Over the weekend, I ran a somewhat unscientific experiment to compare various uses of hashtags on Instagram and see what they did in terms of traffic, likes, comments and so forth. I feel the results are somewhat predictable, but nonetheless interesting to discuss.
Fishing for Interaction
For this test, I ran three variations on the theme. First, I used a post with zero hashtags, then one with three popular ones, then a third post with a slew of spammy like hashtags. I've learned a handful of things from this test, which I will get into at the end.
For each test, I posted an image from California that had a predicted outcome based on how a similar image posted on a similar day and time behaved. For example, the first image was posted on a Saturday at 7 am, and the reference was posted 2 weeks earlier, on a Saturday at 7 am. I know the count of likes isn't a very good barometer of how an image behaves, so I am coupling this with impressions, reach, engagement and who "liked" the image (meaning current followers vs people that happened across this stream).
For this one, I chose an image taken in Coronado, California of a yacht club. I posted a very similar one (same scene, almost, in landscape) some weeks back at a similar time that had pulled in 170 likes and seven unique people leaving a comment. Six comments were left by followers, one by some Spammy McSpammer dude.
So my benchmark was falling within +/- 10 likes and five+ comments to be a statistically similar performance.
After 24 hours, the post had settled down with 110 likes and five unique people leaving a comment. All five people were current followers. With 60 less likes than the similar post, that leaves me to believe about 35% of my "like traffic" is missed when not using a tag, but engagement is pretty similar. Who wants spam jerks leaving comments anyhow?
The 110 likes were (from what I could tell) mostly current followers. There were a handful of handles I didn't recognize that could have come in a number of ways - like following the location
Next, for my second test I wanted to carefully choose three hashtags that seem to deliver on this kind of post. Again, I choose a very similar image - same spot on Imperial Beach in California, but in a horizontal rather than a vertical position posted on a similar day of the week. Benchmark for the original was 218 likes with once again, six comments (no spam on the original!) I knew this was a somewhat high performer, so I was hoping to see this one in the 190-200 range for likes (it did not get there).
I picked the three tags: #beautifuldestinations (11.5 million tagged) #sonyimages (1.6 million tagged) #sandiego (15 million tagged) for this. What I found is you get a swarm of likes in the first 10 minutes, but as these are such popular tags, it dies off right away (gets buried) and the remainder of "likes" look to be from regular followers, with a few stragglers coming in late from tag surfing.
After the 24 hour period on this one, it looked like 100 or so were from regular followers and the remaining 32 (give or take) were from those hashtag chasing. Again, this is using three ultra-popular tags, so the post gets buried really quick. I needed one more test with a bend of popular and not so popular tags to see how the traffic flowed.
The final test was the mix of tags as mentioned above. I wanted a bunch of heavy hitters with 11 million or more tagged to front load the likes, and some in the lower counts, like the #zeiss tag which hovers in the 6,000 range of posts, making the image appear near the top of the tag search for the day and stretching out the likes (which I believe works fine).
For this post I chose an image I felt would perform OK on it's own (these were taken in La Jolla, California), but not be spectacular by any means. Optimistically, it should be in the 100+/- range with a comment or two under normal circumstances. The compare post did 125 likes with no tags at all, and just one comment.
It did much better than I would expect without the tag-storm - 158 likes in 24 hours.
I ended up with four comments from unique followers on this, compared to the one on the reference photo. I give the extra comments to this being slightly more interesting with the ivy house and palms (the other was a bench and palms, right next to this). But again, no real engagement from non-followers.
What I Learned
So, what I got from all this was that hashtags do drive a little traffic to you, usually in a pre-loaded burst and more often than not, with zero engagement. There are some people, but not a ton, that follow hashtags to your post. Chances are they scroll past your image when it's at the top of the hashtag stream (for example, the #a7rii stream) and some will click through to give you a like. Less will take the time to leave a comment, as they don't know you from Joseph. For popular tag streams, I say it's pretty well buried within the hour, never to surface again. For the less popular streams, you can make it hover for days in that sweet "scroll zone" - you know, that amount of images you're willing to scroll through before you're bored with a tag.
You really need to decide on the purpose of your post. Is it engagement, or as many red hearts as you can collect? I know for me, I want the conversation, discussion, and feedback - I can care less about the hearts. Engagement seems to start with followers.
Followers per 24 Hour Period
You may gain a handful of followers with tag use and if that's your goal, mission accomplished! Be warned, a lot of times the followers are merely fishing for a follow back. If they don't get one in a day or so, they un-follow (I should say, their bot un-follows). I am sure you've seen the routine.
These are not your engaged followers.
To the right is what I saw over the 72 hour period. The ratio of new followers to those hit-n-run followers shows that with the 3-tag array, I attracted a few and lost just a couple. With the storm, I attract a ton of bots that do un/follow - at least that's what I imagine they are doing. So, don't tag-storm in hopes of a ton of new followers. It just seems to create a bot-storm.
For me, the real goal is to attract engagement. (And by engagement, I don't mean the Instagram algorhythm that's referenced in the above charts, but real folks having a chat with me.) Tags do not accomplish this at all. I have a less involved test going on right now that reflected above in the 4th blue bar above, finding like minded photographers on Instagram and engaging with them in conversation. In the few days I have been trying this, I have seen my follow count increase as well as my engagement in new posts - but no decrease in followers over these three days.
People really want the conversation more than that little red heart.