Give me Integration or Give me Death

Tweet about this on TwitterShare on LinkedInShare on FacebookShare on RedditEmail this to someone

If you are a SaaS business, heed this warning: Integrate. Or die.

At OCA, we continuously evaluate our investment themes. We look for long-term, sustainable macro trends. If we are right about these themes, it gives us two advantages: (1) our companies will have more margin of error in execution. In niche markets, a company has to execute flawlessly but in large and growing markets fueled by macro trends, you can screw up a couple of things and still achieve a good exit for all shareholders. And (2) once educated in a theme, there are great economies of scale in due diligence, exits, recruiting, deal sourcing and other areas.

One macro trend that has been underway for a long time is businesses and consumers “pushing” data and applications to the cloud. Back in the mid 1990s, some of the first “cloud” applications were not truly Internet applications but instead applications running on a local machine accessed remotely using Citrix or similar. Fewer serious applications were built as websites in those days because HTML, javascript and browsers lacked the interface complexity to compete with an application running locally on a machine. Bandwidth was narrower too, making it hard to compete with a locally-running application. Eventually, these constraints abated and user expectations and desire for simple interfaces allowed developers to greatly close the gap in functionality/user experience between an application in the “cloud” and an installed one. While there is still a gap (e.g. Evernote), I think we reached a tipping point in the last couple of years that has fueled the push toward the cloud.

And the cloud is awesome. Consumers and business are demanding access to their applications and data no matter where they are and on any device. Businesses have gotten comfortable with web interfaces and mobile-constrained screens which were pre-requisites for cloud adoption. Now the largest obstacles are security and privacy (they are different things) which companies like Alert Logic (disclosure, OCA is an investor) and Spideroak are attacking with success.  I have been a cloud believer for a long time: I started using cloud backup in 2007 when the online backup tools were terrible. I patched together my own backup implementation using Syncback, encyrpted zip files and a large FTP repository at a webhost. So I am an adherent.


The Cloud Isn’t all Clear Skies

But in all the euphoria to move data and applications to the cloud, aside from security, users and IT admins are overlooking some fundamental drawbacks to the cloud. The two most significant are (1) data location/backup and (2) integration.

Before cloud migrations, IT admins knew that all of the company’s data was on one or more specific servers, with specific geographic locations and specific applications and datastores. Admins had control over data storage. But with the cloud, that is all different, and there are tradeoffs with outsourcing data management. An enterprise that has fully adopted the public cloud will likely have email stored one place, accounting data in another, CRM data elsewhere and files somewhere else. Besides separate providers and privacy policies, the providers themselves will store and backup that data in locations that may be unknown to the owner of the data, even if a company is using dedicated servers. This might be okay with one application in the cloud, but as more data are moved there, the task of backing up and keeping that data private becomes increasingly challenging. Some larger enterprises are building their own private clouds which address some of this issue, but not all of it. And most of the market does not have the resources to build out a private cloud.

In the past, a solution was to stick with one vendor that could offer a suite of applications and a single datastore. But that is less desirable as those offerings now seem monolithic and outmoded when compared to younger SaaS companies that offer lighter-weight software. SMBs have also become accustomed to using best of breed in each category.


Integrate or Die

The other challenge I foresee is integration. In the old world, applications were tightly integrated. This was easy because data and applications were operating in the same environment if not the same piece of server hardware. There were also fewer vendors so the number of interfaces that needed to be written were limited, and business development energy could be concentrated. In a cloud environment and with an explosion of function-specific SaaS companies, integration is more complicated. And without integration, the usefulness of applications will always be limited.


How to Avoid Extinction

The solution to this has been for many companies to develop extensive open APIs. We see this trend accelerating and becoming a critical part of enterprises getting comfort with the cloud. Software vendors that adopt an antiquated protectionist approach to accessing application data and functionality will greatly impede adoption. Those that are thoughtful, open and aggressive in writing interfaces to other vendors will flourish. With the cloud, data are more portable than ever, and companies that build smooth abstracted from users will win: interfaces that don’t require a lot of configuration or data manipulation, can recover from data collisions and “just work”. Data transportability comes with a responsibility: security and privacy. It may seem that these are contradictions but allowing access to data along with a requirement that the data are protected is achievable. It is different than being completely closed and not allowing access to data at all (abstinence).

Cloud is an inevitability. As companies come to the realization that their data is now scattered and applications less integrated now that they are in the cloud, I believe it will give rise to what I have entitled the “era of integration”. In addition to great products, companies will win on data transportability and integration with other applications that are on par. This will not only be a critical competitive advantage for SaaS companies of all types, but it will also open up opportunity for companies that are assisting other companies achieve this.




Tweet about this on TwitterShare on LinkedInShare on FacebookShare on RedditEmail this to someone

Copyright © 2014 Jason Heltzer