When deciding on the analytics software that will best enable your organization to use its data to its full potential, you have a choice - do you go for a custom or an off-the-shelf solution? This is a tremendously difficult decision which requires a detailed knowledge of both your organization and data needs. It is further complicated by the marketing hype that surrounds different tools, with all of them claiming to be the best. This needs to be cut through first. Ultimately, it comes down to the solution that is most suitable to your organization's needs, structure, and the talent you have available.
Off-the shelf analytics solutions are pre-designed analytics tools that solve a well-defined and understood business process. They work to solve problems common to businesses, such as customer abandonment. An off-the shelf solution will work out for you what data is needed so you don’t have to sit down and try to figure it out yourself, looking at those who have sought the same understanding and establishing the metrics that worked for them.
The benefits are many and persuasive, particularly for smaller companies who may not want to spend the time and resources developing a custom solution. Off-the-shelf solutions have accelerated delivery at lower cost. While finding the right product may be a long process, it is nowhere near as long as developing one yourself, with off-the-shelf software usually available in weeks, compared to months of development for a custom tool. A custom tool will also require continuous business engagement and commercial support to avoid detachment, which means that resources are drained even after it is implemented.
A custom tool also requires data scientists to build it. More so than this, it will require a data scientist capable of building a truly effective tool, and there are not many of these about. It then requires ongoing business engagement, which either means hiring data scientists with pre-existing knowledge of the business, who will be even harder to come by, or forcing business users to work closely with the data scientist, which is time-consuming and diverts resources away from where it is needed.
Off-the-shelf solutions can also help drive data democratization, putting analytics in the hands of business users rather than relying on a single data scientist or team. Instead, they can ask questions of the data on the fly. Reliance on an individual data scientist or team is slow, and should they leave you could have a major problem. The tools are usually incredibly easy to use too, which means that even someone with no experience in data analysis can use them.
There are, however, a number problems with off-the-shelf analytics that mean it may not be suited to you. Firstly, as you grow, there will always come a time when your organization outgrows the tool and needs to scale up. This means having to look for a new tool that can cope, a problem that is likely to re-occur as your needs evolve. With a custom tool, it can be adapted far more easily. With an off-the-shelf tool, you are also relieved of a significant degree of independence around your data and forced to rely on your vendor’s custom algorithms, features, and integrations. With custom analytics tools, you retain full ownership of your data, which is an especially important consideration given the climate around data privacy.
It is true to say that every business is unique, but most are simply not so unique as to require a bespoke analytics solution, and, given the outlay in terms of resources and investment, custom software often won’t provide the necessary return on investment. There are usually preconfigured tools that can solve the key problems you need solving for maximum business impact. At the same time though, your business may need a custom analytics tool if your business model is too complex for something off-the-shelf. You have to ask yourself a number of questions before making the decision and be realistic about your likely speed of growth. Is your company at a level of maturity with its data that it needs a significantly more advanced analytics tool? Does your organization have the resources to see through development and maintenance of a custom tool? Is there any actual business problem that cannot be solved with your current tools? If the answer is ‘no’ to these questions, ultimately, you probably don’t need custom analytics.