Selecting the right tool and vendor is wrought with difficulties, misinformation, over zealous salespeople, and downright confusion within your ranks. Add to this the many facets of office politics on tool selection and vendors, and you have a potential disaster on your hands. The journey of tool selection does not have to be a disaster of epic proportions; however, you must be prepared in advance if you want to be successful. The tool you select will be with you for many years to come, so, unless you have deep pockets, there are no second chances.

1. What Are We Trying to Accomplish?
While this should be obvious, different groups within your organization may have entirely different ideas as to what is needed. You may want the group to be at CMMi level 3, whereas the development group simply needs a good versioning tool just to keep your code straight. So the main goal should be to fix the problem at hand, but be aware that this tool must be scalable for future needs.

2. What Features do we Want this Tool to Have?
This is where collaboration amongst the varying groups in your organization is paramount. I advise getting the groups together separately and ranking your needs on a three-tier scale: must have, nice to have, and not essential. The must haves are not negotiable; if the tool doesn’t provide it, then you drop that tool. Nice to haves are things that would be great but they're not a deal breaker. The not essentials are just that, nice if you get it, but falls at the bottom of the list.

Once the groups write out their lists, representatives from each group should get together and compile them into one comprehensive list. This should be approved by all groups once it is compiled and then get buy in from management. This list is probably the most important part of the process and must be adhered to in the tool selection process.

3. Do We Go with a COTS Tool or Freeware?
I think this all depends on what your compilation of must haves and nice to haves is. If freeware has everything you need and vendor support is not as important, then go with freeware. But also be aware that many of the freeware tools have vendors that will support them and they can be modified by your development teams to fit your needs. There are tradeoffs no matter which route you take. With COTS you are stuck with a vendor for many years, and this relationship is not always rosy. With freeware you can move from one tool to another much easier.

4. How Many Tools Should We Look At?
This is where the list compiled by the groups comes in. I recommend that you do some prescreening of tools by visiting their websites and seeing what tools meet your criteria. The last thing you need is fifteen vendors coming onsite. While there is no magic number, I would say no more than five vendors should be selected. This will keep it simple because after viewing this many, you have a lot of information to consider. I also recommend not having all five vendors come in one right after another. In other words, keep their visits separated by a few days. This will allow the people who go to the demos to sort out what they have seen and you reduce the risk of getting things confused. Trust me on this one, if you select five vendors and have them come in five days in a row, things will bleed together and you will forget who said what.

5. What Do I Do after the First Demo?
You need to write down your likes and dislikes of the demo. Did it meet the criteria you set forth at the beginning? In other words, how many things did you check off of your list of must haves? This leads to a good point, you can provide your vendor with your list beforehand, if you desire, so they can focus their demo to this and you can make sure that their product delivers on your needs. Make sure after the demo that you get the group together as soon as possible to discuss individual findings while the demo is fresh in everyone’s mind.

6. What's Next after all of the Demos?
Now is where it gets tricky. You have seen all of the demos and heard all of the sales pitches and a decision must be made as to who moves forward. Again, this is where the list you compiled plays an important role. The next step is to rank the five vendors on how they delivered on your list of must haves and nice to haves. If they met all of your criteria, then they should pass to the next step. I would narrow the field down to two or three vendors. Then have the vendors come back for a more detailed look at their toolset.

7. Should We Mix and Match Vendor Tools?
This question may come up. You like Vendor A’s versioning tool and Vendor B’s issue management tool and Vendor C’s requirements tool. What do you do with this situation? Well, be careful with this scenario. Vendor A may say, “Oh, yeah, our tool integrates with the other vendor's tool.” Find out specifically what they mean by “integrate.” Simply having pointers from one tool to another via a link of some sort is not integration. At minimum, integration should mean you can operate and perform actions in one tool and never have to open the other.

8. Can We Trust Our Salesperson?
This is wrought with controversy, but must be mentioned. I would like to think everyone can be trusted no matter what. However, remember one thing when dealing with salespeople, their job is to sell you product, period. No matter how nice they are, the meals they purchase, or the way they treat you, they are there to close the deal. Do not let them sell you a sports car simply to drive one mile when a sub-compact will do the trick. This is not to say that all salespeople are not sincere, they do want a long-term relationship with you, they do want you to succeed, they do want you to achieve all of your goals, because this means more and more potential sales.  If you are not sure, ask them one question, “Can we get this in writing.”

9. Do We Need a Proof of Concept?
Absolutely! Do not buy a tool unless you have the vendor prove that their tool will work in your environment. If they are not willing to install on your site or cannot otherwise prove it works with your hardware, end all talks and discussion there. The last thing you want to have happen is for them to come onsite after the purchase only to find out that their tool doesn’t work with your particular OS or hardware. By then it's too late, the money has been spent and you are stuck with two choices, shelve the tools or upgrade your environment. I have seen way too many tools end up being bookends on shelves.

All of the questions I have presented here should be asked, and there are many more that you may come up with. Remember, the tool you purchase will be with you a long time. Your reputation and your very job may depend on its successful implementation, so make sure you choose wisely. The main thing is, if you ask the right questions and get proof of everything that is said, your journey to the right tool and vendor will be much smoother.