In yesterday’s article, we discussed 7 models for achieving startup-like efficiency on large OSS transformations.
One popular approach is to build a proof-of-concept or sandpit quickly on cloud hosting or in lab environments. It’s fast for a number of reasons including reduced number of approvals, faster activation of infrastructure, reduced safety checks (eg security, privacy, etc), minimised integration with legacy systems and many other reasons. The cloud hosting business model is thriving for all of these reasons.
However, it’s one thing to speed up development of an OSS PoC and another entirely to speed up deployment to a PROD environment. As soon as you wish to absorb the PoC-proven solution back into PROD, all the items listed above (eg security sign-offs) come back into play. Something that took days/weeks to stand up in PoC now takes months to productionise.
Have you noticed that the safety checks currently being used were often defined for the old world? They often aren’t designed with transition from cloud to PROD in mind. Similarly, the culture of design cross-checks and approvals can also be re-framed (especially when the end-to-end solution crosses multiple different business units). Lastly, and way outside my locus of competence, is in re-visiting security / privacy / deployment / etc models to facilitate easier transition.
One consideration to make is just how much absorption is required. For example, there are examples of services being delivered to the large entity’s subscribers by a smaller, external entity. The large entity then just “clips-the-ticket,” gaining a revenue stream with limited involvement. But the more common (and much more challenging) absorption model is for the partner to fold the solution back into the large entity’s full OSS/BSS stack.
So let’s consider your opportunity in terms of the absorption continuum that ranges between:
clip-the-ticket (minimally absorbed) <-----------|-----------> folded-in (fully absorbed)
Perhaps it’s feasible for your opportunity to fit somewhere in between (partially absorbed)? Perhaps part of that answer resides in the cloud model you decide to use (public, private, hybrid, cloud-managed private cloud) as well as the partnership model?
Modularity and reduced complexity (eg integrations) are also a factor to consider (as always).
I haven’t seen an ideal response to the absorption challenge yet, but I believe the solution lies in re-framing corporate culture and technology stacks. We’ll look at that in more detail tomorrow.
How about you? Have you or your organisation managed to speed up your transition from PoC to PROD? What techniques have you found to be successful?Read the Passionate About OSS Blog for more or Subscribe to the Passionate About OSS Blog by Email