Gaps Between Goals and Efforts to Drive Desired Outcomes – SpiralFrog Case

I found this article from CNET, which is interesting from web analytics point of view. It clearly depicts businesses are looking beyond hits and pageviews, but shows a wide gap between web site objective and desired outcome. Web analytics analyst should view this as a learning. It also underlines the importance of web analytics data and the leadership of an analyst.
Inside the short, troubled life of a music start-up

Here is the background of SpiralFrog’s business.
SprialFrog is a free digital music download service, which became the first company to convince a major music label to offer downloads on an ad-supported basis. The company’s goal was to give away music and supported itself by selling advertising. SpiralFrog’s management believed that the record companies would rush to do business with anyone who competed directly with illegal peer-to-peer sites, and they have focused heavily on simple plan to grow their business via aggressive marketing to turn visitors to loyal users.

Here are some key highlights from the article. Mostly a reflection of bad decisions. Let’s see how you see it…

• Low registration rate, pages per visit, time on site, high bounce rate
• Company burned $26.3 million while generating sales of just $1.2 million
• The dramatic falloff of site traffic coincides with the board’s decision to cut spending on affiliate marketing
• Management says “it needed to build volume and then swing over to quality. If you didn’t build the volume, you could never get ads on the site from tier-1 advertisers”.
• At one point, traffic grew from 2 million visitors to 3 million in the following month
• Affiliate marketing programs were introduced to spread the company’s brand across its own site and many other affiliated sites.
• In January 2009, SpiralFrog’s 2008 sales and marketing expenses came to $11 million (twice the $5.6 million the company paid in music licensing that year).
• One of their promo’s ROAS was $60 per registered users. In worst case, they paid $490 for registered users.
• When the marketing programs were halted, traffic numbers crashed. SpiralFrog saw just 775,547 unique visitors in October, a fraction of the site’s monthly peak of 7 million.

These are clear indications of two different HiPPOs (SpiralFrog management & investors) with one objective to make money, and outcomes driven by poor management and allocation of efforts. In other words, objective of making money off free music, with one outcome to drive enormous site visitors without putting emphasis to convert new visitors to loyal customers.

HiPPOs: Borrowing Avinash’s term “HiPPO’s”, which means Highest Paid Person’s Opinion.
From web analytics analyst stand point, it is important to diversify your analysis and translate the data into information for business. In SpiralFrog’s case, web analytics analyst should emphasize their analysis and insights on:

• Measure sticky traffic driven by marketing efforts. Look beyond increase in site visitors. 1,000 visitors with 1 person converting, and putting more money just to increase that 99.9% of non engaged traffic makes no sense…
• Translate metrics to further optimize landing pages to increase conversion and ROAS.
• Gauge web metrics data and support CRM efforts to tackle efforts that work in converting new visitors to return visitors, and optimize marketing messages to drive loyal visitors.
• Test various campaigns that drive return visitors/ad clickers.
• Stop investing in marketing campaigns that don’t work, and support business with metrics that eat lower hanging fruits instead of trying to consume the entire tree.

After all, at some point, management should have noticed that increase in traffic doesn’t translate to great services and user satisfaction. Few ways web analytics analyst could do to assess learnings and support increase in satisfied users are (in SpiralFrog’s case):

1. Segment your loyal visitors data, and see who they are and how are they engage with your service.
2. Web Analytics can only tell you the “what”, but not “why”. Take surveys, and assess the qualitative insights.
3. Look at visitors who bounce vs. visitors who don’t bounce when arrived to the site. “High bounces equal throw more money to bring visitors that don’t bounce” is a mistake. Is it a specific ad, keywords, ad group, banner color, or CTA that caused high bounces?
4. Measure and optimize rigorously on CRM efforts. You are dealing with loyal visitors, what can you do to increase and turn them into your “advocates”?
5. Track affiliate networks that drive loyal visitors. Affiliates emphasize on numbers rather than quality. They get paid for what they refer, not for referring satisfied visitors.

It sounds like I’m talking trash about SpiralFrog’s web analytics analyst, but I can really understand the challenges in small companies that is backed up by huge amount of cash. HiPPOs are probably the bottle neck, and is the primary cause of not being able to take best actions based on web analytics insights.

Web analytics analysts are facing a steeper challenge in 2009. If these HiPPOs start to pretend they know all or everything about web analytics, investors and businesses will continue to burn cash for nothing. Plus they’ll blame the analysts and eventually media will use web metrics to put blame at web analyst indirectly. Maybe that is a good stimulus plan for this economy… having these investors burn cash… j/k.

Good luck to all of the web analyst in the world !!