- July 25, 2016

Note: this article was originally published in InfoWorld Data Insights.


Today’s startups are all about bringing to market a great idea, as quickly as possible, and scaling their business to gain adoption and market share before their idea gets replicated. And today, there is less margin for error than ever.

Speed and agility

In an increasingly open and fluid world, it is increasingly difficult to keep a secret. Ideas — the primary currency of startup nations — are free flowing. When you, wannabe or confirmed entrepreneur, get a brilliant idea, the time window to act upon this idea is short. It is therefore critical that you are able to bring to market a first iteration of this magic app, innovative platform or new service, in order not only to put a stake in the ground but also to start measuring user reaction and to collect feedback.

Acting upon this feedback is the other side of the coin: users and consumers are now used to getting their voice heard, and a startup that is paying no attention to its users, that is not acting on what it’s been told, can anticipate a short lifespan.

Engineers are expensive

Especially in startup land such as the Silicon Valley, London, Berlin, Paris, Tel Aviv and more, good engineers — the ones who have what it takes to work in a startup — are sparse, and expensive. You should make the best use of their time, and first and foremost they should not be asked to reinvent the wheel.

Basically, when thinking “engineer”, don’t think “developer” but “architect”.

Need a rock solid platform

First impressions are oftentimes the only impression that stays. An app or service that does not work as advertised, that does not scale under load, is a guaranteed recipe for loosing customers and failing to grow. Simply put, you can’t afford its solution to break, you won’t be given the benefit of the doubt but rather immediately disqualified.

Basing the software stack on reliable, proven architecture and infrastructure is the first step toward success.


In the world of platforms, you should maximize the reuse of component and services that others have built, and are making available for you to use in your own solution.

Needless to say, the platformization trend starts with IT infrastructure. Modern applications run in the cloud, and are accessed in the browser. But cloud platforms are available that provide more than just computing power: the range of Platform as a Service is broad, from databases to business process management to Application Platform as a Service (aPaaS) to API Platform as a Service (apiPaaS).

Many services can also be plugged into a solution. Payment services, accessible through APIs, are a great example. Embeddable business intelligence and reporting tools also exist. Communication gateways (voice and text) are only an API call away.

Let others build it for you

As a startup, you need to focus on your core competency. And that core competency is to provide a service, to deploy on app and gain users to become a standard, before anybody else copies your idea.

So, instead of reinventing the wheel, make maximum use of tools and services available outside the organization. Do not go and negotiate with banks and credit card companies: use a payment provider that has already does this for you through APIs. Do not redevelop a texting interface: user the services of a developer-friendly company that let you build this capability inside your app. Do not build reporting tools, embed existing reporting and analytics components. The examples are numerous.

In addition, these services and the APIs you use to leverage them, will already be field-tested, scalable for thousands or millions of concurrent users, and will continuously evolve to support the latest standards and regulations.

At the end of the day, here is what the day of a startup software architect should look like: 0% coding, 100% integrating.