Archive

Archive for the ‘Email Marketing’ Category

Email Marketing: Groupon’s segmentation strategies across 115 million subscribers

September 20th, 2011

Groupon is one of the biggest email marketing success stories you’ll ever hear. The daily deal company launched in November 2008 and was dubbed the fastest growing company ever last year by Forbes. Much of that growth is powered by targeted email messaging.

Today, Groupon sends about 5 billion emails each month, according to John Becvar, Sr. Director of Relationship Marketing at Groupon. I was lucky enough to hear Becvar comment on the company’s growth and email strategy during a panel session at ExactTarget’s Connections 2011 last week.

“We really began with no personalization and no segmentation,” Becvar said.

Becvar joined Groupon almost two years ago, back when its site featured a single deal each day in markets such as greater Chicago, with no deals on the weekends. Today, Groupon offers about 1,000 deals in the Chicago area in a given week, Becvar said.

“When I joined we had about 2 million subscribers. Based on our most-recent finding, we have over 115 million subscribers … We were in 20-some-odd markets when I joined. Now we’re in 45 countries and 500 markets … So we’ve obviously had to do a little bit of segmentation and personalization to stay relevant,” Becvar said.

Becvar described four types of segmentation Groupon uses for its email marketing: Read more…

Email Marketing: How Microsoft increased product engagement using email

September 15th, 2011

New products with lots of features can be intimidating. A learning curve has to be overcome before customers can fall in love and recommend the product to a friend. And I can think of few products that are as feature-laden as Microsoft Office 2010.

After teaching about email marketing at ExactTarget’s Connections 2011 on the latest stop of MarketingSherpa’s Email Marketing LEAPS Advanced Practices Workshop, I’ve been soaking in all of the email marketing content presented here. Diana Lawson’s case study really stood out to me because it was a great example of how to use email to enamor customers with a feature-rich product.

The senior product manager at Microsoft presented a case study yesterday describing how her team launched a lifecycle email program to build loyalty among Microsoft Office 2010’s newest customers.

“We need people to get the value of Microsoft Office right away… So there’s no buyer’s remorse, and more a feeling of ‘I am so glad I bought this,'” Lawson says.

The emails deliver targeted how-to content and videos for five months. One of their main goals is to improve relationships with customers by offering helpful content from team leaders behind Office 2010.

Here’s how the lifecycle emails work (results below): Read more…

Email Marketing: A toxic misunderstanding that could kill your response rates

August 30th, 2011

Internal challenges are among the hardest to overcome when trying to improve your email marketing. You can know everything about improving results, but if your leaders are unwilling to commit the resources, or have a fundamental misunderstanding about email, then you have some convincing to do.

One of the most common misunderstandings I have heard at our Email Marketing Workshops is that some companies “do not believe in” suppressing parts of their list. The leaders essentially force the marketers to batch and blast, thinking it is always better to have more eyeballs on a message than fewer.

This, however, is a huge misconception. Email marketing is not advertising. If you send a message that is irrelevant to a large portion of your list, then you are encouraging recipients to click “spam” and ignore you in the future. Practiced consistently, this can be toxic to your database. Read more…

Reader Mail: Understanding differences in clickthrough rates and open rates

August 12th, 2011

Recently, my colleague Brad Bortone forwarded me an inquiry from one of our readers, who asked the following:

Can you provide any insight into why my newsletter emails would receive a 10% unique CTR and a 3% open rate? Aren’t open rates generally the larger number?

We use XXXXXXXX as our email service provider. Could this be related to how our newsletter renders in the preview pane of email clients?

In thinking about this, I realized that many email marketers may be asking the same questions, and could benefit from an extensive reply. Besides, I don’t get much mail around here, so I was excited to help out.

Here is what I wrote in my initial reply: Read more…

Email Marketing: Dynamic images get 5 opens per reader on a single email

August 9th, 2011

Email marketers, start your engines! The MarketingSherpa 2012 Email Awards contest is now under way. Enter today for your chance to win this year’s grand prize – a paid trip for two to Email Summit 2012 in Las Vegas. The contest is sponsored by Responsys, and entry is free-of-charge.

