Former Google Engineer Explains how Google Maps Calculates Your ETA

December 27, 2013 - Written By Eric Abent

As amazing as much of it is, technology can be bewildering to folks who never get a look under the hood. Take, for instance, Google Maps – it has a lot of features that are generally very handy, but how those features work is still pretty big mystery to most end users. Specifically, have you ever caught yourself wondering how Google estimates your ETA when you’re using Google Maps to find your way? If you have, you certainly aren’t alone, with a new question on Quora asking precisely that. As luck would have it, former Google Engineer Richard Russell stopped in to answer the question, and his response gives an interesting look into how this aspect of Google Maps works.

Russell says Google bases its ETAs on “a variety of things, depending on the data available in a particular area.” Information like posted speed limits and recommended speeds play a part, as do “likely speeds derived from road types.” Historical average speed times are used, along with the actual travel information from Google Maps users who have taken the same route before. Unsurprisingly, real-time traffic data is the final piece of the puzzle, and all of this information is combined to give Google a snapshot of the situation at hand. Once it has all of the necessary details, it can give you an estimated time of arrival.

Russell also says that the companies that deal in reporting live traffic “compare their predictions against actual time in traffic to tune their algorithms and data sources.” Obviously, this means that the companies with access to the best data will be able to offer the most accurate predictions. Given the number of Google Maps users out there, it’s clear we can count Google among those companies.

However, those predictions aren’t right all of the time, and it wouldn’t surprise us to hear that some people find their ETAs to be at least slightly off frequently. Russell also tackled this little problem, saying that we shouldn’t expect the “predictions to be accurate any time soon.” This is due to the fact that traffic is “inherently unpredictable,” and indeed, all of the data in the world can’t account for things like crashes or other slow downs that are the result of humans rather than road conditions or posted speeds. In spite of that rather bleak prediction for the future, it’s still nice to know just how Google Maps calculates our ETAs.