I’m often asked to share advice about raising the first few rounds of capital for a young startup.  I’m actually on my way to an accelerator right now where I was asked to lead a few sessions on this topic.

There is a ton out there on the internet about the tactics of raising money.  Paul Graham wrote a really comprehensive essay a few weeks ago that mirrors most of the advice I tend to give.  Read it here.

But my over-arching observation is that fundraising is extremely unpredictable.

This is pretty puzzling to most people.  One would think that fundraising could be a pretty rational process.  Pick the 5-8 firms that seem to be the best fit, get a good introduction, focus on the 2-4 that really dig in, and go from there.

I’ve even seen series A firms give this general advice to their portfolio companies.  “Run a small process” and see what happens.  The thought is you can always talk to more folks later.

From my perspective, this is typically pretty bad advice.  The conversion funnel above is accurate, but only mid-way through your funnel.  The top of the funnel needs to be much larger to have an effective fundraise.

This is because there are so many reasons why VC’s can pass on an investment, and it’s pretty impossible to really account for all the variables prospectively.  Here are some examples:

• No experience in the space
• Bad experience in the space
• Partner had a bad experience in the space
• No comparable companies have been acquired recently
• No comparable companies have gone public recently
• Comparable company had an IPO that tanked
• Consumer is out of favor
• Enterprise is out of favor
• Too capital efficient
• Too capital intensive
• Great early traction, but I don’t see how it gets big
• Big idea, but you should start with something smaller to prove traction
• Great traction, but engagement is bad
• Engagement is great, but it doesn’t matter without traction
• The team is unknown
• The team is too well known
• Partner just did a deal
• Partner is dealing with tons of issues with their portfolio
• Partner has personal issues
• Partner is about to join another firm
• Annual meeting prep is taking too much time
• Fundraising is taking too much time
• There is some wierd conflict that is impossible to uncover ahead of time
• Wife thinks the idea is bad
• etc etc etc

This is just a subset.  But there are many more.  Maybe when a company is further along, the realistic set of investors is smaller, and a relationship has been built over time that allows for a narrow scope.  But early on, in the seed, Series A, and even the Series B round, I think it behooves founders to assume a 50% factor of unpredictability, and double the number of groups they plan to talk to.  Then as Paul Graham says, run a breadth-first search and parallel process.  Ideally, you’d like to get to a point where you have 6-8 potential leads really digging in mid-way through your funnel.  So get there, you need to have 2X the number of really quality conversations in the beginning, which probably means that you need another 1.5X the number of potential investors before a bunch of firms get disqualified for one reason or another.

There may the concern that your round seems “shopped” and can get stale if everyone knows you’ve been raising money for a while.  This is the reason to do things simultaneously rather than sequentially.  It’s very natural to think “I’ll focus time on my 4 best prospects, then if those don’t look good I’ll move on to my next 4, then so on.  Don’t do this.  You have to parallel process.

I now that sounds like a lot. It is – and you will feel like a maniac during this process.  But if you are honest with yourself about how investors are receiving your story and triage the funnel appropriately, you can accurately and quickly hone it down to the smaller subset of serious prospects pretty quickly.  But it is almost always beneficial to put in the shoe leather to do this because I guarantee that one of those serious prospects will be a group you didn’t expect to be in the mix when you started fundraising.