I recently received a stark reminder of the types of emails we look for in our “Best Email Innovation” category. Anna Yeaman, Creative Director at Style Campaign, described to me a technology her team developed that can dynamically update an email’s images at the time of opening.

For example, Style Campaign tested this system last year by sending a holiday email to its subscribers. The email gradually revealed a fun holiday image over the course of the day. Subscribers could see a little more of the image each time they checked the email, even if they had previously checked a few seconds earlier. The campaign featured a new image each day for five days.

“This was the first deployment of our dynamic imaging technology. We really wanted to test the whole proof-of-concept and see if we could get users to re-open the same email again and again over a period of time,” Yeaman says.Style Campaign Dynamic Image Mobile Email

The test proved successful. Stats include: Read more…

Email Marketing: Good preference centers can hold onto subscribers even as they change

August 2nd, 2011

One of the most interesting things about life is that we change. Our interests change. Our habits change. Our jobs and locations change. So it goes without saying that our email subscribers will change. The question is whether you can change with them.

Now, I am not advocating that you launch a “singles” newsletter every time a subscriber gets a divorce. However, it can be helpful to make your email program flexible enough to accommodate subscribers’ changing needs.

How do you make your email program more flexible? You can start by having a good preference center that clearly explains your options. Here’s a good example from AAA (click this link to visit the full preferences page): Read more…

Email Marketing: How to capture high-quality subscribers and avoid an inactive list

July 26th, 2011

Remember when you were a kid and thought your parents were getting you the greatest present in the world? Maybe you were expecting a BB gun or a pony, but they got you socks instead? Or some other lame present?

Think back to that exact moment when you saw the present and thought “How did they fit a pony into that tiny box?” Then you opened it, blew out your candles, and wished for new parents.Present expectations

That feeling of crushing disappointment should remind you of what it is like to have your expectations overlooked. That memory can help you improve the quality of your email database.

When people opt-into your emails, they have expectations of what they will receive. Your subscribers will probably not throw a temper tantrum and hurl cake on the walls if you miss their expectations. However, they will be more inclined to ignore your emails and mark them as spam, which undermines your email marketing.

The solution is to very clearly explain your email program at the point of opt-in. This will correctly set people’s expectations. They can decide if your emails have something they want. If they want it, great; they can sign up and become active subscribers. If they don’t want it, great; they will not sign up and not add to the dead weight in your database.

Here’s a good example of how CIO, an IT-management publication, sets expectations for its newsletters (click the image for a larger view). Read more…

Email Relevance Kaizen: 4 categories of data beyond the email database

July 19th, 2011

We reported in MarketingSherpa’s 2011 Email Marketing Benchmark Report that the biggest challenge marketers’ face is targeting recipients with highly relevant content. I do not foresee this moving out of top tier of challenges due to the complexity of reaching a unique subscriber with the exact information they need to help he or she succeed

Crafting an email message that connects with the reader at the moment he or she choose to open the email is difficult. Nevertheless, the alternative of sending irrelevant emails puts a brand’s image, reputation and subscriber relationships at risk.

Relevancy is #1 Challenge

Today, some 70 million US consumers access their email inboxes through a mobile device, and subscribers’ bottomless appetites for multi-tasking, whether it be on a phone, tablet, or laptop during office meetings or in front of the television, make it impossible for companies to capture a reader’s full attention. To break through the clutter, email messages must be easy to engage with, relevant and clear.

A company that sends timely email messages with pinpoint accuracy will not only stand out from the competition, but may be the only brand to make it to their subscribers’ inboxes.

Webmail clients Hotmail and Gmail are using engagement metrics (open, delete, reply, etc.) to determine when and where email is delivered. B2B marketers also need to keep these trends in mind when planning their email campaigns. HubSpot found that 88 percent of email users do not have a separate work and home address. Many employees like to use webmail addressesto have the ability to share larger files over corporate email.

To meet customers’ rising demands for relevant and engaging communications, marketers can focus on three key elements:

  • Content
  • Segmentation
  • Time

All three are essential. If one element is missing, the email can become irrelevant to the reader. Let me show you a recent first-rate email series that fell just short of the goal for my family.

