Twitter, which recently announced its Promoted Tweets advertising program, has acquired the maker of a cloud-hosted Web analytics application, Twitter said Thursday.Smallthought Systems’
Trendly lets Web site owners dig deeper into the usage and traffic data that Google Analytics collects about their sites.
The Smallthought staff has become part of Twitter’s analytics team, where they will integrate Trendly features and technology into Twitter’s existing systems, as well as help develop new products.
In April, Twitter launched with a limited number of partners like Starbucks and Best Buy the Promoted Tweets program, which is designed to let companies market their products and services on the popular microblogging and social-networking site.
As with all advertising services, a key to the success of Promoted Tweets will be Twitter’s capacity to analyze the popularity and effectiveness of these ads so that campaigns can be evaluated and optimized.
The analytics for Promoted Tweets may be more challenging than for other online advertising programs for a few reasons, including the fact that the ad format for Promoted Tweets will be the same as the format for regular “tweets” and that Twitter’s usage is going through the roof, with 2 billion “tweets” posted in May, according to Web monitoring company Pingdom.
“Every day millions of people use Twitter to create, share and discover information, and as we grow, analytics becomes an increasingly crucial part of improving our service,” reads Twitter’s announcement of its Smallthought acquisition.
Credit & Resource: http://www.networkworld.com/news/2010/061010-twitter-buys-analytics.html
=-=-
How Trendly helps your web marketing
Google Analytics does a great job of collecting your data, but it does a bad job of helping you understand it. For example, at Dabble DB, we are very interested in how many people come to us after searching for online database. Here’s what that keyword’s traffic looks like for the first few months of 2009:
It’s definitely going up and down a lot, but we don’t really care about daily fluctuation, we care about the trend. On average, how many people should we expect each day? And is that changing?
Trendly uses a statistical model to answer that question for us. It assumes that our traffic is going to stay the same for days, weeks, or even months at a time, but every once in a while something is going to change — like our ranking for “online database” improving, sending more people our way. The red line shows how Trendly sees this keyword over the same time period:
According to Trendly, our daily visitors from “online database” went up from 24 to 40 in early January, and then again up to 50 in early February. It reports these changes as items in a news feed:
By boiling several months of data down to a couple of items in a news feed, Trendly helps us keep on top of many different things at once. For example, this same feed actually tracks all of our search keywords:
You can see that our traffic from searches for co-founder Avi Bryant had a bump at the end of January (around the time he gave a talk at CUSEC), but went back down again. The sparklines beside each news item show those patterns concisely.
The news feed is continuous, and shows up to three years of history. If we want to see the earlier change to “online database”, we just scroll down a few weeks:
The news feed does a great job of showing what’s changed, but it’s nice to put those individual keywords into a wider context. A chart running down the left provides the big picture:
The chart is locked to the same timescale as the news feed — and so, unlike most charts, time runs vertically. Each colored layer in the chart represents a single keyword. For example, the big green layer represents visits from “online database”. If I click on either the chart or the news item, the layer will pop out:
You can see how the green layer gets wider at the same time as the news item appears. The wider the layer, the more visits we get each day. The wider the chart as a whole — the sum of all the layers — the more visits we get in total from search. You can also see that Trendly tries to make things more meaningful by clustering similar keyword phrases together: The popup for “online database” shows that, while most (86%) of the searches in this cluster are for exactly “online database”, there are less frequent similar phrases that are also being included here, like “make an online database”.
Trendly isn’t just for tracking keywords. It has feeds for many of the reports you’re used to from Google Analytics: referrals, content, ad campaigns, and more. If you use goal tracking or ecommerce, Trendly also helps you track those:
