Roughly a month ago, I decided to give blogging another try, in earnest. I put out an article most days and up to 3 per day when I was experimenting with new channels.
It has been fascinating.
That being said, this is not a post about blogging tactics. It’s about what the experience has taught me about startups.
Find the customers first, then build for them
It took me a long time to understand YC’s motto that you should:
Make something people want.
The first time I heard that, I thought it was the single dumbest piece of advice I’d ever received from such a credible source. Obviously you hope people will want your product. I thought it was a judgement call — that “good” products are the ones people want.
I was missing the implied qualifiers.
Make something a large group of people already want.
So I found some big communities I had overlapping interests with and wrote at the intersection. I wrote for them, while still remaining within my zone of excitement and experience.
The de-mystification of marketing
I’ve spent ages testing different customer development approaches and struggling to come to terms with building repeatable sales roadmaps. I have never had to worry about an on-site conversion rate.
Perhaps I should have! It’s not that hard and gives you some neat options.
Tweaking a button and seeing conversions increase 8-fold is a beautiful thing. Seeing an hour of work turn into 120k visits is magical.
I remember launching my very, very first website, and then stepping back, looking at it, and thinking:
And now, we wait.
Man oh man. What I would give to get back all that code I wrote and go to war with it now. Wait for traffic!? What was I thinking?
You find a community, learn what they love and build something for them in order to re-route traffic and grow your own channel. None of those are passive.
I feel like, in a month, I just got the experiences normally reserved for those who have spent 2 years succeeding or failing with a marketing-driven startup. I’m certain they’ll be better at it than me and be aware of a few more traps, but I’ve gained an awful lot in a short time.
A sense of realistic scale
Hacker news traffic caps out at about 35k direct visits, even if you crack 1000 points. A “normal” #1 of 200+ points will drive about 20k.
A reddit #1 can comfortably drive between 5k (on a 100k subscriber subreddit) and 100k+ (on a default front-page subreddit).
Twitter adds 25% to any other traffic source. Stumbleupon adds an additional 25%, but only to the big hits.
A highly viral post will bring in 100k+ visitors. A solid, successful post will bring 20k.
You can convert visitors into subscribers at 1-4%, depending on how aggressive you are about it. An unoptimised subscription link will convert 5-20x worse than that range.
My twitter followers (hi, I love you!) click at roughly 1%, but that seems to get worse as you grow. For example, Fred Wilson’s 200k followers click 4x less, at 0.25%.
Blog traffic naturally bounces at 85-90% (I haven’t worked at all to optimise this yet). The ones who stick around tend to follow the most obvious call-to-action, which means…
Having no product is tragic
Pulling numbers from a hat, let’s say that 1 in 10 visitors sticks around and half of them will follow the main call-to-action (the other half will click wildly wherever they please).
100k visits can turn into 5k product clicks, assuming you’re building and writing for the same audience and not just trying to make your chart go higher by gaming irrelevant communities.
20k visits, which is much more predictably achievable, is still 1k product clicks.
If you have a product making enough money to profitably buy search ads at $1-5 CPC, that puts the value of a good (in terms of both writing & promotion) blog post at $1k-25k.