The biggest change is Alexa’s decision to drop exclusive reliance on the Alexa toolbar for traffic data, with Alexa now aggregating data from “multiple sources” to compile its statistics and web rankings. As part of the move, historical data from Alexa is no longer available, with data now only going back 9 months (we presume calculated using the new methodology).
Alexa is spinning the decision as a step forward without admitting to previous flaws:
Your ranking wasn’t wrong before, but it was different. Alexa toolbar users’ interests and surfing habits could differ from those of the general population in a number of ways, and we described some of those possible differences on our website. While the vast majority of sites’ rankings were unaffected by such differences, we’ve worked hard on our new ranking system to adjust for situations in which they could matter. The new rankings should better reflect the interests and surfing habits of the broader population of Web users.
A search of tech blogs saw many with significant drops in rank, where as political sites have had big boosts. For example TechCrunch and the Drudge Report were tracking similar figures on Alexa prior to the change, where as now the Drudge Report is a mile out in front.
Although regularly derided in the past for its often bizarre results (like YouTube having more traffic that Google), Alexa has continued to maintain popularity due to its broad global reach and completely free service provision. Time will tell if Alexa has done enough to appease its strong and vocal critics.