"Free Gas!": Make sure your incentives line up with the behaviors you want to encourage

Many digital marketers who are trying to sell products use affiliate programs as a major strategy.  Essentially most affiliate programs work by paying websites a commission on each referral they make to a merchant that results in a sale.  Affiliate programs can be great for increasing the exposure for your products or store, and they can be a win-win for the marketer and the websites who participate.

But when setting up affiliate programs -- and many other pay-for-performance programs like them -- marketers should think through their entire strategy carefully, to ensure they're not encouraging the wrong behaviors or creating a slanted playing field.  This can apply to sales commissions if you manage a sales force, not to mention pay-per-action advertising campaigns, and even promotions/discounts offered to your customers.

Here's an example of a discount program that's poorly thought out on a couple different levels: Giant Eagle grocery store's fuelperks! program.

Giant Eagle is a grocery store with locations in Ohio and three other states.  The company also has a number of gas stations called GetGo, many of which are co-located with their grocery stores.  A few years ago, it introduced a program called fuelperks!, where customers receive a 10-cent-per-gallon discount on their entire fuel purchase for every $50 they spend on groceries.  The discounts are valid for three months and can be accumulated, so it's reasonably easy for a regular Giant Eagle shopper with a large family to build up enough of a discount to earn an entirely free tank of gas, or at least a very cheap one.

While the fuelperks! program's overall incentive (getting people to spend more on groceries to earn more fuel discounts) is good, Giant Eagle seems to have missed a few details in the way their program is set up -- thus inadvertently giving incentives for the wrong behaviors:

• The program gives disproportionately large discounts to shoppers who drive large, gas-guzzling vehicles.  For example, let's say you have two consumers who each spend $2,000 over three months at Giant Eagle on groceries, and the price of gas is $4.00 a gallon.  The $2,000 in spending qualifies them for a $4.00 per gallon discount -- so their gas is effectively free.  But consumer A drives a Chevy Aveo with a 12 gallon gas tank, while consumer B owns a Ford Expedition with a 28 gallon tank.  Consumer A walks away with a free tank of gas worth $48, or a 2.4% discount on their $2,000 grocery purchase.  But because consumer B drives a bigger vehicle, they get a free $112 tank of gas, which is effectively a 5.6% discount off their groceries.
  
  Since the program rules state that your discount applies to a single tank of gas up to 30 gallons, Giant Eagle is giving a disproportionately large incentive to people with huge vehicles.  That doesn't seem like an environmentally friendly behavior to encourage!  (I'm shocked environmental advocates haven't given Giant Eagle a hard time about this yet.)
  
  
• The fuelperks! program also encourages widespread cheating by ordinary people.  In the case above, the Aveo owner might want to get as much free gas as possible, so he cheats.  Nearly every time I've visited a Giant Eagle GetGo gas station, I've seen at least one customer cheating -- by bringing several gas cans to fill up, or by having two cars fill up their tanks on the same transaction.  These are clear violations of the rules, but I've never seen anyone enforce those rules in my many visits.  Widespread cheating of the system takes place, and often I'll see multiple cheaters on a single visit.
  

While the fuelperks! program has been a big success for Giant Eagle, I wonder how many honest, rule-abiding consumers get frustrated watching people abuse the system.  I wonder how many people with small gas tanks get frustrated when they see giant SUVs getting more than double the savings they're receiving for the same amount of grocery spending.  For these reasons, I hate going to GetGo!  (My wife and I stopped shopping at Giant Eagle, so it's rare that we earn much of a discount anyway).

So here are some takeaways for marketers who are setting up an affiliate program or a discount program:

• Make sure you're offering incentives for the right behaviors.  Examine your margins.  Are you encouraging behaviors or purchases that will drive short-term revenue, at the expense of long-term profitability?
  
  
• Think through all facets of your offering, and ensure there aren't any unintended side effects.  Giant Eagle might be driving the purchase of more groceries, but that incentive sprouted some ugly unintended consequences, like rewarding gas-guzzlers and encouraging customer cheating.
  
  
• What is the effect on customers who purchase through the program?  Do they have the same lifetime value as other customers?  In some cases it might be OK to take on less profitable customers to boost volume, but be careful with this strategy.  You might end up with a whole different problem on your hands, if too large a segment of your customer base will only respond to special offers.
  
  
• Watch for cheaters.  Especially with affiliate programs, you will have a certain percentage of sites who will try to beat the system.  They'll look for loopholes, try to use underhanded tactics to increase their revenue (spamming, search engine manipulation tactics, etc.).  These people don't care about your brand, your reputation, or your customers -- they're just trying to make more money for themselves.
  

