Unlike in the Chicken and Egg story... Should not the funding come first? What's the best approach? Can you go meet Angels and VCs with your blueprints? How can that really work in the real world?
You have not provided enough information for your question to be answered. If your "MVP" will cost 350K, then it's likely not an MVP. An MVP is something -- anything -- that you can put in front of people that can be used to secure feedback to make the 'next' decision. If you are having to spend 350K for whatever "it" is (this is what you should have told us), and you don't know how to secure funding, then you probably took a few feedback loop shortcuts. Best to take a few steps back and rethink whatever it is.
Firstly, £350k isn't an MVP. You mentioned AI in one of your posts - which is actually incredibly easy to 'MVP'. I've done it with quite a few startups, both with and without cash. The first iteration of an MVP should be the concierge - doing EVERYTHING manually so you get to know and understand how your system works and how users flow/interact with it. Based on that principle, your AI MVP should be manual - convince your user it is AI in the backend, but in reality it can be someone sat on a computer typing out answers. It's what A LOT of companies do which claim to be AI, Bot etc. This method can also go towards helping your AI to learn!!
Have you looked for ways to build a prototype firstfirst?
I suggest you seriously consider getting a tech partner and build things in house. Plus outsourcing AI is weird. From your responses I see you are doing an AI product. As someone involved with a lot of AI products, the AI (ML model) part is likely fit in any of these categories none of which is sensible to completely outsource:
1. I have a bunch of data, want to use it to predict stuff? This might be out-sourceable but usually a smart engineer can figure this out using online services.
2. I have a problem, no data. Then you need to figure out what data is required and collect it and improve it, etc. etc. Needs cash but is a hard sell. Unless you can somehow figure out a bunch of data and turn this into option 1
3. A brand new algorithm not available in any cloud service out of the box. For example you are creating an AI that helps a robot walk. This is RnD work and cannot be outsourced. You need a bunch of PhDs dedicated to the job.
4. I want to do a common AI work such as detecting common objects in images, text to speach, etc. There are cloud services for this and you only need programming skills
I'd start by securing solid pre-sales/commitment from potential buyers before even raising money.
And then, I'll brainstorm as much as possible with my team to find creative ways to reduce this amount by at least 1/3.
I am sure you can cut/delay some of the expenses or find work arounds.
But as we don't have much details about your product, that's all I can say.
I initially priced my product MVP 35K but this was in a perfect world and without considering the fact that an MVP shouldn't be perfect but functional.
We ended up using elbow grease and it works totally fine!
The MVP should be not more than $50k. I launched recently my MVP. I found the early adopters for my product. I looked what is the minimum needs was needed. Go with "keeping it sweet and simple". I am raising the funds and in my final stage. The person you need the most at the time of MVP is a Product Manager. It's very difficult as a business leader to do that but Product managers can guide you. Good Luck!!
Meet with clients first. That's a perfectly reasonable figure if there's some traction behind it, and a list of clients willing to buy into the MVP (pre-sales) go a long way towards securing the funding you need.
Depending on your industry, many clients will themselves be viable angels. I've received more investment funding from people with a viable *use* for my products than from anyone looking to share profits. And unlike many financial institutions, clients are highly willing to be one-of-many investors.
In the real world, it is very difficult to raise money without some traction. I have raised money pre-MVP from angels. In order to do this, you need to have a solid idea with a compelling pitch.
Hi Alain, in response to your secondary post. I have failed forward in this regard and can report that as a non-coder I went to overseas development for the best price. Project Management had its challenges but I was able to work effectively - until the developer identified he under quoted my fully wire-framed and spec'ed out project and would not be able to finish the remainder as I had presented. I had to stop the project - one of my lowest points. Didn't lose an astronomical amount but nonetheless the cash could certainly have been used in other areas.
Now, I'm a non-coder but technical enough to use a no-code platform so I built the complete MVP from the ground up and automated as much of the back end as I could - its now functional and I'm able to gather the much needed feedback. What's the saying - "Fake it until you make it"? Its amazing what you can do when you're out of money :)
We simply need to guide our fellow entrepreneurs towards getting that feedback - we cannot "donate" that kind of cash without validation. I'd be happy to help anyone starting along this path :)
@Gareth SBIR requires 1) distinct need or area of interest by sponsor agencies as follows (copied from the NSF website):
Biological Sciences (BIO) Computer and Information Science and Engineering (CISE) Education and Human Resources (EHR) Engineering (ENG) Environmental Research and Education (ERE) Geosciences (GEO) Integrative Activities (OIA) International Science and Engineering (OISE) Mathematical and Physical Sciences (MPS) Social, Behavioral and Economic Sciences (SBE)
as a way to de-risk scientific and technological agendas and development, so a consumer app (I infer it is consumer driven due to the "1M-download app for daily sales-generating usage" comment) would likely not be entertained, and 2) requires majority ownership of the company to be by US Permanent Residents or citizens. Alain is in Mauritius so would likely not qualify.