Topsy | Topsy Data Services

Topsy Data Services

Get access to current and historical social data and high-quality measurements.

Topsy Data Services provide access to all publicly available tweet content over multiple years together with proprietary measurements that quantify the volume and characteristics of communication for any keyword, term or domain of interest.

Topsy search and analytics technology ingests and indexes data from public social data sources such as Twitter and Google Plus. We subscribe to the full Twitter firehose, which as of March 2012 is producing over a quarter billion posts (tweets) per day, with volumes increasing at more than a 40% annual rate. Our technology indexes (makes searchable) and calculates a variety of proprietary measurements from these data within seconds of receiving each post. Our index goes back multiple years, and these historical reference points are used to calculate metrics surfacing what’s important from real-time communication across the world.

Each Topsy Data Services request consists of parameters identifying one or more keywords, terms or domains, a time period to be searched, and a list of the metrics required. Topsy fulfills the order by delivering one or more of the full-set of “census-level” content for posts matching the search-query, and/or associated volume and metric information.

Here are the key types of analyses provided:

All Content

All public posts (tweets for Twitter, posts for Google Plus) are output for the historical date range and search parameters specified. Full content tokenization, and support for Boolean and complex queries ensure precise content identification.Historical Mentions

Counts by day for all posts (census-based) show the number of times a keyword, term or domain for the search parameters specified were mentioned. Here’s an example of Historical Mentions used to perform brand analysis.

Share of Voice

Share of Voice analysis shows the percentage share of mentions for any related set of terms over extensive periods of time. An example of how Share of Voice analysis can be used.

Sentiment Analysis

Using Topsy Social Sentiment™, sentiment scores are calculated from the content found in every social post matching a set of search criteria, with scores output by day over extensive date ranges.

Link Trending

Using trending measurements (momentum, velocity, peak period), this analysis shows what the most popular links are on a given set of domains for the past 24 hours.

Keyword Trending

Using trending measurements (momentum, velocity, peak period), this analysis shows what keywords from a given set of terms are taking off the most for the past 24 hours. Cumulative (Cume) Exposure measures the cumulative exposure of communication matching specific search criteria over time. An example of how Cume Exposure was used to depict the Bin Laden tweeter, and this analysis is useful to show exposure over time for any marketing campaign using Twitter Influence Identification.

Key influencers for any keyword, term or domain are identified, now or for any specific period of time going back multiple years.

Related Discovery

Identifies what things are related to terms you already know about based upon an analysis of everything that’s coming over from Twitter in the past 24 hours. Provides the starting point for discovering new things related to known entities.

Results for each analysis type can be queried by keyword, term, domain for extensive date ranges, language filtering and geographic identification.

Measurements are calculated and output for each analysis type, including mention, sentiment, trending (velocity, momentum, peak period) and influence, with output transmitted using a variety of methods. We are happy to provide you with a sample of social data and measurements to demonstrate how we can fulfill your needs.