How Do You Spell API?

It seems like everywhere I turn these days, there's a growing enthusiasm surrounding network APIs as the savior of 5G. I've already lived this moment throughout my career, so every time I see it, it feels like déjà vu all over again!

I can still remember back in the early 2010s when I used to joke that I couldn't even spell API, an abbreviation for Application Programmer Interface. What is the API exactly for? Essentially, it allows software developers to access other software, platforms, or hardware resources through a standardized set of instructions.

It requires a platform or, (in our case) a network owner to develop a set of rules, guidelines, and policies accessible through software code, determining which capabilities they're willing to provide to developers and to what extent of trust/security.

In the 4G LTE era, there was significant interest across the telecommunications community in APIs that would allow developers to build applications that could access key capabilities of telco companies. Think of things like sending an SMS message from an app, billing a telco customer's account, or even simply validating identity with network-level capabilities.?

At the time, we were in the early bonanza of the Apple App Store or Google Play, and the telcos were highly concerned about being sidelined by customers and unable to participate in the app revolution.

However, telco operators needed APIs that interoperated across different operator networks. If I created an app that sends an SMS message, I wanted to make sure that it worked seamlessly regardless of whether the customer was sitting on AT&T, Verizon, Vodafone, Orange, Telefonica, or NTT DoCoMo. This led to the realization that interoperable APIs were essential.

An Unfortunate Acronym

In what was one of many unfortunate acronyms for telcos, the industry formed a new community called the Wholesale Applications Community (WAC).?

WAC represented multiple global operators working diligently to create a set of interoperable network APIs. They successfully developed the first technology around a billing API, but getting 30+ different telcos to rush into a software world was challenging.

Ultimately, for many reasons, WAC was eventually shuttered, and the IP was moved into the GSMA OneAPI program. That was more than 10 years ago and you never really heard of the program after that point.

What now? Last year, the GSMA announced a new program called the Open Gateway Initiative, which, as you guessed it, was another effort at network APIs. At the Mobile World Congress, once again, network APIs dominated discussions as the ultimate solution for the network and 5G.

Since this announcement, we've heard all sorts of major headlines about the potential of network APIs, including that they represent a $300 billion opportunity. Having lived this reality already, I couldn't hold back and wanted to share my thoughts about network APIs as the so-called savior for 5G.

My (biased) Viewpoint On Network APIs

Are network APIs a $300 billion opportunity? The short answer is I doubt it. The longer answer ties to how you monetize an API. Some companies focus on APIs and have a business model around them, but you have to ask yourself what the API's real purpose is.??

In essence, the API is nothing more than an "enabler" of some broader solution or service that utilizes the network. It's like a cell tower or a radio that helps you make a call. We don't think about paying directly for the cell tower or radio; we just call folks on our phone and the telco sets a pricing level for that service that helps them maintain towers and radios.

In the same way, I've seen so many bad business cases about charging for APIs, and each one failed within a telco. Why? The reason is simple: instead of focusing on the customer's needs, the goal is to charge for something, whether or not we're solving the customer's problem. It was short-term focused on getting cash in the door vs. long-term focused on creating unique solutions that “grow the pie.”

The path to value in an API is simply usage. The more people use it, the more they're using your network. The more they use your network, the more valuable your network. Unfortunately, charging directly for the API undermines this entire model.

Can you make network APIs work interoperably?

The short answer is yes. The Long answer is that's the wrong question. I walked into a situation where I had five major global operators trying to make a single network API work across them and quickly realized that we had an N:N problem.?

Effectively, some operators were very interested in a particular API while others were not. The interested ones saw the upside and were willing to overcome obstacles. The ones who were not saw nothing but risk and threw up roadblocks. Ultimately, we solved that problem but then ran into a bigger issue.

What prevents network APIs from taking off?

Commercial terms. While we could come up with all kinds of technical solutions for creating an intermediary platform to translate technical calls from one operator to another, we ran into a buzzsaw when it came to the commercial agreement.

Operators have very different commercial risk profiles, strategies, priorities, regulations, and ways of doing business. While we could create technical solutions, how do you handle the question of who owns the developer, who owns the customer, how everyone gets paid, and how this meets the local regulatory framework at each location?

We were able to go from 0 to 3 working prototypes in six weeks, but we never effectively solved this problem. Until the benefits are sufficiently clear and compelling, no operator is going to embrace this approach. Call me sarcastic, but it's human nature.

What do I think?

The reality is that a network API helps build or access network capabilities to benefit customers. The first question is, what problems are people trying to solve, and what do they need from the network to make that happen? The next question is, what can they already do today, and what would need to be "exposed" through an API?

I read articles with $300 billion market sizes with extreme skepticism. It's not that I don't want that to happen, but as I recently shared on a similar LinkedIn exchange,

Whenever we saw these massive market estimates for an API, it didn't take long for it to start breaking down.

The first question I would ask is – "Who's going to pay for it?"

Answer – "Some developer or the client."

Second question - "Why?"

Answer – blank stares

What else do you want to know about APIs and their future?in the telecommunications industry?

Subscribe to ?????? ?????????????? ???????????? newsletter and be the first to receive weekly high-quality content to help you tackle complex strategic challenges in the telecom industry.

#newsletter #telecom #telco #value #vision #ruralbroadband #API