Here's the thing: when there are a lot of services that do something similar, I sign up to the free trials of lots of them at once to evaluate them. If you require a credit card, I probably won't include your program in the initial evaluation.

So yes, if you require a credit card, you're getting fewer people with a higher likelihood of conversion, but you're also missing out on a lot of people who are doing comparisons of services. How many of those people you convert is likely to be a function of how good you are compared to the other options.

If you are offering something unique or have amazing word of mouth, this probably won't matter. If on the other hand you think your product is one of the best solutions to the problem out there, you're probably losing more customers than you want by demanding credit card details.

In the case of the OP, doing something that on the face of it anyone who needed that service could hack together a script to do roughly the same thing within a relatively short period of time, I would have thought that it would be very important to remove as many barriers to trying the thing as possible (but yes, there's no need to let the free user have access to checking more than 2 sites).