The Apple Map Dilemma
J. Kim Fennell, CEO - November 2nd, 2012

The problematic launch of Apple Maps in September has quickly become one of the bigger technology stories of the year. After a string of successful product launches, Apple came out with a map product that not only had serious flaws but also replaced one of the most heavily used apps on the smartphone: Google Maps. The pressure was strong enough to force Apple CEO Tim Cook into the unprecedented move of a public apology coupled with a recommendation of several competitive apps that iPhone users could use in the interim.

At the same time, many knowledgeable pundits (here and here) pointed to the strategic necessity Apple faced that led them to invest in their own mapping and navigation technology. Apple likes to control its own destiny, both in terms of user experience as well and financial models. Continuing to rely on rival Google for such an important piece of the mobile ecosystem was not strategically tenable.

We feel Apple absolutely made the right strategic business decision to go it alone - but they clearly mishandled the execution on some basic, core geospatial technology.

So this is Apple's dilemma... and maybe yours as well?

Apple needed to control its own mapping and navigation platform, but was not able to produce a product that met customer expectations even after four years of serious investment. This dilemma is not Apple's alone. All the same business reasons that compelled Apple to be independent from Google also apply to many other companies, whether those are large platforms that want to compete with Google for mobile internet dominance or smaller specialty sites that worry that continued reliance on the Google Maps platform will leave them both undifferentiated as well as unable to control their own monetization opportunities.

Developers are increasingly looking for alternatives to Google's Mapping API - for differentiation, control over monetization, avoiding Google's restrictive ToS or to get technical functionality not offered by Google. The dilemma is HOW to achieve that.

Mapping is Hard

Maps are actually very hard to do well. It's relatively straightforward to generate map tile images. But the really hard geospatial work is under the covers: geocoding, geosearch, cartography, integration of map and POI data from all over the world from various suppliers, and routing across all the road types in the world.

Apple's problems are in data integration, not bad data. Some have speculated that the Apple Maps problem lays in the "data" they're using - mostly from TomTom. Let's clear that up right away - it's NOT a data problem. TomTom's map data is used by many applications worldwide (including Google) without these issues. deCarta uses both TomTom and NAVTEQ/Nokia map content. Like others, we enhance and refine their data to improve geocoding and routing, but it works very well out of the blocks. That's not the source of Apple's problems.

The key to successful map platforms is data integration. While Apple Maps has a very nice UI and some cool 3D imagery, they seem to have underestimated the complexity and magnitude of the underlying geospatial platform technology such as geocoding, map rendering, and global data integration.

Core geospatial work is the proverbial combination of Art + Science. It's not simple and it requires constant refinement. But to do it well you need experienced geospatial people. I know that sounds self-serving - but if you ask any knowledgeable geo-types out there about the Apple Maps issue you'll get the same response.

The best quality data set actually comes from blending, de-duplicating and prioritizing data from multiple sources. It's not clear that Apple always sourced the best of breed data from these multiple sources. For example, in a recent project in Europe for T-Mobile, deCarta blended and merged four different data sources to produce a "master" authoritative POI database. While Nokia and TomTom had very comprehensive data sets, there were better sources for certain categories such as restaurants and bars... sources like Qype in Europe. And therefore an intelligent process of merging and de-duplicating these data sets was necessary.

Geocoding is not something you do easily by throwing a few engineers at it and assume it can be done quickly. Heck, we've been doing it for 16 years and are still refining it - as we also do with our routing algorithms. It's never done, but it gets better and better through experience, user/customer feedback and with the manipulation of multiple sources of data.

We think Apple will eventually solve their mapping problem - they are smart people and have always been committed to quality products. But it will take time - especially to do it right for every country in the world.

That Doesn't Mean the Google Mapping API is the Only Alternative

For the record, we think Google's Map API is very good. For many developers, Google's API is fine, but increasingly developers want alternative mapping APIs because they want:

  • The ability to customize and differentiate their location services
  • More advanced functions than are offered by Google Maps' API
  • Their own branding, not that of a potential competitor
  • Control over the way content is used (both their content and the user data they capture)
  • The ability to monetize the LBS service themselves rather than just be a publisher for others

As is often the case, the "bar always rises" and developers increasingly want more capabilities, more control, more tools, more flexibility to build LBS Apps the way they want them - without giving up anything.

deCarta Takes a Different Approach: Our Technology, Your Business Model

For 16 years, deCarta has provided the critical elements of local search, mapping and navigation for hundreds of customers via our white-label geospatial platform. Our customers have included some of the best-known names in the online world (Google, Yahoo, Samsung, T-Mobile, etc). Our value proposition is simple: You control the customer facing application, the branding and the business model. We manage the geospatial technical challenges - the data integration, geocoding, geosearch, routing and the scalability that lies underneath.

It's a classic "Core versus Context" business argument. You focus on your core strengths - your business/consumer application, and leave the geospatial "plumbing" to a core expert like us. That means that you can come to market in weeks/months with a fully featured mapping platform that supports your brand, highlights your data and enables your business model. And it will work. At scale. Worldwide. From Day 1.

We use our patented, time-tested technologies to do the mapping, navigation and local search with a user experience that matches or beats anything on the market. All of this can be customized to emphasize your brand and your content.

We have perfected the underlying technologies that make mapping so hard - and we improve it constantly. That means cartography, geocoding, routing and integration of all the various kinds of geodata, including:

  • We integrate map data from all the major map data providers plus many regional players and OSM giving us coverage in 103 countries.
  • We use POI, traffic and other real time data from multiple providers, including yours if you have custom datasets.
  • We support single line geocoding and local search in all major countries
  • We have proven routing technology used by Ford, GM OnStar, FullPower's MotionX GPS Drive and 25 mobile operators that can power directions and turn-by-turn navigation.

You get the picture. We've been there, done that...worldwide. And we bring unique capabilities that you won't get elsewhere, like traffic-enabled routing, search-along-route and more.

Here are a couple examples that demonstrate how our customers use us:

FullPower's MotionX GPS Drive: The No. 1 Paid Navigation app on the iPhone in N. America for 38 months in a row. FullPower has a passion for quality and it shows in the reviews you'll see on the App Store - so we'll let you do the research. They have developed a beautiful, fully functional nav app that offers unique features, unique U/I and unrivalled performance. It runs on the deCarta platform and we host if for them.

MotionX is winning because they have the best app on the App Store. They have the best app because that's what they focus on - their core strength, and they leave the underlying geospatial routing to deCarta because that's what we're good at.

Sensis is the best-known local search company in Australia, with its own street map database, POI information and traffic data. They are owned by Telstra - the largest telco there. When they wanted to develop an online map service, they recognized that Google was a direct competitor, and that using Google and giving them their data was a long term path to irrelevance. So several years ago Sensis launched the mapping portal. The site fully utilizes the Sensis data and shows their own brand, without the threat of a major competitor lurking behind the map. And underneath is deCarta's LBS platform performing all the geocoding, map rendering and routing.

We've been at this a while - and that's a "good thing" because experience in this stuff counts as much as technology. Nevertheless our technology goes though continuous revision and we just launched our newest JS APIs, as well as new single-line, local search engine.

Check out what we've got today. The easiest way to do that is to sign up on the deCarta DevZone where you'll find our APIs and everything you need to get up and running quickly. You can download our APIs, documentation and source code - for FREE. We even have code to help to transition from Google's API to ours in a matter of hours. At deCarta we specialize in doing the complex mapping technology under the covers to enable our customers to build unique, successful applications. Maps are hard, but that just means that if you do them right, you separate yourself from the pack.