal server errors with status code 500. Can your servers handle the extra traffic from a successful campaign?
- Cancelled Requests – This is a request to the server where the response could not be delivered. Did the user just give up on loading the page? Maybe he or she accidently clicked on a banner then quickly hit the back button or closed the browser. This will at least give you some clues.
- Server Gen Time – Create buckets of times for the server generation time of a web page. If the page is taking more than 30 seconds to load, this is bad news and most browsers give up on looking for a response from the servers. If the user has to wait more than a couple seconds for the page to load it’s a bad experience for that user.
- Network Time – Is your network slowing down the response back to the browser? Though this is not often an issue, you’ll still want to rule it out.
- Page Render Time – How long is the page taking to render on the browser. If it is too heavy consider making the landing page lighter or modifying it by browser version/type.
- Round Trip Time – From click-through to having the landing-page loaded, how long did it take to serve up the campaign landing page to the end user? If it took more than a couple seconds, start looking at server page generation, network or page render times.
Also, don’t forget your customer struggle metrics. Make sure to measure process restarts, form-field errors, time-to-complete, etc. The next section lists dimensions that you can use for your campaigns. Once you create the dimensions, don’t forget to add report groups and make sure all the events mentioned above are using the same report groups.
Tip #2: Group Lists
Adding your campaign IDs to a group list allows you to quickly identify campaigns that may be having an issue. Group lists are easy to manage and you can export/import from an excel file. Populate multiple attributes/dimensions with the campaign tracking code ID. For each attribute/dimension use a group list to classify the tracking codes as part of a value group. Some popular value groups and their uses are shown below:
- Campaign Code – Make sure the campaign code is in its own attribute/dimension to hone-in on the individual campaign that may have a problem.
- Campaign Type – Was this a paid keyword? A banner display? This shows how performance and user experience may differ from one campaign type to another.
- Campaign Name – The general name for the campaign that is running. If you’re running multiple campaigns, it shows how the user experience may differ from one campaign to another.
- Campaign Creative – What creative group was this added to? This shows how a creative helps the user experience or creates a disconnect in the user experience.
- Paid Keyword – If the campaign was for a paid keyword add the keyword to its own report. This shows how popular keywords may have low conversion because of user experience disconnects once they land on the site.
- Search Engine – Find out if users from different search engines are expecting different experiences.
- Branded Keywords – Track whether users click through from branded or non-branded keywords. Brand aware users often have different expectations from non-branded users.
I will share additional tips in my next post on this topic. Coming soon!
How are you measuring and monitoring your campaigns to ensure they are as successful as they can be?