As a CTO, I find it it fascinating to watch the explosive growth of the API economy. APIs have become the fuel that’s enabling companies to create and launch new business models at unprecedented speeds. According to programmableweb.com, there are more than 16,000 APIs. HBR cites that Salesforce.com generates 50% of its revenue through APIs, Expedia.com generates 90%, and eBay, 60%.
However, the unintended consequence of the API economy is that it exacerbates the data fragmentation problem. The use of APIs to access more data in different systems causes more data fragmentation, on top of our already highly fragmented and dispersed enterprise environments
With the API economy, companies are transforming digitally by leveraging existing data and assets within the enterprise, and delivering new services and offerings to the market faster. As such, we are seeing the proliferation of disparate apps, hybrid environments, and the rise of self-service that’s creating a situation where data gets produced almost everywhere, in almost any format.
That means hundreds or even thousands of new endpoints exist inside and outside your enterprise. Consider what this means in a self-service app, data or reporting environment, where more people enter more data; unless that data is carefully managed and governed, there’s enormous opportunity for data to be mismatched and irreconcilable within systems, let alone across systems – leading to even more data fragmentation, plus data distrust.
The biggest challenge this presents is how you connect all this data and make it useful through an API. Here’s my take on the top data considerations to achieve success in the API economy:
- Carefully define reporting requirements with end-user report/data consumers – there are far too many data source permutations to be able to chase them all so you must narrow your scope.
- To achieve data access, start with data connectivity – once you know how to get access to the data you need, you can consider how to use it.
- Consider carefully how you can guarantee data trustworthiness and in particular how to harmonize data across different sources. Data from different sources will likely have been created with different usage in mind so pulling it out from the original system and context and merging it requires careful design.
- Always remember – having data access is not the same as data and system knowledge, either seek out subject matter experts to validate the information, or look for solutions that include packaged content you can trust.
Is your business equipped to handle the API economy? Let’s have a conversation – share your comments with me here.