Where to Start When Building AR for Enterprise

In a previous post, I described the difference between markerless and marker-based augmented reality (AR). While technology development will one day make marker-based AR mostly a thing of the past, today it is good for providing information, while markerless is good for discovery and delight. In this post, I am going to talk about which approach makes sense for what kind of customer. You can learn more about this from my talk at AR In Action at the MIT Media Lab.

A big component to any AR experience is context. With enterprise use cases, you have a lot of prior information and therefore context. You know the machines, the instructions, and what you’re trying to do. For consumers, you want to let them explore, so you cannot be so rigid about what you allow in terms of context. Or if you are rigid, you have to make sure you are hitting the sweet spot your customer cares about.

What exactly does context mean for enterprise situations? It means knowing what information you need, when you need it, and how to present it in such a way that it can be acted upon. With enterprise situations (the manufacturing floor or disassembling a jet engine for example), you know what information you will need or want. There is less discovery, and more information retrieval and presentation.


Consumers are not huge fans of marker-based augmented reality. They don’t get enough value out of it due to its prescriptive nature. Consumers like to explore and experience delight. In effect, they don’t want to visualize data; they want to experience it. An aspect of delight is surprise, so the very nature of marker-based AR is not going to fit that framework. This is why Pokemon Go took off – it allowed for both exploration and discovery. Snap (previously known as snapchat) led the way with facial filters that augmented without markers. Recently, both Facebook and Google have followed Snap’s example by using the camera as a way to provide AR experiences. It is interesting to note that while both companies are presenting consumers with augmented reality solutions, they’re doing so in very different ways that resonate with the company core. Facebook released their Camera Effects, which allows people to interact with their world in an augmented fashion in such a way you want to share and tell stories. At Google’s I/O Developer Conference this year, they introduced Google Lens that allows users to now search for information with their cameras using augmented reality technology. In a sense, this is what Google has always provided their users.


Enterprise customers are driven by costs and are therefore always looking for technology that saves or produces more than it costs to develop, implement, and maintain. Unlike consumers, enterprise customers are not looking for delight. They’re looking to optimize and enhance their operations with an eye towards minimizing errors, downtime, training time, and all other forms of cost and waste.

Sensors and connected devices in capital equipment can feed data back to a centralized database to allow for predictive maintenance. So you will know when something needs to be fixed, but what then? There is still some time involved in repairing a piece of capital equipment even if it has not yet malfunctioned. This is where enterprise augmented reality comes into play, as it can measurably reduce the time involved and increase the accuracy of repairing/maintaining equipment through a mix of solutions providing remote expertise and the hands-free nature of the platform. In 2006, a study showed that automotive downtime can cost $22k/minute. Per minute! If a company in the automotive industry spent $10M on an AR pilot project, it will recover its investment if it eliminates only 8 hours of downtime, or just over a minute per day in a continuously running environment.


Augmented reality allows us to better understand the world around us and to interact with it in new ways. For consumers, AR will be a new way of experiencing surprise and delight. For enterprise customers, AR will be the interface for humans to take part in the digital conversation that machines are having on manufacturing floors. Right now that can be most efficiently enabled with the use of marker-based experiences.

Do enterprise customers have use cases for markerless AR? Absolutely, they will. For now, while markerless technology and solutions develop and mature, only a few forward-thinking customers will embrace it. The majority of enterprise customers are still confirming the hypothesis that these technologies will save them money. Utilities are experimenting with AR technology, and they realize that while it is going to keep getting better, it’s currently at a point worth their attention. Enterprise has a lot of data they aren’t using, and technologies such as IoT and AR will finally allow companies to build and leverage such data sets. Before they feel the need to explore with markerless, there are many quick wins by just making that data accessible on-demand.

So what about discovery and more open-ended uses of AR for enterprise? Right now, that is mostly covered by remote expertise solutions. Rather than trying to code all the logic and know-how into the application, we rely on the expertise of well trained employees. At Augmented World Expo, PTC introduced Project Chalk, which provides such solutions for enterprise customers. In the future, as artificial intelligence (AI) grows stronger and we rely less on the remote expertise of a human being, we may see AI as the third leg of the stool (with AR and IoT) to really provide value to customers.