Reflections on and impact of our own email ‘oops’

Yesterday’s email oops from the New York Times brought back seriously bad memories of a time when an email newsletter meant for 25,000 got sent to 2.5 million on our “all subscribers” email list.

For us, it was a simple training issue. The main email “sender” went on vacation and trained another user in her department to send the weekly email. The documentation wasn’t detailed enough and the replacement clicked the “all subscribers” list instead of the subscriber list associated with the newsletter. BAM! We’ve since had the “all subscribers” list removed from the send options, but the damage was already done.

Here’s a rundown of what happens when an email meant for a few thousand people gets sent to millions instead.

Deliverability issues when users mark as SPAM:

When a user marks a sender as SPAM, the email service provider (in our case, ExactTarget) makes that address “inactive.” Once those users hit that “master unsubscribe” list,  that means no more email newletters, breaking news alerts, etc… massive list depletion. Even if further emails do reach that user, you may end up directly in the user’s spam folder instead of where you want to be: the inbox.

In our case there was a significant impact on most of our email newsletter lists, up to a 30% drop in active subscribers for some of our smaller/niche newsletters. When you sell display advertising in your newsletters based on list size, a significant and rapid decline can crush future sales.

Budget buster:

Large-scale emailing isn’t exactly cheap when you send through an ESP. Generally, companies purchase yearly email allotments and pricing is based on the volume of email you send. A mistake of a million emails could cost thousands of dollars if you burn through your yearly allotment ahead of schedule because of it. If you send a correction email, that doubles your cost. Luckily for us, we did not end up paying for overages but we were close.

Reputation killer:

I’m not talking about send reputation based on spam complaints and deliverability, either. I’m talking about the trust you’ve worked so hard to build with your customers. Breaking the trust of a user is a serious violation; it’s a fragile relationship where one wrong move can cost you the customer forever. With rampant data breaches, spamming, phishing and everything else going on, users are always on high alert to bail at the slightest whiff of something foul.

So, if you have others in your organization who send email, check in with them regularly to make sure they understand proper workflows but also the consequences of carelessness or misunderstanding. Make sure any new hires are trained thoroughly on email sending and be available to answer any questions they have. From experience, taking the time for proper training is a much better option than the giant mess like this.  It shows you even an industry leader like the New York Times can make big mistakes and one errant click can cost you thousands.


Things I learned in 2011: Litmus is a kickass tool

LitmusIf I were Baby Jesus and the three wisemen were bringing me gifts, I hope one of them would be Litmus (the basic $49/month version will do!). In contrast, the other two wisemen should bring Super Turbo Trains and an endless supply of Chicago-style pizza, respectively.

Previous to the joyous gift that is Litmus, email client testing at my company went something like this:

“Is there a GMail account on this test list? Yahoo? Outlook? Sweet.”

Inexcusable, I know. But so was the organization of our email “program.” I’ve worked tirelessly this year to make us respectable.

So, why do I love Litmus? For starters, it is so easy to use.

Here is my quick and dirty version on how to run a test (after you have your account):

  1. Login to Litmus
  2. Click “New email test” button
  3. Select all email and mobile clients you want to test with (you can “select all” for desktop clients, mobile devices and web clients instead of selecting individually).
  4. Click the “start test” button
  5. Copy the testing email address provided and send your test email to that address.
  6. Click the “I’ve sent it” button.
  7. View results in  all selected clients by clicking the thumbnail images. Or, cycle through clients using the left or right arrow keys after clicking on one of thumbnail images.

That’s it! Seven simple steps to see how your email looks in nearly all available (relevant) email clients.

Or, you can skip steps 2-6 and send your test email to your account’s static email address. Your test is created using the default email clients for your account. It would take hours to just create email addresses at all the web clients and create test lists alone.

Another reason to love Litmus: the price! We use the $49/month option and it is perfect for us.

Are you spending 4,5,6, even 7 figures for your email service provider? What kind of rendering tools do they offer? With ExactTarget (I do love them, though), we get nothing like Litmus in the rendering department. The “send preview” option is clunky and complicated. If you think you can “get by” with that preview tool, think again.

So, I’m sure your budgets are already set for 2012, but beg and plead to your manager that you NEED Litmus. You even get 2 months free when paying for the entire year ($490). You won’t be sorry!

Coming tomorrow: Things I learned in 2011: Get a go-to email coder (that isn’t you)

The Power of Connections

It’s easy to see why interactive marketers would want to go to Connections: Top-notch keynote speakers and entertainment, the food, the production, the soundtrack (available on iTunes and Spotify), the inspiration! Connections is a first-class event.

While all of those things distinguish Connections from other events in its class, most of the 3,000+ marketers really came to learn from top industry experts in a variety of disciplines including social media, email deliverability, mobile marketing, analytics and design.

Like many other attendees my company pays for me to go to Connections, and I have tried each year to focus my experience around helping my company make giant leaps forward in email marketing.

Last year — my first Connections — was about learning the ropes. I was new to my role as email marketing product manager; I’d never been to a conference of this magnitude. In the past, I had been a light user of Exact Target software, but I was thrust into being in charge of nearly all aspects of our email communication with consumers. Quite a shock to the system.