Here are a couple old but good articles from ClickZ that talk about similar issues:
Rewarding Profitable Behavior
Free Shipping and Handling Aren't Free

Google's Timeline feature in beta

It's been in beta for quite some time -- more than a year already.  But Google's Timeline feature, currently part of Experimental Search, is an interesting technology that might someday make its way to mainstream search.

The timeline parameter is meant to show search results sorted by time period.  So for example, let's say I want to find information on the computer chip maker Intel.  In my search box, I'll type:

intel view:timeline

And it shows me results sorted by year (see full-size screenshot).




When I click on a year, it takes me to results sorted by year.  It extracts these results from each page.  Pretty cool, huh?  Google's explanation says it works best for people, companies, events, and places.

From a marketing and PR standpoint, if Timeline ever gets rolled out as a mainstream feature in Google, it could change the way we think of search results and SEO.  It instantly becomes easy for someone to look back at your company's history and see what was being written about you in the past.  That might be a good thing, but it could also be a bad thing if you have something to hide.  With this new "time dimension" to search, it instantly becomes harder to bury bad publicity you may have received a while back through search engine optimization.  A simple timeline search could turn up the page you were trying to keep quiet.

Right now Google's Timeline feature isn't quite ready for prime time.  Although it's pretty good, there are still plenty of problems with Google recognizing dates, or figuring out that a company name on that page isn't relevant to a date.  For example, let's look at Intel again, but let's take a peek into the company's distant past...800 years ago!  When you set the filter to the date range 1200-1300, you still get a ton of results (see full-size screenshot):




In the top result from Wikipedia, Intel created a chip called the 1201, and Google thinks 1201 is a year.  Then the third result is from a court case in 2002, but the PDF happened to have the number 1200 included on one of the pages.  Oops.

I don't know how Google will figure out how to differentiate dates from numbers.  It seems like a big obstacle for timeline search.  But if it's a high enough priority, I'm sure Google can find a way to do it, and make timeline search a mainstream tool.

Additional thought: Google's new business video tool

Here's a relevant comment about Monday's post that I received from Ben at Wistia.  Now that he mentions it, I'm a little surprised Google didn't build some sort of analytics into their new business video offering, since most programs are so metrics-based nowadays.

I believe that there is one area which you didn't hit on which is one of the larger failings of the Google offering: tracking.  With Google's YouTube mindset of user-generated content, tracking and analytics has largely been ignored.  However, the reality is that companies are spending thousands, if not tens of thousands, of dollars on creating professional video with in-house departments and third party production companies.  With this level of expenditure, they need to know more than just "23 people watched this video" in order to justify their costs and recognize a true return on their investment.  In a sales, training, or marketing environment, having full transparency into how specific people are viewing your video content is extremely valuable.

Google's new business video tool

Last week Google made big news when it launched Chrome, its new web browser. In a blog post on Sept. 3, the New York Times pointed to the potential death of the search toolbar, thanks to Chrome. But while Chrome is an interesting new development in the browser wars, I don't think it will have any major effects on digital marketing, at least not immediately.

Google also put forth a quieter launch last week -- its new Google Video for Business as a part of Google Apps. It's meant to be an enterprise solution that allows companies to share videos internally. Some of the suggested uses are corporate announcements and internal training. I could also see it being used for sharing of best practices and video newsletters. Because it's made to be secure from the prying eyes of competitors and the public (unlike posting videos on YouTube where a video is accessible to the world), Google Video could enable companies to accelerate their use of video internally.

According to the Google Video for Business site, "no large files or complex infrastructure" are required. It looks like the new product works similarly to YouTube works for mainstream video -- just a simple upload of a video file and it's available within your company. This type of simple, turnkey solution might make sense for smaller companies, and maybe even the smallest of mid-sized companies. But I'd expect the vast majority of bigger enterprises already have streaming servers in place to host videos. Google Video for Business isn't anything new or wonderful for them -- in fact it might even be a step backwards because it forces companies to sacrifice control. Only videos of a certain size are allowed, so big videos won't work. And on top of that, if you're not using Google Apps across your company, you can't share across your company. Since I doubt the market penetration of Google Apps among midsized and large companies is very big, Google Video for Business won't gain significant adoption in this market segment.

If you're a smaller company that's already using Google Apps, the new Google Video for Business offers some great tools for pushing the adoption of video across your company. If you're already considering deploying Google Apps for other reasons, the addition of video support might be another minor attractive feature as you're making your decision. But outside of these two situations, I don't expect Google Video for Business to win over too many people.