A
Abhishek Tandon
Guest
It's a common question—how do you unlock your analytics potential? Should you build your own tools in-house, or invest in buying from a vendor? We analyze both options.
One of the most talked about topics with all of my customers is whether they should build an analytics application with an in-house team of data scientists working on open source technology, or whether they should buy a product which serves their purpose of creating a foundation or even better provide their industry specific solution for faster go-to-market. The clear bent is towards the build side as the general feeling is that they would have more control and hence a better chance of success.
To see which tactic makes more sense, let’s study the pros and cons of building an analytics application from ground up:
Build Pros
However, as rosy as it may sound, the build strategy is more often than not, marred by massive project delays, poor execution and very minimal output based on the amount of time spent.
The reason is simple. Analytics is still in its evolution phase. There is not one correct answer to the problem and there is definitely a dearth of talent which is causing more harm than good to organizations.
Build works best when you have the ability to spend tremendous amounts of man-hours in research and have the patience to wait for the right solution. Companies that have been largely successful in building in-house analytics solutions and capabilities are Google, Uber, Amazon, Facebook (you get the point here). Companies that are way ahead of the curve. Companies that rely on data analysis for their daily bread. It's core to their business and their existence.
Now lets’ look at the Buy option. Buying a product or solution has its own set of pros and cons.
Buy Pros
Skeptics may argue that products are not flexible enough for easy implementation and it can be hard to integrate with other applications. Thank god for Google, which created APIs that have made it very easy for products to talk to each other and ensure that there is no friction. The other very big advantage with buy is that the investment lock-in can be minimal till the time value is not derived. You are not wasting time trying to decide the best technology stack but are more focused on ensuring success of applicability.
Moreover, due to the nascency of the technology companies and plethora of VC money in this space, proof of concepts are provided by almost every provider. However, for the buy strategy to work, it requires a very defined mindset and team configuration which I will take up in my subsequent blogs.
In summary, stick to your core business and take advantage of the research dollars put in by others. As technologists it's easy to get carried away with the excitement and buzzwords around analytics. But my suggestion would be: “Don’t Try this at Home”!
Continue reading...
One of the most talked about topics with all of my customers is whether they should build an analytics application with an in-house team of data scientists working on open source technology, or whether they should buy a product which serves their purpose of creating a foundation or even better provide their industry specific solution for faster go-to-market. The clear bent is towards the build side as the general feeling is that they would have more control and hence a better chance of success.
To see which tactic makes more sense, let’s study the pros and cons of building an analytics application from ground up:
Build Pros
- More control, so you can constantly track how you are doing against your goals
- Possibly cost effective assuming the team is working/experimenting on open source technologies
- SMEs can interact with the technologists on a regular basis to ensure that the final outcome is in line with what they need
- There is no restriction on the product features and hence, the extent of build can be anything
However, as rosy as it may sound, the build strategy is more often than not, marred by massive project delays, poor execution and very minimal output based on the amount of time spent.
The reason is simple. Analytics is still in its evolution phase. There is not one correct answer to the problem and there is definitely a dearth of talent which is causing more harm than good to organizations.
Build works best when you have the ability to spend tremendous amounts of man-hours in research and have the patience to wait for the right solution. Companies that have been largely successful in building in-house analytics solutions and capabilities are Google, Uber, Amazon, Facebook (you get the point here). Companies that are way ahead of the curve. Companies that rely on data analysis for their daily bread. It's core to their business and their existence.
Now lets’ look at the Buy option. Buying a product or solution has its own set of pros and cons.
Buy Pros
- Its definitely a faster GTM strategy—much of the stack foundation is already there
- There is already significant R&D that the analytics product companies have put in, which you can reap the benefits of
- The onus of success is as much on the product as it is on you. Hence, you have a very dedicated team working with you towards your success.
Skeptics may argue that products are not flexible enough for easy implementation and it can be hard to integrate with other applications. Thank god for Google, which created APIs that have made it very easy for products to talk to each other and ensure that there is no friction. The other very big advantage with buy is that the investment lock-in can be minimal till the time value is not derived. You are not wasting time trying to decide the best technology stack but are more focused on ensuring success of applicability.
Moreover, due to the nascency of the technology companies and plethora of VC money in this space, proof of concepts are provided by almost every provider. However, for the buy strategy to work, it requires a very defined mindset and team configuration which I will take up in my subsequent blogs.
In summary, stick to your core business and take advantage of the research dollars put in by others. As technologists it's easy to get carried away with the excitement and buzzwords around analytics. But my suggestion would be: “Don’t Try this at Home”!
Continue reading...