The Nine Rules of Viral Marketing -- Hitman 2 Videogame Promoters Reveal Their Tactics

CHALLENGE
Approaching the launch of the sequel to the Hitman "thinker-shooter" video game, Eidos Interactive faced the perennial games marketing problem.

It is relatively easy for a company to reach frequent game buyers through, for example, relevant print publications and websites. For a bigger sales impact, Eidos needed to reach those console or PC gamers that do not read gamer magazines or visit gaming websites.

Eidos marketing manager Jon Rosenblatt: "We needed to try and move away from the traditional advertising that games companies tend to do."

Just to compound the problem, Hitman 2's target demographic was the British male, aged 18-34 segment, one which is notoriously difficult to reach anyway.

CAMPAIGN
Rosenblatt teamed up with production company Maverick Media, and viral marketing service Digital Media Communications Ltd. (DMC), to produce two entertaining Hitman 2 video clips for an online viral campaign aimed at raising brand awareness.

The clips were designed to appeal to the target audience, contain material similar to the kind of material this demographic passes around anyway, and reflect the game brand, by being somewhat "different" and "underground." (Link to samples below.)

The two 25 second "f**ktalking" clips featured "everyday" real-life scenes (one a shop queue, one a police line up) ending with a humorous and shocking twist. There is some gun violence and a British-sense of black humor. According to Rosenblatt, the online environment let the company, "show something a little bit grittier than we could perhaps with a print or TV ad."

The clips were ready to launch in late Sept 2002. Next, the team began a viral marketing campaign to get lots of young men to view them.

Here are the nine rules of viral marketing that Rosenblatt and DMC's Justin Kirby base their campaigns on:

Rule #1.
Do not obsess over producing meaningful metrics for viral campaigns. Kirby admits the tracking numbers can only provide a general estimate of what is going on.

He suggests some other ways of evaluating success are to count the number of sites offering, for example, the download and to do appropriate Google searches to evaluate how people are talking about the viral content.

Rule #2.
Do not try and compare a viral campaign with, for example, banners. A banner view is passive, while a video clip view involves a proactive, involved experience. There is also the added benefit of third-party endorsement through download site editorials and comments made by people as they pass on the creative.

Rule #3.
Achieve critical mass as quickly as possible by seeding all the bigger sites. This is particularly important if you are trying to create word-of-mouth buzz before a more formal marketing campaign.

Rule #4.
The best way to ensure prominent and positive coverage at relevant sites, and a higher pass-on rate, is through quality.

Kirby: "The idea that you send good material to a list or your family and friends, and it then goes around the world just isn't true. It might have happened at the beginning but now there's a hell of a lot more material floating around competing for people's attention."

He favors video clips for viral work, as other formats suffer from too much competition; "There are plenty of people making games, but video clips have a cost, so there's a barrier to entry and less competition."

Rule #5.
Get immersed in the scene. The insider view gives you a better understanding of what viral content works and helps you develop relationships which will keep you in good stead when it comes to seeding material.

Rule #6.
Find advocates, people who run a community or are linked to other communities. "People who'll in effect say to a large audience, this is a really cool clip, check it out." These may not be people traditionally regarded as influencers offline.

Rule #7.
Learn to let go. Once the viral content is out there, you have little control over how it is presented, which can be hard for brand managers to accept. It is another reason for ensuring high quality. Rosenblatt: "lack of control is only an issue if we're going to show something which is just plain revolting."

Rule #8.
Avoid looking like a marketing campaign. Kirby: "These clips are not being judged by TV advertising production values and aesthetics. In fact, the less they have TV production values and aesthetics, the better they seem to do. People don't want to be part of someone else's marketing campaign."

Rule #9.
Ensure the viral effect is not geographically limited (if you want to hit a global audience). Rosenblatt: "If you're going to do a viral campaign you need it to be as universal as possible. In this campaign, the humor was quite English and didn't translate well enough perhaps on a global level."

Bearing these rules in mind, the team launched the viral marketing campaign using three steps:

-> Step #1: Identify seed sites

Luckily Kirby monitors and reviews the viral video site community online as a dmoz games category editor and as publisher of the ViralMeister weblog. The team relied on his experience to pick seed sites.

Seed video sites act as central hosts of (or portals for) viral video material, attracting visitors looking for entertaining downloads. Visitors can download the clips, view them and pass them on to friends as email attachments.

Kirby prioritized the sites he chose in terms of reach (essentially traffic levels) into different groups:

Group 1: Large dedicated networks like iFilm, which generate more monthly users than the combined circulation of many national (UK) magazines.

Group 2: Large professional portals and hubs, where the viral sections are a significant sub-element of the site, such as the Lycos UK Viral Charts.

Group 3: Semi-professional sites which generate significant traffic.

Group 4: Small hobbyist and amateur sites likely generating little traffic.


-> Step #2: Seed the video clips

Next, DMC emailed chosen site owners (mainly in groups 1 and 2) about getting the clips featured. Kirby's own personal relationships with meant site owners paid attention to the email.(Links to some of the sites below.)

The real key, though, to getting prominent positioning and a suitable endorsement or editorial from a site is simply quality.

Since the sites need content, and there is plenty of competition within the genre, payments are not required. (Kirby says payments are generally only ever involved as part of a wider media buy at a portal site, like Lycos, where the viral seeding is integrated with other online media campaigns.)


-> Step #3: Track the results

Most of the sites redirected clip downloads to DMC's own server, so DMC could measure the number of downloads and record the IP address of the PC downloading the video.

The video files were programmed such that the media player made an http call to DMC's server when loaded, i.e. viewed, allowing "opens" to be tracked and giving a second set of IP addresses for views.

By comparing the first and second sets of IP addresses, it was possible to gauge the degree to which people pass round the downloaded material.


RESULTS
Within 8 weeks, the campaign produced 404,448 recorded views. The cost per view (which includes creative, video production, seeding, tracking and any media placement) at that time was around 7 pence (about 11 US cents).

More metrics:

- The download link with site editorial (if any) was exposed to an estimated 9,150,000 people. This figure comes from watching how long clips are featured at a site and then cross-referencing this with what is known about that site's traffic levels and pageviews (as supplied by the video sites themselves).

- Some 3,000 visitors clicked through directly to the Hitman 2 site (since it was a branding campaign, clickthroughs were seen as a bonus).

Rosenblatt: "The game has sold over 1 million units. It's impossible to attribute sales to a particular campaign but a big part of the success was the viral campaign because it got us into a lot of great places."

These results underestimate the true impacts of the campaign, since offline views etc. are not measured, and the clips are still doing the rounds of the Internet, so the cost per clip will continue to fall.

The obvious tracking limitations mean Eidos/DMC did not want to quote a specific pass-on rate. However, by looking at the two sets of IP addresses recorded, and comparing the pattern with other DMC viral campaigns, they have rated the viral element as "successful."

Related Links:

View the video clips at: http://www.dmc.co.uk/index.php?bz0yNg==
http://www.hitman2.com/ - game website
http://www.eidos.com/ - Eidos Interactive
http://www.dmc.co.uk - Digital Media Communications Ltd.
http://www.maverickmedia.co.uk/ - Maverick Media
http://www.ifilm.com/ - iFilm
http://viral.lycos.co.uk/ - Lycos UK viral chart
http://www.viralmeister.com/ - ViralMeister
http://dmoz.org/Recreation/Humor/Audio_and_Video_Clips/ dmoz category