Archive for April, 2009

Don’t Confuse E-mail Selling with E-mail Marketing

April 27, 2009

In E-mail Marketing vs. E-mail Sales, e-mail marketing expert and independent consultant, Jeanne Jennings, discussed how some companies are confusing e-mail marketing with e-mail selling, and thus not reaping much of the benefits e-mail marketing can offer them.

Jennings points out that e-mail marketing is focused on longer-term objectives, while e-mail sales are geared towards immediate revenue, and that companies who send nothing but promotional e-mails tend to fatigue their lists, as well as limit their audience only to customers and prospects who are already at that stage in the purchasing cycle.  I could not agree more.

Jennings reiterates what we marketers must never forget: E-mail marketing – as any marketing –  is more than selling; it’s also brand-building, relationship-building, keeping your company at the top of your customers’ minds, and exchanging information between you and your customers.  Concentrating your e-mails solely on short-term sales can cost you greatly in foregone future repeat sales that often accompany good customer relationships.

The ideal proportion of your e-mail marketing messages that should be non-promotional vs. promotional varies by industry, product category, and other factors.  However, your company can reap great benefits from a healthy mix of these two types of messages.

Sending a promotional e-mail will likely succeed if a prospect is in the buying stage.  Another e-mail that offers news and tips on how to use your product may help increase product usage and create customer loyalty.  An e-mail that illustrates the benefits of your product or service may help a prospect who is still in the needs discovery phase of the buying cycle to think of your company when he/she is ready to buy.  Sending a confirmation e-mail after a purchase – coupled with additional information –  can trigger some impulse spending by your customer.

Another useful benefit of taking a long-term focus to e-mail marketing is that link-clicks are trackable.  You can track the behavior of someone on your e-mail list when he/she opens your e-mail and clicks on a link.  This can yield valuable clues about the type of content that interests the prospect, and can help you tailor both your non-promotional and promotional e-mails to the prospects’ preferences.  When you send your prospects and customers e-mail that interests them, they believe you have their best interest in mind, and they are more likely to buy from you.

In these recessionary times, companies need to make sales.  But hard selling, whether online or off, is a sign of desperation.  Companies whose marketing demonstrates  – in every channel – that they understand and care about their customers will more than make up for today’s lost sales tomorrow.

Three Metrics for E-mail Marketing Excellence

April 24, 2009

The principles of direct marketing apply just as much online as they do offline.  The process for tracking the performance of an e-mail campaign is essentially the same as for that of a direct mail campaign.

How do you know if your e-mail campaigns are working?  Start with three basic statistics: your bounce rate, your open rate, and your click-through rate.

Bounce Rate

The bounce rate tells you the percentage of your e-mails that were returned because they were undeliverable.  If you sent 10,000 e-mails and 1,000 were undeliverable, your bounce rate is 10%.  The 9,000 e-mails that were delivered are known as your non-bounce total.

Use the bounce rate to assess the quality and recency of your e-mail list.  Eyeball the list of e-mail addresses that bounced back.  You may find that some are simply invalid addresses (“,com” instead of “.com”) which can easily be rectified.  Others may be incomplete and thus useless.  Still other addresses might be old, which suggests you should have a continuous process in place for your customers and prospects to update their e-mail addresses.

Reducing bounce rate should be an ongoing objective of your e-mail marketing strategy.

Open Rate

The open rate is the number of recipients who opened an  HTML version of your e-mail, expressed as a percentage of your non-bounce total.  The open rate can give you an idea of how compelling and attention-getting your e-mail is.  Continuing with the example above, if 1,800 recipients opened your e-mail, then you have an open rate of 20% (1,800/9,000). 

The “HTML version” and the non-bounce total are very important components of this definition.  E-mail Service Providers (ESPs) can only track HTML e-mail messages, not text.  And the use of the non-bounce total has its own share of problems, because the non-bounce total isn’t synonymous with the total e-mails delivered.

E-mails may not be considered bounced because some e-mail servers inadvertently send them to a junk folder on the recipient’s computer, which he/she cannot access.  Furthermore, if the e-mail isn’t bounced by the server, but by a portable device or software on the recipient’s computer, it will not show up in your e-mail tracking report.  Hence, you are basing your open rate on the number of e-mails sent, as opposed to delivered.

An additional problem with the open rate lies in the definition of “opened.”  Your e-mail is considered “opened” if the recipient either 1) opens it in full view or lets its images display in the preview pane, or 2) clicks a link in the e-mail.   The preview pane is a double-edged sword: If the recipient let the images of your e-mail display in the pane, your open rate may be overstated.  On the other hand, if the recipient didn’t allow images to show in the pane, but scanned the e-mail, you open rate will be understated.

You might want to use some qualitative methods to estimate the degree to which these flaws exist.  For example, a survey may give you an idea of the percentage of your customers who use the preview pane and allow images to display; a test of 100 pre-recruited members of your list to receive your e-mail (who report whether or not they received it) might give you clues into how many non-bounced e-mails weren’t sent.  This will help you place a margin of error around your open rate.

