I was just about to make the same point as Inquisitive Mind above. If your idea is truely unique, it's the patent that could potentially hold (and protect) the value of your idea. If this isn't watertight, then all of the time, money and effort you spend in setting up the business, could all count for nout if it's easy for people to bypass.
I'm in the same boat as you (I'm also from the South East too).
I'm just starting out, and had the exact same questions as you a few months back. I've put together some of the info I've learnt so far (some of it I've learnt the hard way already!) which I hope may help:
1) You dont need to get an NDA signed by a Patent Attourney (just make sure they're part of CIPA first ;-) ) as they're governed by certain rules and charters.
However, just for your own peace of mind you may want to just follow up after the meeting via email with something like "
Hi John, thanks for the meeting on Friday. I found it really helpful and I'll be in touch again soon", etc etc
- that way at least, you have something to confirm you had a meeting with an employee
just in case you ever needed it. (hopefully you never will, but it's better to have something you never need, than need it and not have it

)
2) Get
anyone else you chat to to sign an NDA. There are loads of templates out there so what I ended up doing was reviewing about 10, and then choosing the elements I felt I needed. You can also get them done by a solicitor, but that's obviously going to cost you - you need to decide if it could be false economy not to have a watertight one though.
If you're dealing directly with a non-UK based company, that may be worth the cost.
So, the first thing I would say is have a chat with a patent attorney using their "Free consultations" which they offer - they can help give you an idea of the IP costs for your specific product, and they may also sometimes do a very quick preliminary search to see if anything jumps out right way as being a potential issue.
THEN (before you commission them to do a patent,full patent search, etc for you), I would take the advice offered above and do some initial user research.
There are lots of good online survey tools out there (survey monkey, etc) but all of these tend to either have limits, or charge. I did some digging and found a really good survey company which enabled me to do a free survey which didn't have any limit on the number of questions. (Plus, I needed some help setting up some aspects of the questionare and they were brilliant!)
I dont know if I'm allowed to link to them, but if you Google "Sensor Pro" you should be able to find it.
3) Create an online survey - make sure you don't use "leading" questions which may bias the output. This can be tricky so you really need to spend some time on this.
I trialled this inititally just on some friends and family and got some really interesting results (some of it was contrary to what I was expecting to get!)
After you've trialled this across a small sample (say, about 20-30 people), review the results and see if anything needs tweaking (i.e. how many people didn't complete the survey? did they all understand your questions or did a lot of people choose a generic "dont know" answer - if they did, you may need to retweak those questions)
Once your happpy with it - use the power of social networking to push it out to a wider audience.
Oh, and if you're getting all these people to fill in the questionaire, you may as well take the opportunity to ask them if they're happy for you to email them again in the future (
dont make it compulory!!) - for those that do, you're already starting to build up a bit of a potential costumer base!)
4) Other market research - surf, surf and surf the web some more.
I've managed to find a mass of really interesting stats, etc which not only help prove the need for our product (which can help if you need to secure funding), but also, loads of additional information regarding user groups (which helps with building up an even better picture of your market) - and all of it for free - it's just costs your time and effort
5) Don't give everyone the full picture too early.
Again, I may have learnt this one the hard way. I tend to trust everyone; it's a bit of a floor of mine.
Long story short,
even if you have an NDA in place, only provide people with enough information to help them do whatevery it is you're asking them to do. Try to hold off giving them "the bigger picture" (i.e. specs for additional future product lines, etc) until you've established a really good relationship with them.
6) Product development - Get
all costs in writing no matter how much you trust the person you're speaking with - This includes your per unit costs if you have a specific value you need to hit.
I've just
very nearly found this one out the hard way!
Oh, and I've heard that China isn't necessarily as cheap as it used to be, so don't be too heavily focused on getting it made in China. Do a global search for companies who will be able to help
create what you need, ask for some estimates/quotes (as applicable) and
then see what comes out of that. Chances are, it could still be a company in China, but at least you haven't ruled out a potentially cheaper/better supplier purely based on location.
I think that's about it for now, but I hope it helps. Good luck!