I’ve been exploring Instagram’s new video product for the last several days. I consider it the most significant move from Facebook since Facebook bought Instagram more than a year ago.
Executing a big shift like this on a product used by 130 million people is always a challenge. You have to introduce new product elements in a way that doesn’t upset your loyal users. But at the same time, the new product needs to be discoverable. In the days leading up to the announcement, I had three big questions on my mind:
- Would video disrupt the elegance of the Instagram experience?
- Would there be filters on video?
- Are people capable of making interesting videos?
One of the nice things about Instagram has been that every item has roughly the same cognitive load. If you scroll through your Twitter feed, each item can take very different amounts of processing. Some are self contained, some are pictures, some link off to stories that could be 500 words or 5,000 words. Your brain is constantly shifting modes. With Instagram, it was consistent.
The design of Instagram video perfectly balanced that. Someone who wants to scroll through can continue to do so. Videos don’t start playing unless you pause on an image long enough. That provides for discoverability without breaking the expected experience. As a content creator, you can also select an image to be the image that’s shown in the feed. That’s key because the first frame in a video isn’t always the one that will catch someone’s eye. (Users who are annoyed by auto play can go into settings and turn it off; the default experience is auto play.)
Given that filters have been a key element of the Instagram brand, I considered filters a must-have in video. Instagram has 13 of them. I have to admit I was hoping for one that would replicate old-time news reels with inconsistent lighting and streaking.
The last question is the most critical: Are people capable of making interesting videos?
Creating video is hard
There’s no doubt that people can create videos. According to Facebook, 5 million videos were uploaded within the first 24 hours. If you were to take all of the videos uploaded within the first eight hours and play them back to back, it would take a year to watch them all. At its peak, 40 hours of video were uploaded per minute when the Heat won the NBA championship.
But creating interesting video is much harder than taking interesting photos. And because video requires much more attention, the ROI on time invested has to be there.
“Video is really hard. I make video for a living. Nobody wants to pay for it, so when they don’t, they get shit,” said Mac Premo, a Brooklyn-based artist who responded to comments I made about Instagram video on Facebook. “Let me just say this to the world: World, you are about to find out that you can become very, very bored in the span of only 12 seconds.”
I’ve been closely tracking videos in my own Instagram feed over the last few days. On the first day, there were 26 videos. I liked seven of them. (Most of them were along the lines of “hey, I’m testing out Instagram video.”) Since then, the volume in my feed has dropped to about 5 to 9 videos a day. But the quality has gone up dramatically — I find that I like more than half of them now. That’s to be expected; any new product release that gets the kind of attention a Facebook product does will see huge initial experimentation followed by a drop off as people figure it out. (Facebook PR did not respond when I asked what the usage trends looked like beyond the first day.)
What’s “interesting” is, of course, entirely subjective. You can see the videos I’ve liked on my Facebook page.
Over time, I expect average quality to increase because people will get feedback. If people get likes on their videos, they will continue to produce them. If they don’t, they might learn from other videos that get likes.
Part of the reason creating good short-form video is hard is that we all know what good short-form video looks like. Most of us see it nearly every day. Once a year, we gather around TVs at parties to watch short-form videos. As much as we might claim otherwise, TV commercials influence our opinions of short-form video. Advertisers often spend millions just to produce a 30-second spot. (Not including the cost of air time.)
Instagram isn’t charging for videos now, but it’s not hard to imagine that if people become accustomed to video in their feeds, advertisers will be able to pay to promote videos. And that could be significant in terms of revenue.
There are two challenges to overcome: Much of the existing ad video content is 30 seconds and not in a 1:1 aspect ratio. Either Instagram would have to change its format for advertisers or advertisers would have to develop custom creative.
For startups, these are really touch challenges to overcome. They don’t have the volume to justify custom creative work. (Especially in video.) But with 130 million users, Instagram reaches more people than most TV shows. Even so, when you ask advertisers to do more work, you’re going to reduce your addressable market.
Instagram vs. Vine
Instagram has provided a number of tools to help make creating interesting video easier:
- Increasing video length from 6 seconds to 15 seconds. To be fair, this also allows for longer boring videos. But it can be a lot harder to be creative in 6 seconds than in 15. And there’s no requirement that you use the full 15 seconds.
- Filters. Filters can make uninteresting photos look a lot more interesting; they can do the same for video.
- Minimal video editing. With Vine, I sometimes found myself wanting to delete a segment. But it requires starting over from scratch. By that time, the first moments are gone. Instagram allows you to delete the last segment easily.
- Image stabilization. Shaky video can be nauseating. Instagram’s built-in image stabilization exceeded my expectations.
From a technical standpoint, these features are replicable. But replicating Instagram’s huge user base is a much tougher challenge.
Areas for improvement
A number of small improvements would make Instagram video even better:
- The camera retains state. If I recorded a video last, next time I go into the camera, I want it to be on video. Currently, it always defaults to photo.
- Tick marks at the 5- and 10-second marks. This would help with composition.
- Ability to drop audio. Many of the videos I see on Instagram are scenic videos and panoramas that suffer from wind noise. An even better solution, long term, would be to implement wind-noise reduction algorithms. Dropping audio would also be helpful when making videos with quick cuts.
- Audio cues. These would assist when you’re creating videos with others. I created a video where someone needed to know when to toss a bean bag. I did it audibly and you can hear me saying “OK” in the resulting video. Ideally, there would be a tone that would be subtracted out of the resulting video.
- Ability to tag people in videos. You can tag people in photos, but not in videos. I don’t expect you to be able to tag frame-by-frame (that would be tedious in terms of authoring), but you should be able to indicate that someone is in the video.
Overall, Instagram video is a terrific product that manages to introduce new functionality without breaking an existing user experience. That’s a significant feat.
It’s clear that the Instagram of video will be Instagram.
Rakesh Agrawal is a consultant focused on the intersection of local, social, mobile and payments. He is a principal analyst at reDesign mobile. Previously, he launched local, mobile and search products for Microsoft, Aol and washingtonpost.com. He blogs at http://redesignmobile.com and tweets at @rakeshlobster.
VB's research team is studying web-personalization... Chime in here, and we’ll share the results.