The Big List Blog

| All |

NVIDIA: “The Global Snowball” Video Exceeds Expectations — Live from BlogWell

Coverage of this session by Angela D’Arcy of The OutCast Agency. Connect with her by following her on Twitter.

2:10 —’s Erin McDaniel introduces NVIDIA‘s Social Media Strategist, Shanee Ben-Zur.

2:11 — I’m going to talk about a viral video, how many of you have been asked by your company to create a viral video?

2:12 – I’m not going to tell you how to make a viral video today, but I will tell you how to collaborate across teams to make a successful video.

2:13 – Most of our social team is here, we don’t spend money on ad buys, we did this with a small team across product engineering, social, PR, and marketing to create this video.

2:14 – We had a mobile product coming out  and decided to have a phased announcement to show people different features, first video was “Glowball 1” and then a few months later announced our second video “Glowball 2”. I’m going to tell you about how the first video did very well, and the second video did not so well.

2:15 – Let’s jump into Glowball 1, not going to show you the video now, but it was interesting with a creepy clown and glowing ball that moved around in a circus environment. It was devoid of marketing, showing the features in a creative way.  The video told people what they should know about the product without marketing talk.

2:16 – We worked with our product engineering team to make a demo, marketing to create the video, and PR identified targets to pre-brief and the social team identified where to post the video to make it easy to access and embed.

2:17 – We put the video on YouTube as unlisted, gave press that link, got on the phone with them and walked them through the video, followed up by email to get them the embed code for the video, gave them the draft version of the blog post with an embargo time, and handed this all to them. If you spoon feed them content in an easy way, they will use it. They are busy, this makes their job easier.

2:18 — Once press posted stories and the video, we made the video live and sent to tier 2 press. Our goal was to get views, but more than that to get people to understand this product. We have a hard story to tell so used the media to help us tell the story.

2:19 – In the first week we had over 600K views, now at 900K views, our most watched video, with 800 comments. We always get asked how to measure, so we just benchmarked against ourselves,  and this was the most watched, most commented video we’ve ever posted.

2:20 – One thing we didn’t do so well with Glowball 1was to respond to video comments, but comments were generally positive and asking about product availability and price, which is great for us.

2:21 — The majority of our views came from referral sites with the embed code – 77%, we know that this happened from our outreach because of the embed code used. Engadget was the #1 referrer.

2:22 – Months went by and we were getting ready to launch the product along with a video, “Glowball 2”. We worked with product engineering to pull together the video to show functionality that we didn’t show before. Overall, this second video was less successful, but when we were putting it together and planning everyone came out of the woodworks since the last one was so successful.

2:23 – Overall, it was less compelling and got less pickup. When we launched 1, it was the only thing we were pushing. When we launched 2 we had 3 other blog posts and 2 other videos, too much competing content. Press and users were confused. Viewers and press have a limited attention span, so you should focus all of your attention on one thing.

2:24 — We are very lean, we are very scrappy, that’s what we are known for. But we had less advanced planning for Glowball 2 than with Glowball 1, so overall the numbers were low. Only 11.5K views, 122 comments (weren’t very positive), didn’t do pre-briefs so only 30% were embed traffic. 5 videos launched at once, which fragmented our audience.

2:25 – We hurt ourselves in trying to sell too many messages. Across all 5 videos that were launched along side 2, we had 350K views. This was a great, horrible experiment, where we learned that early planning, coordinated efforts, and strategic deployment leads to success.

2:26 — Key takeaways: make a good video, no marketing, make the message easy to understand, and make it short. We looked at the heatmap on YouTube, and saw that at 1:30 is when people typically start to click away.

2:27 – Keep the topic focused, if you have more than one video, make sure that they are different enough to warrant separate outreach plans. If they are similar, see how you can combine in to one, so that you are not hurting your own efforts.

2:28 – Plan early and coordinate a multi-team effort. And be sure to deploy strategically. Is the launch around an event? Are you working with influencers/evangelists to distribute? This is just as important as figuring out how you will distribute through your own channels.


Q: How the internal stakeholders reacted to the key learning of 2? Were they receptive or did they have their own conclusions?

A: This question is asked at every meeting, and within our company we can say “we failed”, when we said what didn’t go well, it was very well received. Case studies are very helpful. I collect them from our company and others for people to read.

Q: Measurement, are you just using YouTube insights or additional platforms?

A: Because we don’t have a lot of budget or people, we use the insight information. They provide a lot of information and we also use Google Analytics on our own blog. Also looked at click throughs for what we put out on our own Facebook and Twitter channels. While the metrics were successful, you can also see as a company that the video was valuable to promote the features of the product.

Q: How did you decide to do a video for this launch?

A: We create products that are visually stunning, so video is a natural way to showcase the content. We knew this would be a good way to show people who may not  understand the product, and better understand the features.

Q: Can you tell us more about the ratios/views of the video?

A: We just looked at our previous videos, and benchmarked against ourselves. And then also between 1 and 2 could look at the comments to see the sentiment, which is how we knew that the reaction to 2 wasn’t so good.

Q: There was such a significant difference between the content in the 2 videos. Were there discussions around what kind of content would be more completing?

A: I give all the credit to product and engineering teams. They were creative in showing the features. For 2, it was evident that in creating so many videos we weren’t able to focus our efforts.

Q: You talked about too many cooks in the kitchen, can you talk about who is responsible for prioritizing/leading?

A: The key takeaway for 2 was that if we had brought in all the videos earlier, myself and our blog manager would have seen that it was too much content to put on the blog and social all at once. Even though we have a flat company, we have a lot of teams that need to reach their goals. We are trusted partners for these announcements, so it’s our job to bring that up. Unfortunately, we didn’t have the discussion on 2 until 24 hours before the launch, so didn’t have the chance to do that. If we did we probably would have gone down a different path.

Q: What was the time difference between 1 and 2 video? Could that have had something to do with the views?

A: it was several months, maybe 5, the difference was the first one announced the codename, the second was showing the product and all the specs. We lacked the strategic planning, and it was diluted from all the other content.

Q: Have you found that this case study has given you a bit more power internally?

A: Absolutely, my first year was me swimming upstream, no one would listen to me. And when I pulled together these case studies, there is this weird power in pulling together a study like this. Everyone, at all levels, reference this case study. We are planning for an event soon, and the video discussion came up, and a VP was able to reference all of the YouTube stats, including the heatmap in the meeting.

March 27, 2012 0 comments

Comments are closed.