If you find your open rates declining over several campaigns, that is a sign to make your messages more compelling.

 Click-Through Rate

Your click-through rate tells you the percentage of unique individuals who click at least one link in your opened e-mail.  If, from the 1,800 e-mails were opened, 180 recipients clicked at least one link, then your click-through rate is 10% (of your opened e-mails).  It’s important that you subtract multiple clicks by a recipient (whether he/she clicked more than one link, or one link several times), in order to prevent double counting.  Most ESPs do this for you seemlessly.

Your click-through rate is a measure of how well your e-mail calls your prospects to action.  Low or declining click-through rates suggest your e-mail message isn’t generating interest or desire.

Always remember to:

  1. Track every e-mail campaign you do;
  2. Look at non-click activity (increased store traffic, phone inquiries, etc.) that occur immediately following an e-mail campaign;
  3. Look at activity to your Web site immediately following an e-mail campaign; and
  4. Track your metrics over time.  Look for trends in these metrics to refine and improve your e-mail marketing results.


Isolating the Drivers of Customer Satisfaction

April 20, 2009

In the last post, we discussed how to set up a customer satisfaction survey.  Today, we’ll discuss how to analyze the results of that survey to identify the attributes that drive satisfaction.  Let’s again assume that you’re a restaurateur and you’ve administered a C-Sat survey. 

Gather your dependent and independent variables

Your dependent variable – the variable you want to measure – is the answer to the question: “Overall, how satisfied were you with your dining experience tonight, on a scale from 1 to 5?”  Assume 5 is “very satisfied” and 1 is “very dissatisfied.”

Your independent variables are derived from the questions about the actual service delivery:  “The server greeted my party as we entered the restaurant.” This can be a point scale ranging from “strongly agree” to “strongly disagree,” or it can be a dichotomous variable: 1=”yes” and 0=”no.”  Other independent variables can be derived from questions such as: “The manager was visible on the floor,” “the food was of adequate temperature,” and/or “the server checked back with us frequently.”

Use Regression Analysis

The easiest means of identifying satisfaction drivers would be regression analysis.  Regression analysis with a few variables can easily be performed in Microsoft Excel.

When you perform regression analysis, Excel (or any other statistical analysis tool) will examine all the data points and generate an equation.  This equation will provide coefficients for each of the independent variables as well as values called a t-statistic.  Also, the equation will generate an intercept term.

Interpreting the results

Look at the t-statistic.  You want to make sure it is significant at either a 90% or (preferably) 95% confidence interval.  If the t-statistic is at least 1.645 (absolute value), it will be significant at the 90% level.  If the t-statistic is at least 1.96 (absolute value), it is significant at the 95% level.  At these levels, you can be confident 90 to 95% of the time that the independent variable is associated with satisfaction.

Also, look at the R-squared statistic – also known as the coefficient of determination.  This is a number from 0 to 1, and indicates the percentage of variation in the dependent variable that is explained by changes in the independent variables.  You want this number to be as close to 1 as possible, as it would mean 100% of the variation is explained.  If you have significant independent variables, but your R-squared is only 55%, that means that 45% of the variation in satisfaction is being explained by changes in factors you are not measuring.

Use your equation to predict satisfaction

Now that you have your equation, see how well it predicts satisfaction changes.  Using some satisfaction surveys that were not used to build the model, plug in the scores these respondents gave into the equation and see if the equation estimates an overall satisfaction score that approximates the one the respondent gave.  The equation’s predictive quality will be dependent on how close its estimates come to the actual.

Keep in mind…

  • While independent variables that are significant are associated with the changes in the dependent variable, that does not mean they caused the change.
  • You can model more than one dependent variable – but separately.  In this example, we modeled overall satisfaction.  In another case, you might want to use “how likely are you to recommend this restaurant to a friend.”
  • A high R-squared doesn’t mean your model will be an accurate predictor of satisfaction.  Some models with high R-squared predict poorly while some with low R-squares predict quite well.

“Must Haves” for Designing an Effective Customer Satisfaction Survey

April 18, 2009

Customer satisfaction surveys have unique features which make their designs and approaches differ from other types of surveys.  For one, they are much more time-sensitive (the customer must get the survey while his experience with your product or service is still fresh in his mind); their sample size is not known beforehand (the customer’s engagement with your company triggers the survey); and frequency of surveying customers must be carefully monitored, since some customers purchase more frequently than others, and surveying them each time can fatigue them and actually lower their satisfaction.

Aside from these differences, your C-Sat survey follows the same survey principles that other surveys do.  Today, we’ll discuss the ingredients for a winning customer satisfaction survey.

Remember to define  your objectives first!

As with any survey, if you don’t know what you want the information to tell you, you’re wasting time and money.  Let’s assume you own a restaurant.  Every restaurateur wants repeat customers, and higher check amounts.  Similarly, every waiter or waitress wants greater tips. 