Moving beyond a single data point for determining relevancy

A year ago, my wife joined the Chuck E. Cheese’s email club to receive coupons and free tokens for our kids. In exchange for her email address and dates of our children’s birthdays, we receive emails twice a month promoting their latest coupons or specials. The emails nicely tie into holidays, school calendars, or kid activities, like receiving a bonus of tokens for cleaning their rooms.

As you might expect, with the information of my children’s birthdays, Chuck E. Cheese’s sends a few emails leading up to the big day. About thirty days before my son’s birthday we received this email with the personalized subject line “[Parent’s name], Get 100 Bonus Birthday Tokens for [Child’s name].” I did not speak with Chuck E. Cheese’s, but from the numerous party invitations my kids have received from their classmates starting in preschool, these emails are very successful at enticing parents to book birthday parties at Chuck E. Cheese’s locations.

The week of my son’s birthday, my wife received an email with the subject line “Happy Birthday to [Child’s name] from Chuck E. Cheese’s..” The principle message was to wish my son a happy birthday and encourage us to stop in and play some games with complimentary tokens.

Close behind and at the top of the email was the copy “Haven’t planned a party yet for [Child’s name]? No worries, it’s never too late to reserve a birthday party at Chuck E. Cheese’s. Last minute walk-in parties are always welcome, so come on in and celebrate!”

As you might agree, these are excellent emails containing personalized and relevant information, and are delivered at appropriate times leading up to a child’s birthday.

So why did they fall short of relevancy for my family? Well, I guess you would classify us as outliers; we planned our son’s birthday 90 days out at an indoor gymnasium. (Full disclosure: Chuck E. Cheese’s does send half-birthday email cards, but we are not that organized, and yes, in the copy, Chuck E. Cheese’s promotes walk-in half-birthday celebrations).

I am confident Chuck E. Cheese’s identified two groups of parents/guardians with their internal sales data: a profile of party purchasers that plan within 30 days and another that plans the week leading up to the event. Combined with the child’s birthdates collected during email registration, Chuck E. Cheese’s sends highly relevant and targeted communications with these two types of buyers.

Moving forward, Chuck E. Cheese’s may choose to expand their email program to cater to “overly-organized” parents. An email message sent 90 days prior to the celebration, with more highlights on specific cake, food, and party favor options may appeal to this customer profile who enjoy delving into the details of the planning process.

Mining for data

For marketers, the quest for relevancy and engagement is not complete until a one-to-one relationship is established. Due to budget and resource constraints, this is an unrealistic goal for most companies. That should not discourage marketers from moving towards one-to-fewer, even if their budget doesn’t allow for a true one-to-one relationship to develop. Remember, with each successful email campaign targeted to a select customer profile, brands can expand their reach and look at new segments of their customer base to exploit, such as outliers like my family.

To do that, organizations need to go beyond traditional demographics like birthdates to understand their customer’s motivations. I do not think you would ever want to have an email preference center with a checkbox to identify your customers as an early bird, prompt, or fashionably late party planner. But that doesn’t mean such valuable information is out of your reach.

In addition to your email database, mining all of the in-house data found inside CRM, sCRM, web analytics or financial software systems may be beneficial. When you begin, search four categories of data; endemic, transactional, behavioral and computed.

Endemic data is data unique to a particular record. Examples are contact and demographic information commonly collected during registration or  in a brands preference center.

Transactional data consists of data captured during any transaction a customer has with an organization. Financial software is the most common source of transactional data, but transactional data can also encompass a subscriber’s IP address.

Behavioral data is perhaps the most actionable of all database information and can come from a wide variety of both online and offline sources. Website clicks, email opens, calls to customer service, and redemption of coupons at store locations are all examples of behavioral data.

Computed data is the outcome when one or more variables are used to create a third variable.

For example, the variable of “number of miles from a retail store” is computed data, and is the difference of the distance between the customer and store’s address.

