IBM Coremetrics. If your company is considering using it, run to the hills! Sure, it might offer greater accuracy than Google Analytics, and there's good live chat support, but even the simplest of reporting tasks can end up being hugely frustrating. I liken it to trying to run a kitchen when you only have access to the serving hatch.
If you're looking to develop an alternative, recognise that analytics tools should give you easy access to the raw data. I wouldn't want anything less.
To give an example, with CM you can choose to build ad hoc reports with either a full data set or a sampled data set. GA will sample data and give you an accuracy rating, but inaccuracies can grow from this depending on how you use the reports (e.g. comparing multiple metrics over a long time period). I have no idea if the premium version of GA gives you access to full data sets but that's even more expensive than CM.
Comments
IBM Coremetrics. If your company is considering using it, run to the hills! Sure, it might offer greater accuracy than Google Analytics, and there's good live chat support, but even the simplest of reporting tasks can end up being hugely frustrating. I liken it to trying to run a kitchen when you only have access to the serving hatch.
If you're looking to develop an alternative, recognise that analytics tools should give you easy access to the raw data. I wouldn't want anything less.
I input raw logs to Splunk and built great looking dashboards with full access to raw data.
GA is a free joke.
How is it more accurate? Are they not using javascript? Do they do less sampling for larger properties?
To give an example, with CM you can choose to build ad hoc reports with either a full data set or a sampled data set. GA will sample data and give you an accuracy rating, but inaccuracies can grow from this depending on how you use the reports (e.g. comparing multiple metrics over a long time period). I have no idea if the premium version of GA gives you access to full data sets but that's even more expensive than CM.