Over the course of the last couple of articles we have been speaking about the different types of data integration strategies that are available. One of the consistent questions that we receive as feedback is, "How do we determine which method is right for our enterprise?" It is safe to say that there is no such thing as a one size fits all answer. Even within the same enterprise, varying groups may have needs that preclude one tool being right for every situation.
There is absolutely nothing wrong with having more than one integration solution. The only thing that an enterprise needs to have is one integration strategy. There is a significant difference between having one tool and one strategy. Any tool is exactly that: a tool. The issues get cloudy when departments within an enterprise begin to try and establish solutions because what happens is that every group begins to make selections. This is when the solution becomes expensive because there is no oversight insuring that duplicate functionality isn't creeping into the enterprise. Like anything else, tools are a significant part of the IT infrastructure and, as such, need to be treated with the respect and care you would give any other company asset.
When an enterprise establishes a clear-cut partnership between the business needs of the organization and IT, then that oversight function is filled by IT. Business needs to communicate clearly what is required, and then IT needs to judge what the best solution is based on the specific needs of each business case. This scenario is not meant to imply that business is at IT's mercy. There are issues with the use of each tool, and it is clearly the responsibility of the business users to articulate completely what their needs are. It is perfectly acceptable for one of the conditions to be "ease of the tools use" or any other particulars that the business users feel are pertinent to fulfilling their mission. IT must decipher the needs and translate them into technical requirements. This integration of the business needs with the technical requirements is a vital piece, ensuring that everyone has what they need to get their jobs done.
From last month's column, The Advent of E3," we want to review the table that has the pros and cons of tool selection.
Figure 1: Tool Selection Guidelines
As can be seen from Figure 1, perhaps the best way to approach the tool selection is to make a list of the business requirements and the technical capabilities of the enterprise infrastructure. Once you have a clear and concise idea of what you are trying to do and what your IT organization can support, it becomes a matter of looking at the pros and cons listed in Figure 1 and finding the appropriate fit. In some cases, even though a tool may not be an exact fit, taking the organizational infrastructure into account may yield a tool that will satisfy many, but not necessarily all, of the requirements. This then becomes an issue of filling the gap. There is a portion of the selection process which demands that common sense and finances be taken into account. Selecting a tool that fits 100 percent of the requirements but costs more than the entire IT budget to install, forces the issue of practicality. Anything that breaks the budget or fills none of the requirements is a bad investment, and it is up to the partnership between business and IT to insure that the business requirements are met with the practicality of budget constraints.
The constraints involve not only the initial purchase costs but should also take into account things such as supportability - hardware, software and personnel - and ease of use. Selecting a tool that requires the end users to undergo substantial retraining is another issue that must be addressed. In addition, an enterprise must also look at the first line of support - be it a help desk or a designated individual or group. Training and first line support can be extremely costly, and these costs are sometimes totally ignored in the decision process. The costs are either totally overlooked or minimized until the first time it comes up.
In summary, selecting integration tools is only a small part of the enterprise's overall integration strategy. The strategy must take into account 1) first and foremost, the current IT infrastructure and what it is capable of absorbing and supporting; 2) the enterprise's short-term (3 to 6 month) needs and its long-term (1 to 2 year) needs; 3) the support issues that the new tools introduce; and 4) the costs of the new tools in terms of software, hardware and support personnel. As you can see the selection is not something that should be taken lightly. Only by insuring the partnership between business and IT will everyone understand everything involved, and the enterprise will be assured that the end solution was correct for the company.