Some people love our headlines. Some hate them. Here's the math behind why our headlines are the way they are.

By Anand Sanwal (@asanwal), CEO / Co-founder / Customer Service

We get a good amount of love for our newsletter headlines.

great newsletter headlines



But they’re not for everyone as you’ll see below.

newsletter headlines


I get about 3-5 emails per week like this. It’s a very small number in a newsletter of 175,000+ subscribers, but when people write in to complain about the newsletter subject lines, they are often pretty fired up aka hardcore.

We usually ask them if we should unsubscribe them (the answer is usually no), and although it’s a small vocal group that hates our headlines, we wanted to provide a view into what drives these headlines and what we’ve seen work and what doesn’t work.

Join 250,000+ others who get our newsletter
Get free reports, data and graphs on emerging industries, startups & VC.

Here’s the process we use to pick our headlines.

  1. For every newsletter, we try 3 different headlines as part of an A/B/C test
  2. 6.67% of our newsletter subscriber base gets headline A, and 6.67% get B and 6.67% get headline C
  3. Typically after 1 hour, we pick the winner based on highest open rate (while also looking at % clicking and # of unsubscribes)

So the headlines you ultimately see are based on the data.

But, there is a rhyme and reason to our headlines based on lots of data and learnings over time.


Newsletter Goal #1 – Open Rates

The #1 goal of a newsletter is to get folks to open it.

That is all.

If they don’t open it, all the great work our research team is lost. So we’ve optimized for getting that initial open. And headlines are the key determinant of whether the newsletter will be opened.

Now some of your will say “If you built trust with your subscribers, they’ll open it irrespective of the headline.”

That sounds nice and all, but it’s not how the world really works.

The open rate data we have shows the difference in open rates can be massive when we pick a headline we like but which is not optimized for the open. So while I love the sentiment behind this idea, the data tells a different story.

Headlines matter. A lot.

On a newsletter of 175,000+, a 5% difference in open rate means 8,750+ folks either see or don’t see the newsletter. That difference also translates into revenue.


Getting newsletters opened. What works?

We issue the CB Insights newsletter 6 times per week. It’s growing by about 1300 net new subscribers per week. So it’s working. As this article in Fast Company about our newsletter highlights, we have above industry average open rates as well.

And the upshot of sending so many emails is we have lots of data to know what works.

Here’s what works for us. (Note: There are no rules with newsletters. Do what is right for you)

  • Brand names – There is a reason big tech blogs cover Google, Apple, etc so much. People care about them. Brands work.
  • Short titles – Less is more.
  • Negativity – Unfortunately, “A unicorn takes flight” will do worse than “A unicorn falls to earth”. The news adage that “if it bleeds, it leads” seems to hold here.
  • Surprises – Contrarian views or data on widely held beliefs or areas do well.

Here’s what doesn’t work:

  • Obviously, the opposite of the above. Long headlines. Positive headlines. Proof of trends everyone knows. These don’t work. The worst headline I can think of would be “Koch Brothers says big data is powerful”. Literally, nobody would open a newsletter with that headline.
  • Questions – The old Buzzfeed or Upworthy style curiosity gap style headlines don’t do well for us. Things that end in questions in particular do terribly. Our audience is pretty technology proficient and so our assumption is that they are probably a bit tired/jaded of these headlines and so they don’t work.
  • Niche important – This one makes me sad a bit but if we do a brief on cancer therapeutics or wearables focused on Parkinson’s and indicate in the headline, they don’t do as well on open rates.


The money from higher open rates

When we A/B/C test campaigns, the gap between the winner and loser averages out at about 2.6% (this is the gap between the highest, winning open rate headline and the the lowest one).

That means that the winning headline results in 4,620 incremental opens per newsletter.  Over the course of a month, that is 115,000+ incremental opens. If we apply a conservative estimate based on some of our data from 2015 of how many of those incremental opens try the CB Insights free product (0.09%) and that only 1% of those trials buy a subscription, we’re looking at nearly $625,000 of incremental revenue.


Said another way – that means an incremental $625,000 in revenue per year as a the result of better headlines. The rough math is below. These are extremely conservative estimates as well (based on 2015 data) so in reality, the contribution is much larger.

But irrespective, headlines matter. And there is a method to the headline madness.

Thanks for reading. Let me know if any questions.

Now go subscribe to our newsletter.

newsletter headlines


  • Rees Cahen

    Insight to the insight

  • Per Edström

    Love your transparency on insight.

  • Jonathan Drake

    While opens are a good metric my strong hunch is that revenue per click is a stronger metric to track. You should be able to see a click to revenue cohort where you can determine the $ per click. Also believe that you will find a few “magic” numbers of clicks to optimize towards. So many more metrics/indicators to measure/monitor/react to that will provide a clearer path towards what to optimize for with click cohort to revenue being a great place to start.

  • JohnT

    I CB what you did there.