MetaMuse

How Effective are Groupon Promotions for Businesses?

Just read How Effective Are Groupon Promotions for Businesses, which was a survey-based study of 150 businesses that ran and completed Groupon promotions between June 2009 and August 2010. (If you don't know what Groupon is, skip to the end.) Below are my key takeaways:

1. The key factor contributing to whether Groupon worked for the small businesses (as measured by profitability of the promotion) was employee satisfaction within the small business. (Happy employees is good businesses ... whoda thunk?!)
2. Restaurants appear particularly susceptible to negative outcomes; spas appear particularly susceptible to positive outcomes.
3. 42% of the business would not run Groupon promotions again, even though 66% of them thought it to be a profitable promotion.  ("There is widespread recognition among many business owners that social promotion users are not the relational customers that they had hoped for or the ones that are necessary for their business’ long-term success.")
4. Groupon competition will be tough. "Based on our study’s responses, the news for Groupon’s competitors appears to be decidedly bleak ... few respondents had positive things to say about other social promotion sites."
5. Social couponing is in its early days yet, with innovation likely necessary. "Although the majority of Groupon users are satisfied and intend to run another Groupon promotion, an industry in which two in five customers are hesitant after a first purchase, and where the customer base is a relatively limited pool of small businesses with strongly interconnected social networks that could quickly spread news of dissatisfactory results, may need to modify its overall strategy."

For those unfamiliar with the Groupon model, the study describes it succinctly:

Marketing circles have been abuzz in recent months with the sky-rocketing popularity of social promotion sites. At present, Groupon is perhaps the best known and certainly the largest one of these sites. It features a daily deal for each city it operates in, offering consumers a significant discount for a local business or event, such as $40 worth of sushi for $20, or a $175 facial at a spa for $59. Consumers buying the Groupon must pay its price upfront, and then have a certain amount of time, up to a year, to redeem it at the business. Groupon promotions have a social aspect. Each promotion is valid only if a certain minimum number of consumers – pre-specified by the business – purchase the deal.

NAI Research Study Validates Effectiveness of AudienceTargeted Advertising

The Network Advertising Initiative (NAI) released a study today called "The Value of Behavioral Targeting" -- derived from explicit data from 12 ad networks, including nine of the top 15 ad networks by total unique visitors according to comScore's December 2009 rankings. These ad networks shared their average CPM rates and conversion rates for run-of-network, behaviorally targeted and basic retargeted advertising. Click the image below to view a larger image ....

They reach 3 important conclusions:

• average CPM for behaviorally targeted advertising is just over twice the average CPM for run of network (RON) advertising.
• behaviorally targeted advertising converts better -- more than twice the rate for RON advertising.
• since ad networks get their inventory from Web content and services providers, this makes BT an important source of revenue for publishers as well as ad networks.

Not sure I'd fully agree with that last point -- it's depends entirely on how ad networks are working with publishers. Most of these ad networks are buying as cheaply as they can still, through auction-based ad exchanges and such where they benefit from the fact that supply is greater than demand. This results in an imbalance in the online advertising marketplace, where publishers are actually at a disadvantage. The publishers working with the Rubicon Project though are in fact seeing higher CPMs as a result of audience targeting.

See the full study here.

How Search Engines Limit Ad Inventory to Optimize Revenues

In classic supply and demand analysis, increased supply will result in reduced prices and decreased supply in increased prices -- at any given level of demand. Similarly, at any given level of supply increased demand will result in higher prices and decreased demand in lower prices. This is why CPC rates for search campaigns have become so damn high -- demand is high, and supply is being limited.

The reasonable question is "if search engines want to make the most money, then why not always list 10 ads on every search engine results page?" I always knew that the search engines (well, Google anyway) used scarcity tactics to make the most money, but I'd never actually done the math. It seems Brad Libby at The Search Agents has. You should really check out his post for the details, but here's the punchline:

Note how in his model, search engines maximize clicks with 3 ads yet optimize revenues with only 2 ads.

Google Targeting Ads by Credit Score

Neal forwarded me the MediaPost article "Google Tries Hand At Targeting Consumers With Good Credit" which describes how Google has been testing the ability to lay consumer FICO scores on top of its Google Content Network to identify/target people with good credit. Some interesting takeaways:

• will be offered to Google text and display advertisers.
• partnered with Compete.com and their panel of 2M opt-in users, which means it will not be "explicit" FICO score information but rather paneled data (Web sites which index well for high credit scores). Compete's sister company integrated the FICO data with searches done by participating consumers who applied for a credit card between January and March 2009.
• According to their marketing person, "Google's Content Network can reach 70% of credit card applicants with a high FICO score, 87% of mortgage applicants with a high FICO score, and 90% of the people who visit small business sites who have a high FICO score."
• Consumers with high FICO scores use non-branded search terms more than branded -- approximately 60% of high FICO searchers.

Top Search Advertisers on Google

Interesting stats found this morning at AdQuants, a company that does a lot of SEM analysis. They list the top advertisers for Google, Yahoo and Microsoft by number of advertisements, and by average daily ad spend.

What's most interesting to me is the volume of ad arbitrage revenues Google takes in. I looked at the top 100 advertisers on Google (by daily ad spend) and did a quick SWAG on how much is pure ad arbitrage (they advertise to get traffic, which they monetize thru ads themselves) or affiliate arbitrage (they advertise to get traffic, which they monetize thru affiliate relationships), versus real businesses (United, Nordstrom, HomeDepot, Chase, etc.). Based on this rough analysis, appears that:

• >19% of Google's ad revenues are based on pure ad arbitrage.
• >39% of Google's ad revs are based on affiliate arbitrage.
• Which would mean that the majority of Google's ad revenues are based on arbitrage! For instance, AOL, Yahoo and Ask.com are all in the top 12 advertisers.

Full lists below. And if you click on the links, AdQuants gives you more info.

Drop in Social Media eCPMs and Forecasts

Just read two separate data points regarding social networking advertising, and the struggle for advertisers to figure out how to make it work -- resulting of course in a hindrance on average eCPM rates earned by social media companies.

First I read that eMarketer has lowered their estimates for US social networking ad spend in 2008, from $1.6B to $1.4B. They claim the economy is partly to blame for this decrease, as companies cut experimental ad dollars first. As important, they claim that social networking sites are still trying to figure out what works. Net net, it's the same reason -- the ROI isn't there yet for advertisers.

Next, I read PubMatic's AdPrice Index for May which is real data (not a forecast). They found that eCPMs earned within the social networking category of sites dropped 47% , from $.37 in March to $.19 in April. Average eCPM for large Web sites (>100M pageviews/month) dropped from $.38 to $.18, and 95% of all Web sites with >1M page views per month earned less than $1 eCPM. (Keep in mind that PubMatic's data set is only remnant inventory, not direct sales.)

I've been talking a lot lately with social media companies, and they are indeed having a tough time monetizing their page views. Ironically it's such a rich source of information in terms of describing an audience and their affinities, yet it's so undervalued by advertisers. What they need are tools to help them harness that information and segment their audience into concise targeting parameters.

Social Media Research

Great stuff on social media usage, activities, trends, numbers, etc.

| View | Upload your own