[time_to_smile]

> "you need it to promote your OSS stuff"

I have found this to be the biggest myth of Twitter: it is nearly useless for content marketing.

A long time ago (so it's possible this isn't true any more) I did some analysis on a pretty large number of domains and found that Twitter had virtually no impact on traffic. There were brands that had a large twitter following, but in nearly every case I could find this was because a popular brand happened to have an audience that also was active on Twitter, not because Twitter activity helped to grow their brand.

I also had/have (much more recently) a Twitter account with > 5k follows, that was pretty active (i.e. I didn't just game Twitter to grow followers, something I have done in the past, which is a separate issue). I had a couple of Tweets with more than 1k likes, many with more than 500. First off, most really viral content is viral because it has nothing to do with your "brand", so that gets no traffic. However, even tweets that are relevant, linking to specific articles etc. only get a small blip in traffic and no visible long term improvement.

tl;dr in both my research and anecdotal experience Twitter is not a good marketing tool

Twitter can be good for communicating with your users/audience if they happen to already be there, but it won't grow your brand.

Twitter can be useful for meeting people. I have a small number of friends that I have met there.

But as a marketing tool it is a complete waste of time.

[lapcat]

> I have found this to be the biggest myth of Twitter: it is nearly useless for content marketing.

I've found Twitter to be essential to my business. Perhaps you're looking at it too "directly", in terms of traffic analysis. What's important in my opinion is "worth of mouth" advertising. You have a circle of friends on Twitter, your friends recommend your product to their friends, and so on. If your Twitter content is interesting and valuable, you might even get members of the media to follow you. Or at least members of the media might be following one of your followers and happens to see your product that way.

[time_to_smile]

I'm genuinely curious if you've measured this, and how you determined the impact of twitter on your business.

It's certainly possible that other people have different success with twitter.

In my case I was looking over thousands of websites traffic history primarily concerned with long term patterns of growth based on activity among different social channels, SEO, news aggregators etc. The vast majority of sites primarily saw permanent growth in traffic as a function of SEO success. One high ranking article can get you a increase in average traffic that can last as long as you're on top. The only sites that didn't benefit from SEO are buzzfeed-style sites that exist solely from a constant stream of traffic coming from new articles in news aggregators. But Twitter doesn't generate a substantial burst in traffic that reddit or other sites do. Plus if you're popular on reddit, you will be reposted to twitter without any effort.

I should add that when I discovered this I was unhappy about it, as I was currently spending a lot of time on Twitter promoting my own project. I really enjoyed twitter at the time, and it felt like I was succeeding. I wanted all my tweeting to be a valuable activity. After doing the numbers I periodically would just stop using twitter for months with zero impact on website traffic.

[lapcat]

> I'm genuinely curious if you've measured this, and how you determined the impact of twitter on your business.

Well, Twitter is basically my only form of promotion. :-)

I've tried a number of forms of advertising, the results of which I can measure directly, but I've lost money every time, so I usually don't bother with that.

There are times when I can measure Twitter's impact directly. For example, I once had a very viral tweet that immediately resulted in a massive sales week. Admittedly, that's pretty rare though.

[jklinger410]

> I did some analysis on a pretty large number of domains and found that Twitter had virtually no impact on traffic. There were brands that had a large twitter following, but in nearly every case I could find this was because a popular brand happened to have an audience that also was active on Twitter, not because Twitter activity helped to grow their brand.

How did you perform this research?

[time_to_smile]

Many years ago at a small private company that had access to thousands of Google analytics accounts ranging from recognizable name brands to people just trying to get started. You could actively watch small brands grow big over history and in all cases it was clear that twitter success (in terms of traffic from twitter and follower count) trailed brand development.

This was also not universally true of all social media. Facebook for example did precede growth in quite a few cases.

I also want to repeat the caveat that this was quite awhile back, so I wouldn't give it too much credence for today.

However my Twitter critique also grows out of several cycles of activity and inactivity over many years running a reasonably high traffic (for it's niche) blog and seeing these same observations. Twitter activity only caused minor bumps in traffic, and never correlated with a shift in the baseline readership.

Others here claim different experience, and it's quite possible they are correct. I have found personally that it's very easy to conflate the "feels good" of getting a lot of likes/follower/etc with the illusion that "this is working". Twitter is like a micro HN front page, only at least with HN you'll get a pretty ego inflating spike upon success (which is in practice is bad for marketing because it makes it easy to focus on the things that feel good over the things that work).

For better or worse the best source of consistent increases in traffic was (and from my experience still is) SEO. A high ranking post will have the highest probability of shifting your baseline traffic up.

[affectsk]

Yeah, I wouldn't rely on it as a primary marketing channel in most cases, especially if I had to build it from the ground up.

Conflating "feels good" with "this is working" is right on point, too.

I'm sure it works better as a marketing channel for specific niches, though (just like Instagram works best for certain types of brands).

Other than that, it's just a good place for all kinds of organic engagement in general if that's where your audience is. Not necessarily marketing.