Twitter’s Edwin Chen and Alpa Jain wrote an interesting post on the company’s engineering blog about improving Twitter search “with real-time computation”. They discuss the challenges of real-time search from an advertising perspective.
I’ve written quite a few articles about real-time search, often comparing Twitter’s offering to Google’s lack of one since the deal between the two companies came to an end, effectively killing off Google’s. Yet, I’ve never really talked about the advertising-related challenges that must accompany such an offering. Twitter’s post makes you consider what’s going on behind the scenes.
“The queries people perform have probably never before been seen, so it’s impossible to know without very specific context what they mean,” the two say. “How would you know that #bindersfullofwomen refers to politics, and not office accessories, or that people searching for ‘horses and bayonets’ are interested in the Presidential debates? Since these spikes in search queries are so short-lived, there’s only a small window of opportunity to learn what they mean.”
“So an event happens, people instantly come to Twitter to search for the event, and we need to teach our systems what these queries mean as quickly as we can — because in just a few hours, the search spike will be gone,” they add.
To tackle the problem of serving relevant ads on real-time queries, Twitter monitors for which queries are popular at the moment, tracks stats on them, and then sends them to human evaluators, who are tasked with answering a variety of questions about them. To make a complicated story simple, when a response from the evaluator is received, Twitter pushes the info to its backend systems, so the next time a user searches for the query, it can serve a relevant ad based on context.
“For example, suppose our evaluators tell us that [Big Bird] is related to politics; the next time someone performs this search, we know to surface ads by @barackobama or @mittromney, not ads about Dora the Explorer,” Twitter explains.
Today, it might be better to see ads for Knicks tickets than for cereal on a Honey Nut Cheerios query.
So, when you see ads on Twitter as you’re searching for information, just know that a lot of work is going on behind the scenes to make it fit the scope of what you’re actually looking for. Check out Twitter’s post for a much more in depth explanation.