Mining these four types of data will uncover insights into your customers and understand the steps they take to make a purchase. At MarketingSherpa, we recently launched our annual Email Marketing Benchmark Survey to identify the barriers that exist in preventing subscriber engagement and the best practices to increase the velocity and accuracy of email communications. I encourage you to participate to help us learn more from marketers like you who are in the trenches, battling for relevancy every day. In exchange for your valuable time, we will provide you a complimentary copy of our Special Report: CMO Perspectives on Email Deliverability.

Related sources

Email Deliverability: How a marketing vendor with 99 percent delivery rates treats single opt-in lists vs. double opt-in lists

Email List Growth: Finding low-cost and no-cost ways to grow your database

Email Marketing: Three lessons learned at the MarketingSherpa Email Marketing LEAPS Advanced Practices Workshop

New Chart: Most effective email list growth tactics

Email Deliverability: How a marketing vendor with 99 percent delivery rates treats single opt-in lists vs. double opt-in lists

July 8th, 2011

Yahoo! Mail, Gmail, AOL and similar providers track senders’ reputations by IP address. This is partly why sending emails from dedicated IPs (instead of IPs you share with others) is the most effective tactic for improving deliverability, according to the MarketingSherpa 2011 Email Marketing Benchmark Report.

Many smaller companies, though, cannot manage a dedicated IP well enough to build a strong reputation and are better off using shared IPs, says James Thompson, Email Systems Manager, Infusionsoft.

“What we found is that most small businesses really don’t have the resources to be able to dedicate the kind of attention and maintenance required to maintain a good status on a dedicated IP,” he says.

Thompson oversees the sending of about 40 million emails each week for clients at Infusionsoft, a marketing automation software provider that caters to small businesses. He is tasked with maintaining the email architecture and deliverability for clients across the system.

Thompson has been through the trenches of email deliverability and helped to pull the company’s average delivery rate from the mid-90-percent range two years ago to consistently above 99 percent today.Email Three Funnels

Thompson shared several stories on how his company handles deliverability. His examples can help email marketers of all sizes understand how their reputations can extend beyond webmail providers and ISPs and begin to affect their relationships with agencies and vendors.

Three tiers of email IPs

One of many changes Thompson’s team made to improve deliverability was to shift its email architecture onto a message management platform. The new platform allowed the team to establish three groups of IP addresses to send from:

Group #1. Transactional emails

This first group of IP addresses was used to send invoices, order confirmations and other transaction-based emails for Infusionsoft’s clients. Thompson’s team wanted to isolate these emails because they had some of the best performance metrics of any email type, and because getting them delivered was critical to Infusionsoft’s clients.

Group #2. Single opt-in lists

This group of IP addresses was reserved for client lists that were built using unconfirmed- or single-opt-in tactics. The team isolated this group because its lists generated a higher number of spam complaints. Grouping them would prevent the complaints from hurting the reputations of IP addresses that sent emails to more qualified lists.

Group #3. Double-opt-in lists

This final group of IP addresses was reserved for email lists that were generated with confirmed- or double-opt-in tactics. These lists generated fewer problems than single-opt-in lists, and therefore benefitted by being associated with each other rather than with lists of lower quality.

Why this matters:

Thompson noted that the delivery rates between Group 2 and Group 3 originally differed by as much as 5 percent, but is now down to about 1 percent.

“However,” he says, “we are talking about averages here on millions of emails, so that does add up.”

So even if your company is relying on a shared IP address from a provider, it can still be in the best interest of your deliverability to build high-quality lists with confirmed opt-ins.Baseball batter

Three strikes and you’re out

Thompson’s team realized it had a few bad apples in the barrel when it began work to improve delivery rates. The team reached out to clients who were hurting the sender reputation of Infusionsoft’s IPs and tried to enlighten them on the issue.

“What we have come to find is that most of our customers who generate these spam complaints don’t necessarily understand or aren’t aware that their practices are abusive,” he says. “So essentially what the solution has been is to work with them, give them a set of criteria to meet, and then the problem is usually resolved.”

Unfortunately, not every situation was resolved. Some clients were unable to make the necessary changes and the team had to let them go to protect the reputations of other senders who used the system responsibly, Thompson says.

Now the team has a three-strikes system for handling clients that generate a high number of complaints.

