I started a B-2-B business just under a year ago and I took a few different approaches, some of which follow the lean start up method... The key to the whole thing is to know what you need to validate at each stage.
1. Test it manually to verify that there is acceptance of your product idea. Depending on the idea, you may be able to replicate the entire idea with a website 'facade' and manual back end. Your service time might be a bit slower, but you can set expectations for that. With my business, I wanted to test whether small businesses 'got' the process methodologies I was planning to build into a cloud software. So I tested them as a consultant and contracted with two organizations who would pay me to facilitate them through the process. At the end, I'd know if the process resonated or not and how I might need to tweak it.
2. Figure out the ideal (but not overdesigned/built) product and create a clickable prototype of that. (This can be done very cheaply and I can tell you more about that process if you want.) Then do "demos" for people in your network who fit the target audience. With a clickable prototype you can make it as sketchy or as finished looking as you like. But either way, in less than 30 minutes you can essentially demo the product and get initial feedback - does this resonate, is this something you'd consider using and why/not? I got great feedback this way. And there is no "ask" other than for up to 30 min of your time. Plus, you'll start to build up a list of people who are excited to learn more as you get to the next phase.
3. Build the skinniest MVP you can that will get you validation and provide value to the customer. Then pitch the hell out of it to anyone who will listen. Everyone in your network knows people who might be in the right business/industry for your product. I'd tell everyone what I'm working on and say that I'm looking for beta clients to partner with me in testing the first version. I was surprised by how many people made introductions to other people who they thought would benefit from my product. Some of those went nowhere, but I ended up securing 6 organizations to participate in my beta. In exchange for using the product for 6 weeks and providing specific feedback along the way, I gave them a personal onboarding, rolled out features they requested (that I agreed with) every week, and allowed them to use the product for free forever. I pitched (via web screen sharing) to about 15 orgs in order to get 6.
Thats as far as I've gotten, but its worked really well for me thus far.