Hence, your objective might be: “To maximize the enjoyability of my patrons’ dining experience so that they will return often, spend more, and leave greater tips for the waitstaff.”

Base the survey questions on your objectives

Now that you know your purpose for the survey, you can design the questions.  You should always ask the standard satisfaction and loyalty questions:

  • Overall, how satisfied were you with your dining experience at Sophie’s Gourmet Soups and Sandwiches?
  • How likely are you to return to Sophie’s Gourmet Soups and Sandwiches?
  • How likely are you to recommend Sophie’s Gourmet Soups and Sandwiches?

You should ask these questions first; I’ll explain shortly.  Next you should ask questions about the elements that can influence satisfaction.  For a restaurant, these can be numerous.  Recall that my March 16 post Customer Loyalty Key to Restaurants Surviving Recession cited some research that recommended asking patrons whether they were greeted by the server; whether the server checked back with them regularly; whether the server thanked them for their business; and whether the manager was visible to them.

You might also want to ask about the taste of the food; the temperature of the food; the cleanliness of the restuarant, and other features.  However, you must still keep the survey brief, especially in a restaurant, since the time patrons will be completing the survey is when they’re about ready to leave.

Your survey should ask the satisfaction and loyalty questions first because if you start out with questions such as “Did your server greet your party as you were being seated?” or “How would you rate the taste of your food,” a very bad or a very good experience with the subject of that one question can influence the answers to all the other experience questions, creating a “halo effect.”  Then, if you ask the overall satisfaction question at the end, you’ve essentially led them to the satisfaction answer.  Your findings will be biased.

Know your independent and dependent variables

Generally, your dependent variables will be the actual measures of satisfaction (overall satisfaction, likelihood of recommending to friend, etc.).  If you can tie the customer’s transaction amount to the survey, you can also use the check amount as a dependent variable.

Your independent variables include whatever drives your customers’ satisfaction.  In this case, the greetings, checking back, taste of food, etc.

You will want to determine which independent variable(s) has the greatest effect on satisfaction, so that you can adjust your customer service policies accordingly.

The next blog post will discuss how you can determine which independent variables drive satisfaction in more detail.

Last but not least!

Also remember to take into account the timing of your survey.  It should always be done while the experience is still fresh in the customer’s mind; you should not inundate the same customer with the same survey over and over again.  If a customer dines in your restaurant every day, do not expect him to take the survey each time;  it will get old very quickly.

Perhaps you can give the survey to every fifth table.  That would ensure that someone isn’t getting over-surveyed.

Also, offer an incentive for taking the survey – perhaps $5 off their next dining experience.

And, as metnioned above, try to match the patron’s survey back to his/her spending amount.  For a restaurant, the waitstaff may write at the top of a (completed) survey the check amount and the tip amount.  This will help you determine how much satisfaction correlates with customer spending.  You might also want to track back spending of customers who declined to take the survey.  This can also yield valuable clues about these customers.  

As you can see, C-Sat has lots of ground to cover.  Next post, we’ll talk about identifying the drivers of satisfaction.

Customer Satisfaction: Your Most Cost-Effective Marketing Tool

April 9, 2009

When many firms do marketing, they spend thousands, if not millions, of dollars on advertising: newspaper, radio, television, billboards; anything that builds awareness of their product or service and gets the customer to buy it.  Then, in bad economic times such as these, advertising and marketing are the first things they cut.

One would think that when customers aren’t buying, marketing spending should be stepped up to keep the company’s products and services front-and-center in the customer’s mind.  Yet, the companies all say that the results of their marketing were too costly to justify the costs.

Whether or not that’s true – some don’t get the full potential of their marketing because they do it wrong – most companies overlook what may be the most cost-effective marketing medium available to them: a satisfied customer!

Acquiring a new customer is, on average, six times as costly as retaining an existing customer.  That alone makes customer satisfaction an essential marketing tool.  Another reason – a dissatisfied customer will not only stop doing business with you, but will also complain to about 9 other people.  (When I’m dissatisfied I reach 9 easily: I have 8 older siblings!)

When a customer is dissatisfied – and his or her friends and family have heard the rants against your company – no amount of good marketing will help.  The deck is already stacked against you in their minds.

So how do you find out whether your customers are satisfied?  Simple.  YOU ASK THEM!  Call them up after a sale.  Encourage them to be straightforward with any criticisms they have – you cannot fix a problem you’re not aware of!  Send out a survey.  Invite your customers for a discussion group.  Talk to them any way and any time you can.

Just the fact that you show interest in your customers’ experience with your product will set you apart from your competition.  While a satisfied customer may not tell 9 people how satisfied he/she is with you, that customer does tell people.  That’s why C-Sat as part of an integrated marketing strategy is useful.  Give your customer a satisfying experience, keep him informed of your company, and then when one of his/her friends asks “Do you know of a good…,” the customer instinctively thinks about and tells the friend about your company! 

Over the next few blog posts, I will discuss designing an effective C-Sat survey, identifying the key areas you need to act on, and building an effective customer feedback system.