Today, more than ever, a wide range of complex business challenges drive the need to unify information and make it readily available. This, in turn, impacts what is needed from your integration engine.
Selecting a data transfer systems involves identifying and quantifying strategic requirements (summarized in our Blog Post “Six Easy Steps to Identify Data Integration Needs”), and then matching them to integration options. As there are, however, literally hundreds of integration tools to choose from, this Post suggests ways to home in on the types of integration systems that are likely to meet your business imperatives.
The first step is to look at the various categories of integration solutions. There are many ways to sort different systems – for instance ETL versus EAI versus data visualization – however small to medium sized companies, which make up around 99% of businesses and up to half of private sector employment (in the US, at least), can benefit from a more pragmatic sorting, as suggested in the table below.
Select the Category that Best Fits Your Needs
Once you have identified categories and their key attributes, match these to the list of “must-haves” identified from a review of your business requirements. It is best to keep to high-level objectives at this point, as demonstrated in the example below:
The trap to avoid is simplifying needs. For instance, where development time needs to be short, the project is a one-off, or performance monitoring is not important, point and click solutions can be a great bet. On the other hand, if the project is a work in progress and requirements could change suddenly, or if volumes are expected to increase significantly, a point and click may be a bad choice. In other words, don’t be short-sighted. Look beyond the immediate requirements to avoid wasting time and money on an inappropriate approach.
An expansive view is also helpful when reviewing pricing options. Across and within all categories, there can be large differences between initial and ongoing fees. For instance, adding users, increasing volumes and building in more automations/jobs (a complex project can have thousands of these), can all lead to a dramatic increase in cost. As an illustration, the graph below shows how some typical pricing models are related to volume increases.
Too much, too little, or just right?
There are two big problems when selecting an integration approach. The first is to take the process too seriously, the second is not taking it seriously enough. In other words, the evaluation process can become bogged down in detail, or not look beyond the immediate project requirements. These are easy traps to fall into given today’s data complexity and the allure of “quick-fix” tools, however keeping sight of key business desiderata, and making sure these are met, will help identify the integration approaches needed. Done correctly, the final match of requirements against options will result in an approach that helps springboard your company to ongoing success.