This scenario happens all too often, leading to extra website project costs.
It’s a couple of weeks before a website launch, the site’s almost finished development, heading into testing, and discussions about measurement begin. Analysis of the site leads to a high tagging implementation cost estimate. Sticker shock is felt by all. Less is implemented and tracked because there isn’t enough budget, or deferred, so the site launches without all conversion events being tagged. Sound familiar?
This can be avoided by starting measurement discussions during initial scoping and carrying on to design.
True, the above scenario won’t always happen. And for a simple site built in HTML, adding web analytics tracking code may indeed be a simple job.
Let’s take a deeper look at why costs are higher if measurement discussions occur after most of the design and development work is complete:
- Implementation costs to install tracking may be higher, particularly to track conversion/success events that are deployed within rich media or within client side elements such as flash, modal/pop-up windows.Costs are higher because developers likely have to go back into their code to add tracking code. And some changes may have to wait for the next development cycle because the cost is too high. So you can see that it would have been much easier to integrate tracking during the initial build.
- Custom coding costs could have been avoided. Perhaps there was an alternate rich media design, one that would not deflate the user experience, whereby a task completion results in a URL change. URL changes are typically automatically detected by web analytics tools. The most common opportunities for this type of trade off are forms and video presentation.
- Tracking code is added in a sub-optimal area of the site’s pages, resulting in less complete tracking. For the most complete data collection, the tracking code should to be added close to the top of the page, so that tracking has less chance of being blocked by slow loading page elements.If the templates have not been designed to allow the tracking code to be included easily near the top of the page, the code is typically added to the footer. In this situation, traffic will be lower because some activity may not be logged.
To avoid this problem, integrate measurement and analytics work earlier in the creative process, starting during the initial Discovery meetings, carrying through Design. Don’t wait until Development to add measurement and analytics.
Here are some questions business owners should ask during the early stages of Discovery and Design:
- What are the business goals for the web project? What activity do you need to be able to measure to allow you to decide if the web project is a good investment of time and money?
- What actions completed by visitors will show they’ve had a positive experience and have converted? What are the ‘success events’?
- What do you want to know about your web site visitors’ behaviour, your content, your campaigns, etc…? What type of analysis will be needed to help you scale the analytics value pyramid?
- Are any of the ‘success events’ identified in Discovery not automatically tracked by the web analytics tool? Is there a way to deliver this functionality so that they will be automatically tracked (tool specific, to reduce custom coding)?
- Do ‘success events’ include downloads, video, mailto links?
- Are there any clicks to separate domains that are a ‘success event’, and should therefore be tracked?
- How many sub-domains or separate domains are involved in a visit?
- What type of Content Management System and/or platform is being used?
In an actual project, we ask many more questions than the above. Consider these a starting point, to get everyone thinking about measurement and analytics early, instead of the week before go-live.
Has this been helpful for you? Please let us know.