To buy or to build, that is the question

Posted by Christophe Volard, Global Head of Product Development, BISAM on Jun 9, 2015 9:00:00 AM
Find me on:

buy or build

Do you have any idea of the number of apps that are available for smartphones and tablets? According to the statistics portal, as of May 2015 there were 1.5 million apps available to Android users and 1.4 million available to Applephiles. I am not shocked that the numbers are in the millions, but these are still astonishing numbers that really bring home the extent to which mobile devices have changed the way that we use application software in our everyday lives.

But it isn’t just the way that we use application software that has changed.  There has also been a revolution in the way that it is developed. Gone are the days when personal application software was developed by teams inside specialist companies with upgrades being released once a year. Today, apps can be developed and released by individuals with limited resources, with monthly or even weekly upgrades.

Thinking about these points led me to wonder what changes we are likely to see in the near future in the way that business applications are developed. What changes will we see in the form, structure, focus and organization of in-house IT departments? Will in-house IT teams become “app integrators” and in-house development a distant memory?

I don’t really think that this will happen, at least not in the near future. There are fundamental differences in the nature of business and personal applications which mean that custom-built software will remain important in the development of business applications for some time to come. But I do think that the old conundrum that is faced by IT whenever the business needs a new system, namely “to buy or to build” has been given a contemporary twist.

The pros and cons of buy vs build

In the world of smartphones and tablets, the most successful apps are the ones that provide good support for something that lots of people want to do. This is similar to the situation for business applications, where the most successful vendor systems are the ones that provide strong support for a market requirement. Many businesses, however, have substantial in-house IT resources, and so for business applications there is the question of whether to build applications or buy them.

The answer to this question will always be weighted more heavily in favor of “buy” whenever business processes are more standardized rather than more customized. To see why, look at the main generic pros and cons for the two options.

For build over buy:

  • Pro: the system contains functionality that is an exact match to the business requirements.
  • Con: it takes time and costs money to design, develop and implement a system from scratch, and so the time-to-market can be longer than, and the Total Cost of Ownership (TCO) greater than, for vendor systems.

And for buy over build:

  • Pro: vendor systems exist and are in production use, allowing references to be taken on how well they meet business needs; time-to-market and TCO are not just less than for custom-development, they are more predictable.
  • Con: the system may not contain functionality to meet all of the business requirements.

A recent research report suggested that asset management firms’ middle and back office systems will become entirely commoditized and outsourced in the future owing to two major shifts:

I)        Software as a Service: technology is becoming easier to provision and use.

II)       Emergence of new technologies: embracing new technologies is becoming essential to competitive success, and 3rd-party providers are well equipped to deliver new technologies, because they can spread R&D costs across a wide client base.

What is behind the predictions in this report is the belief that technology is enhancing the pros and diluting the cons for the ‘buy’ option. To see why, let’s look again at those generic pros and cons with a contemporary perspective.

The increasing adoption of web-based applications running on Cloud-based infrastructure is dramatically reducing the time and cost required to implement vendor systems. And system vendors can now release new functionality more frequently than in the past, which means that any gaps between system functionality and business need can be closed in less time. Both of these developments have further strengthened the case for buy over build.

Performance processes – are they standardized or customized?

But is it true to say that the business processes in performance teams are standardized? Don’t different performance teams use different methods to calculate returns and analytics? Doesn’t this mean that the business processes are more customized, and doesn’t it weaken the argument for buy over build in the specific case of performance systems?

Well, there are really two levels of business processes in performance teams. The first level can be thought of as micro business processes. These processes calculate performance and risk metrics. In the past, when calculation was characterized by a high degree of manual processing, one could argue that these processes were more customized than standardized. But these days, any good performance system will provide strong support for all calculation methods. So these micro processes are no longer customized business processes because the customization is now embedded within vendor systems. Business processing has been elevated to a higher level.

Which brings us to the second level of business processes, which can be thought of as macro business processes. These refer to the end-to-end processes extending from the collection of source data, through the validation and cleansing of that data, into the calculation of performance metrics and the checking of the results, all the way to the distribution of performance metrics to everyone that wants to see them. Details such as where source data is collected from, and where performance metrics are distributed to, will be different in different organizations. But the overall end-to-end business process is the same across organizations.

Good vendor systems contain features that simplify the task of integrating performance systems into the end-to-end system infrastructure. And remember, system integration isn’t a one-off activity. Change any of those upstream or downstream systems, and you have a re-integration challenge. Vendor systems are designed to be integrated with many different systems; in-house systems are much more likely to be more hard-wired into a specific system platform configuration. So the integration cost of vendor performance systems is lower than that of in-house systems when measured over the lifetime of the systems.

“The future ain’t what it used to be”

You may not agree with all the points made above, but I hope you will agree that the way we develop and deploy business applications in the future will be different to the way we’ve done it in the past.

The heading above is a quote from Yogi Berra, a very famous baseball player, who is also famous for a series of pithy quotes, containing paradoxes or contradictions. One of Yogi’s other famous quotes is “If you come to a fork in the road, take it.”

On the buy or build issue, I think we’ve come to a fork in the road. Which way do you think the industry will head? 

For a deeper dive into the Buy vs. Build discussion, please read our related  paper

BI-SAM offers a suite of sophisticated and flexible digital solutions for performance, attribution, risk, GIPS compliance and reporting – via our best-in-class B-One platform, and our new web-based BI-SAM GO solution. Complete our info request form to learn more.

Topics: Performance & Attribution

Leave a Comment

If you have any comments or have any questions about the subjects raised in this post, please fill in your details and message below. We'd like to hear from you.