Get the newsletter · 2018 essays (PDF) · Featured · Recent

How to fool VCs into thinking you have traction, Part 4

OK Go with point #3
Finally, here’s point 3 of a multi-post series on the question, "How can you fool VCs into thinking you have traction?" Here are links to older posts:

"Standards" and metrics like date ranges
Today, we’re going to talk about manipulating standards and metrics – in fact, let me roll in hits versus pageviews and ad impressions, since it’s all part of a similar issue. Basically this sleight of hand has to do with the fact that people treat $0.99 as much cheaper than $1 when they are both only a cent off. So if you pick the right metrics to measure yourself by, you can get a big boost in credibility.

Here’s what you say
When someone asks you what kind of traction you’re seeing, say something like this:

"Since we started public beta, we’ve served over 300 million ad impressions on the site"

I also love "peak" numbers or "run rate" info:

"We currently have a peak run-rate of 240 million pageviews"

This sounds great right? It’s actually more complicated than that.

What’s weird with these stats?
Even though they are a bunch of equivalent numbers, people really perceive the bigger numbers a lot differently than the smaller ones.

What’s misleading?
People don’t have a good sense of the ratios
to translate between these numbers – some numbers sound bigger than they really are!

What’s a typical set of ratios? Every site is different, but these would all be reasonable:

  • 1 registered user = 10 unique users
  • 1 user = 10 pageviews
  • 1 pageview = 2 ad impressions
  • 1 pageview = 10 hits (or transactions)
  • 1 dollar = 1000 ad impressions

What this means is that rather than giving a number like 500k registered users, you could in fact give a number closer to 500k x 10 x 10 x 2 = 100 million pageviews. And if your site is growing 20% a month, growing 500k users 20% a month is a lot less impressive sounding than 100 million pageviews growing 20% a month.

Even worse is giving someone a dollar amount for your advertising. 100 million pageviews doesn’t sound so great when it’s monetizing at $1 CPM or less.

Another way to compound this is by giving big numbers based on non-fixed time intervals. For example:

  • "We’re hitting a run-rate of 100 million pageviews"
  • "Our peak run-rate is 200 million pageviews"

This also includes what I wrote about earlier with "Since we did X, we got Y." It’s very confusing what these numbers entail. If you wanted a great peak run-rate number, just take the HIGHEST number of pageviews you’ve ever had in ANY time interval – let’s say the 10 minutes right after you get a Techcrunch spike, and multiply that out to a month. You’d get a ridiculous number.

How do you figure out the truth?
It’s hard to figure this stuff out, but ultimately you want to heavily bound the problem. Rather than waiting for them to spit out a crazy metric, instead you should ask a very specific question:

"In the last month, what did your analytics tell you in terms of # of pageviews to the site?"

Asking a question like that, and not accepting a cryptic number, will give you the best chance at understanding what’s really happening on their site.

Ideally, you want to be able to open up their metrics and understand:

  • How long is the average session length?
  • How many unique users come to the site per month?
  • How many pageviews did the site receive?
  • How many ads are on each site?
  • What’s the average CPM of the site?

The trick, of course, is understanding what typical metrics might look like for each of these.

The best case scenario is when you’re doing something a little weird and not a typical consumer internet site. For example, if you have a publisher ad network, you’d roll up a bunch of stats from sites you really don’t own or control. Or just disclose certain stats (like peak stats) but not the rest.

PS. Get new updates/analysis on tech and startups

I write a high-quality, weekly newsletter covering what's happening in Silicon Valley, focused on startups, marketing, and mobile.