Apple’s maps

Apple's maps problem isn't about software or design. It's about data.

Apple Maps screenshotI promise not to make any snarky remarks about Apple’s maps disaster, and the mistakes of letting a corporate vendetta get in the way of good business decisions. Oops, I lied. But it’s good to see that Tim Cook agrees, at least about quality of the maps. It’s humbling for a company like Apple to issue an apology.

The real issue isn’t the apology, but what happens next. Google seems to be in no hurry to submit a maps app. It’s unclear how much patience Apple’s customers have; on my Android phone, I probably use Google Maps more than anything else. Not having public transit information when I’m in New York would be a deal breaker for me. I suspect Apple’s fans are more loyal, but even that has limits. How long can the fanboys wait?

One article put Apple’s mapping efforts 400 years behind Google. That’s a lot of catch up. And Google certainly isn’t standing still: their addition of underwater photography to “street view” is spectacular, and may serve us well when sea levels rise. But that’s not the point, either. Apple doesn’t have to “catch up” to Google, though I’m sure they’d like to. They just have to get a product that’s good enough. I don’t think that’s a three-to-six-month proposition. But it could be done in a year or two.

Here’s the difficulty. As Stephen O’Grady has pointed out, the problem with maps is really a data problem, not a software or design problem. If Apple’s maps app was ugly or had a poor user interface, it would be fixed within a month. But Apple is really looking at a data problem: bad data, incomplete data, conflicting data, poor quality data, incorrectly formatted data. Anyone who works with data understands that 80% of the work in any data product is getting your data into good enough shape so that it’s useable. Google is a data company, and they understand this; hence the reports of more than 7,000 people working on Google Maps. And even Google Maps has its errors; I just reported a “road” that is really just a poorly maintained trail.

Maps isn’t Apple’s only data problem. Apple’s spelling correction is an embarrassment. Google has this nailed, both from the standpoint of accuracy and user interface, even to the point of auto-suggesting the next word (with uncanny accuracy). When I’m typing on my Android phone, I don’t even bother correcting mistakes: I can trust Google to pop up the correct word, often before I’ve finished. On my iPad, it’s another story. As Google’s Peter Norvig has said, “We don’t have better algorithms. We just have more data.

Again, I really don’t mean to be snarky about Apple, so I’ll stop here. Apple has been successful by being a great product company. But to move forward, they have to become a great data company. Likewise, to succeed at offering services (including map services), they have to become a great operations company. It isn’t just about product design. I have no doubts that Apple is capable of making the shift; if they do so, the year or two it takes them to get a map product that’s “good enough” will be well spent. But if they don’t make this shift, they could be in for a rough time.

tags: , , , , ,