“AOL, Yahoo!, MSN and those types of companies want to see average complaint rates of about 0.1 percent,” Thompson says. “So that means for every 1,000 emails we send, they only want to see one person complaining.”

Thompson’s team uses this same guideline when monitoring its system. Senders who regularly generate two or three complaints per 1,000 emails sent (0.2% or 0.3%) are given suggestions to improve their practices. Clients who fail to respond or make changes can be asked to leave (though this is rare).

So if you’re using a shared IP address to send email, the impact of a complaint can extend beyond your email program and even begin to impact your relationship with your marketing vendor — so keep those rates down!

Related resources:

Members Library – Email Marketing: How Publishers Clearing House uses “blacklisted” words yet achieves a 99.2% delivery rate

Members Library – Webinar Replay — Improve Email Deliverability: Tactics for Handling Complaints and Boosting Reputation

Email List Hygiene: Remove four kinds of bad addresses to improve deliverability

Email Deliverability: Riddles answered on spam complaints, feedback loops, and dedicated IPs

Message Systems – the message management platform used by Infusionsoft

Email List Growth: Finding low-cost and no-cost ways to grow your database

July 1st, 2011

A common challenge we’ve heard at our Email Marketing Workshops is that marketers want to know how to effectively grow their email lists. This was true at the Seattle and Boston workshops, and I expect it will continue through Washington, New York, San Diego and Austin.

One tactic we describe is how to leverage existing channels between your brand and your audience to grow your list. Marketers can study how their audiences receive information from their brands and test adding an opt-in request to those channels.

For example, the New York Public Library used its transactional emails to grow its list. The library already sent automated emails to remind patrons that books were due. The marketers tested a simple addition to these emails:

“Get the latest NYPL news and events. Sign up for our newsletter at: [shortlink].”

This two-sentence request increased the library’s opt-ins by 120%.

More than transactional emails

I really like the above example. It shows how a simple idea in an existing channel can grow a database at nearly no cost.

Your company is not likely to be a library, but this idea extends beyond late notices for books. I saw another example while flying to our Email Workshop last week.

Delta In-Flight Email Opt-in

I took this picture on a Delta flight from Memphis to Seattle. I was poking through the touch-screen personal television at my seat and noticed, lo and behold, that Delta wanted my email address. As you can see in the second sentence, the page does a good job of emphasizing the value I’ll receive by handing over my information.

Here’s the next screen:

Delta In-Flight Email Registration form

I am not very sharp on the history of in-flight televisions, but I assume these screens were not installed solely to collect travelers’ email addresses. Instead, I assume that the opt-in request was added later at the suggestion of a clever email marketer. This is another example of a company leveraging its existing channels of communication to build its list.

Look for relevant audiences

In our Email Marketing Workshop, we spend the first half of the list-growth section emphasizing the importance of list quality and how it can be improved. We really hammer on the point that the size of a database is not as important as its quality.

For example, a large database that is loaded with disinterested subscribers is not going to help your marketing nearly as much as a smaller database with subscribers who are interested in your content and offers. You want high-quality subscribers who enjoy receiving your emails and clicking your links.

One way to help build a quality database is to invite only relevant people into your email program. The two examples mentioned above do just that.

  • The New York Public Library’s transactional emails reached people who had books checked out. These people had visited the library, so they might have been interested in receiving news about the organization.
  • Delta reached people in-flight on a Delta plane. These people were travelers, and they traveled with Delta, so they might have been interested in receiving offers and updates from the airline.

So when you’re looking to your existing channels to help grow your list, be sure to test the channels that reach the most relevant audiences for your email program. You don’t want to invite just anyone into your database. You want to build a high-quality list.

Related resources:

Email Marketing LEAPS Advanced Practices Workshop

Members Library — Growing Email Lists with Social Media: KFC’s Facebook tool adds subscribers

New Chart: Most effective email list growth tactics

Maximizing Email List Growth: How the New York Public Library drove a 52.8% lift in newsletter subscriptions

Email Marketing: Three lessons learned at the MarketingSherpa Email Marketing LEAPS Advanced Practices Workshop

B2B Marketing: Building a quality list