For years now, companies have been combing through Twitter postings, trying to glean any information that may help them improve their products and services. But with more than 250 million tweets sent every day, it would take an enormous staff to analyze the data. That task gave rise to hundreds of social media-monitoring companies, but they were still limited to the tweets that could be accessed by individual users. Twitter, however, has decided to make it easier for these companies to mine billions of messages for valuable marketing data. The company will open its archives and sell its old tweets. One of Twitter’s new customers, DataSift, has formed an alliance with the social network to get access to tweets going back to January 2010. “Twitter has really become an incredibly valuable information source,” said Rob Bailey, DataSift’s CEO. “There are a flood of companies wanting to get more use from it.” Starting next month, DataSift will launch a cloud-based service that will allow other companies to analyze these two-plus years of tweets to learn more about their customers. According to its website, DataSift promises customers will be able to “unlock trends from public tweets” and “access the full Twitter firehose.” Users will pay DataSift for only the data it retrieves. DataSift will then share part of the revenue with Twitter. The company emphasizes it won’t have access to deleted tweets or users’ direct messages. Twitter also has partnered with Gnip, a Colorado-based online data-mining company, to license its historical tweets. Because Twitter is a public forum, privacy watchdogs may face difficulties raising objections. “We welcome the privacy debate,” said Bailey, who is based in San Francisco. Twitter so far is not commenting publicly about its new partnership.