Using a Viral Game to Get Email Newsletter Subscribers

CHALLENGE
Business newsletter publisher MarketingSherpa has always relied nearly 100% on viral marketing of the most basic kind to build its subscriber lists.

"Readers pass copies to colleagues who then hopefully sign up," explains Managing Editor Anne Holland. "We grew our lists from a few hundred initial contacts to more than 100,000 that way over three years."

"I expected the more readers we had, the more pass-alongs we'd get and then the more subscribers we'd get. I thought, aside from obvious marketplace size limitations, the math would be a percent of our total readership, so if we had 20,000 readers we'd get far more new subscribers per week from pass-alongs than if we had 2,000 readers," explains Holland.

That did not turn out to be the case. "We get around 125 new subscribers from pass-alongs per newsletter per week. That number has not moved one iota in three years. It doesn't appear to be remotely tied to how many readers we already have."

When the team decided to launch a new newsletter, MarketingSherpa, in May 2002 they knew they needed at least 10,000 subscribers for the it to be economically viable from ad and related report sales. At a steady 125 names per week, it would take almost two years to get there.

"We're a small publishing company," says Holland, "We can only wait about three-six months for a title to begin to break even."

The team decided against super-heavily promoting the new title to their current lists for a quick boost. "We definitely did some cross promotion so that current readers who were really interested in the topic would join; but, it's not worth launching another title unless you can reach a new audience. Otherwise your product sales and your advertisers' sales can't really grow much beyond your current boundaries."

CAMPAIGN
As expected, after MarketingSherpa launched in May 2002 it quickly grew from cross-promotions, bartered co-registrations and initial publicity to about 4,000 names, and then leveled off to slow-but-steady pacing.

One afternoon in August, Holland and MarketingSherpa's Tech Editor Alexis Gutzman were grousing about spam overload when they had a brainstorm: Why not create an online game that email list owners could play to vent their frustrations, and incidentally sign up to get MarketingSherpa?

If the game was fun enough, perhaps it would viral out beyond their lists and the rest of the world would play.

Holland contracted interactive agency Cyber-NY to create the game based on her specs. She learned a big lesson from the experience:

"Game developers and Flash animation people will focus on what they love, which is the game. It's a great game, but they spent 99% of their time and energy on that goal, instead of our end-goal which was collecting qualified opt-ins."

The delivered game, entitled 'Torture a Spammer,' was fun to play, but the homepage had obviously been tossed up as an after-thought. "It was a logo plus pics of few of the cartoon characters and a start button," says Holland.

Holland knew that getting conversions (getting visitors to continue clicking their way into the game and joining lists) is tough. She delayed the game launch date a week and contracted a Web developer with some direct response experience to work hands-on with her in creating game entry pages that would achieve her goal. "It wasn't as interesting to the game dev guys because it's boring old HTML, but it was everything to me."

The revised game homepage was inspired by Classmates' homepage. "They have gotten tens of millions of people to join their site by starting off with questions right away. It's brilliant. You're not reading their marketing copy or wondering where to click, you're immediately involved and sucked into interacting with the site," says Holland. (Link to Classmates Case Study below.)

The second page also drew on a Classmates idea by presenting an offer the site wanted visitors to accept as a "gift." Visitors had three options:

1. Join MarketingSherpa's list (clearly marked "for marketing professionals only") and get a no-cost 'Guide to Permission Email Marketing' which was a how-to article sent via autoresponder. All names that joined were cookied with an affiliate code so Holland could track their future buying patterns at SherpaStore and compare them to "regular" readers.

2. Get a copy of Sherpa's Consumer's Guide to Reducing Spam, which was another how-to article sent via autoresponder. (All names gathered for this option were erased and never used again as Sherpa is not a consumer publication, nor did it gather permission to use the names.)

3. Get info on having a viral game built by Cyber-NY, which was another autoresponder email message.

In order to make the game as viral as possible, Holland had a friend-tell-a-friend form added to the end. "I didn't want to encourage anyone to spam though, so I was very careful about the wording."

The game launched Sept 18th 2002. Holland ran brief mentions with a link in two of Sherpa's publications reaching about 24,000 names, but did not push it very hard at all. "The point was to reach outside our current readers."

In addition she ran a bartered space ad in a related industry newsletter reaching 9,000 online ad professionals, and wrote articles about the game for two other outside newsletters reaching an estimated 30,000 readers. The Company also sent out a press release via BizWire.

"Then I had to tie my hands behind my back and not market or promote it any more because the point was to see if it would viral of its own accord after we got the initial word out," says Holland. "It was really really hard for me not to promote it. I was itching to!" She adds.

RESULTS
As of press time today, the game has had 74,441 visitors, just over 5% of whom have joined MarketingSherpa's mailing list. "We got about 4,000 subscribers we might not have gotten otherwise. It's nice but it's not gangbusters," says Holland.

She is extremely glad she had the system set up to track these names through to sales conversions via the SherpaStore affiliate system. Turns out that names who subscribed via the game are far less likely to purchase reports on related topics than names that subscribed from any other promotion such as co-registration, Google search results, and cross-promotions with other MarketingSherpa lists.

"Each month there are about five times more sales from 'regular readers' than sales from game-generated readers, even though it's double a opt-in list you have to really want to join," says Holland. "That part of the list is simply less valuable."

More metrics:

- Roughly 15% of game visitors checked the box to receive the Consumers Guide to Reducing Spam. "It was a PSA (public service announcement), and not a revenue generator for us, but I'm very glad we made it available," says Holland.

- 2.7% of game visitors used the refer-a-friend form at the very end of the game to send a message to someone else. Of these, 42% sent a message to one friend, and 58% send a message to two friends. (Players were only allowed a maximum of two.)

- 58% of referred friends clicked through on the link in the message they received to visit the game. (Wow; the power of referrals!)

- Although reportedly some popular online games make worthwhile revenues selling associated products such as hats and t-shirts, the Torture a Spammer game link to a themed CafePress site only sold $120 worth of merchandise, resulting in a net of $19.

- Holland used 10 different links to the game in order to track clicks through various promotions, but quickly learned tracking was imperfect because once someone gets a link and shares it with friends you never know where it will show up next.

"13,862 visitors used the link that six Sherpa staffers added to their routine SIGs [email signatures]. We don't send all that much personal email - what happened was that link got picked up by a few outside email newsletters and passed around. So it didn't really track SIG effectiveness specifically."

- Just 1.7% of game visitors were from mentions in MarketingSherpa's own newsletters.

- Blogs were a bigger source of traffic than anyone expected. "Our site traffic stats showed random spikes of hundreds of people coming from various Blogs we'd never heard of before, who linked to the game. The word got around in the blogging community, especially for some reason in Eastern Europe and Japan. It was crazy," says Holland.

Crazy is what a viral campaign is all about.

Links:
The game http://www.tortureaspammer.com

How Classmates Got 1.5 Million Paid $29 Subscribers:
http://library.marketingsherpa.com/barrier.cfm?CID=1856

Cyber-NY: http://www.cyber-ny.com