Epiar Learning Center
<<- previous next ->>
Perhaps the best way to summarize our approach to data collection is to consider the metaphor of the needle in a haystack. While the popular reaction to the “haystack” is one of resistance, inconvenience, and futility to even begin the search, Epiar takes the opposite point of view. We actually invite more hay.
As a result of our ability to process and merge incredible amounts of data, Epiar seeks out as much extra raw information as possible. By constantly filtering through more and more data, we reduce the probability of false conclusions plus increase the chance of finding extra “needles” where we didn’t expect to find them. Using this holistic approach to data collection and hermeneutical analysis also permits us to see the bigger picture and provide a more well-rounded summary of a given topic. In fact, not a single research project passes where the results don’t surprise us too!
Epiar currently collects information from the following resources:
- Wordtracker
- Trellian
- Log files
- Yahoo!
- Alexa
- Our own internet spider, tools, and processes
Each datasource above offers unique and limited types of information. As a result, Epiar uses different resources for different types of research. The ultimate goal of every Epiar research project is to provide meaning and insight surrounding any given topic, and as every topic is unique, every topic requires a customized set of data to adequately answer the questions posed.
For instance, while Alexa, Google, and Yahoo! do not provide search query data, they do provide internet connectivity data, website summaries, traffic estimates, and their estimations of “relevance” in the online world. While Wordtracker and Trellian provide excellent search query data over unique date ranges, they do not provide internet connectivity data. Similarly, other datasources corroborate, challenge, and complement this information for entirely different questions.
As a result, we choose between a wide variety of resources for every research project.
In short, our approach is: “The more data, from more places, the better.”
