The best way (in my experience) is try to deliver the value but in a old-school way (not software related). Sometimes this is not possible, so you want to try to get as close as possible to what would be your product.
Example: I wanted to know if people would pay for software that would help them curate and manage their social media feed but get it deliver by email (for 40+ audience). I curated Twitter, LinkedIn and other social network where I could get graphs and started and email list. I got feedback that it was very useful and then I ask them if they would pay for it and how much.
My advice is: find the real value that your SaaS will deliver, and try to do it manually in small scale. Then try to get paid for it. If you get customers, then you may build something to serve them.
**sometimes this is hard to do if the problem you are solving is very complex. In this case you should pitch the solution to possible clients and see if there is enough interest for pre-sales or if they commit to a pilot