At Connections 2010 I focused mainly on the “101”-type sessions, getting a better handle on the basics and learning best practices. I came back with pages and pages of notes, much to the detriment of my co-workers, whom I nagged multiple times to read them and discuss.

This year, I wanted to take the next step in my marketing education while also getting inspired by the strong lineup of keynotes. I was excited for the Day One lineup of speakers which included Exact Target’s Scott Dorsey and Scott McCorkle,  author Aron Ralston and Wikipedia Founder Jimmy Wales.

I won’t recap details of the keynotes but the lineup was a great mix of technological and emotional inspiration. My favorite parts were seeing a demo of the Interactive Marketing Hub and listening to Aron Ralston’s story of triumph. I also got into a discussion with Jimmy Wales in the elevator about how much Wikipedia spends on server space (only $3-to-$4 million!) each year.

While my company will not use the Interactive Marketing HUB in a way that will take advantage of its best features (yet), it is exciting that Exact Target has invested so heavily into developing a revolutionary product like it. I’m looking forward to trying to evolve our email program into a cross-channel program that could utilize the HUB in the way it is intended.

My co-worker Kim and I striking a pose at Connections

I must say my breakout sessions on Day Two were hit and miss. When I scheduled them, I looked for sessions pertaining to my use of the Exact Target software and my job duties. I also wanted to avoid partner demos and integrations because there is definitely no room in the budget to add new products to the mix. So why were they hit and miss? I think it was because of my own knowledge explosion in the past year, spurred by last year’s conference.

The first session I went to was about email design. Most of the presentation discussed Subscribers, Fans and Followers research on how users interact with email, most of which I’ve become very familiar with in the past year.  As a user who is building emails and trying to improve email designs, I was craving more examples and insight I could take back and incorporate into my own designs.

My favorite session of the day dealt with deliverability. It was called The Death of Inbox Neutrality. The panel — Al Iverson from ExactTarget, Andrew Kordek of Trendline Interactive and Angela Heckler Knox of Cloudmark unleashed a fury of deliverability knowledge on the group. My notebook was filled with tips and information that will help me down the road.

The session I had the highest hopes for but ended up disappointed was “How Exact Target Uses Exact Target.” It essentially was being a basic overview of the Exact Target tool suite. Perfectly fine, but I’m already a user of most of those tools. I guess I was looking or something really powerful to inspire an “Aha!” moment. It never came. However, I did learn a few things and saw the mobile platform in action for the first time. Nice tool.

Day Three ended on a high note with CNN’s Soledad O’Brien keynoting. The Power of One message really hit me in this session as she discussed her foundation that helps underprivileged women (Read my blog post about O’Brien’s session here).

So don’t let my small disappointments about breakout sessions deter from my overall message: Connections is an amazing event that I would recommend all interactive marketers attend. But really, Connections is what YOU make of it.

In my two trips I have learned so much about marketing concepts, email, social and mobile that I have taken back and applied much of it to my everyday job. I’ve also used it to evangelize best practices throughout my organization to folks who don’t consider themselves marketers. Being inspired at Connections has pushed me to learn more and more because there is so much knowledge to tap from the great people I have met and interacted with there. The networking opportunities alone are enough to fill your LinkedIn profile with dozens of new connections.

So, if you get the opportunity to go to one conference next year, make it Connections. Tweet me when you get there and we can meet up in the Social Media Lounge.

Lessons learned from first subscriber re-engagement effort

What do you call thousands subscribers not opening or clicking links in my company’s emails? Gonzo!

In my last post, I described the email re-enagagement plan I devised to clean up 10 years worth of neglected data. The plan has become reality. It’s always great when things go smoothly, but it’s even better when you learn a lot in the process. 

After the clock ran out on the un-engaged respondants, I provided the list of unsubscribes to our web development team, who then executed the unsubscription to our email lists. There were so many that it took 3 days for the process to run.

The Big Unsub” happened around May 20. I will say that I was very nervous to remove so many people. I doubt there was an easy way to undo it once it happened. So what has happened since then?

It woorrrrrrrrkkked! In the month since the unengaged were removed, we have seen an open rate increase across the board in our email products, anywhere from 3-10%. We have seen % clickthrough increases in most of our products. Oh, and I haven’t received one complaint from a user who stopped receiving emails ::knocks on wood::.

Wait, only “most”? Shouldn’t the math dictate increases across the board as well? Agreed, but it’s hard to pinpoint one reason why clickthrough % didn’t go up for all emails. I have a few ideas:

1. The content of the email isn’t being presented in a way that will spur clicks.  We need to write more enticing descriptions and headlines instead of using the web story headline and first paragraph of the story. It may be easier and faster to do that, but it could be hurting interaction with us.  I was already planning changes for most of our emails, but the ones not seeing immediate improvement from mass unsubscribes might jump to the top of the list.

2. I don’t have enough data points yet. Many of our emails are weekly, so I had just 3-4 sends to go by. Also, we send emails with news about sports teams, some of which are not in season right now. I’m definitely concerned, but maybe the next few sends will show increases in clickthrough rates and open rates.

I also know that re-engagement is just a piece of this puzzle. My email project has multiple elements that are all focused on vastly improving our email products. In future posts, I will be able to actually show you some of the things I’ve accomplished. For now, my advice to you is to stay on top of your email subscriber lists. I’m planning a re-engagement effort every 4-6 months to make sure that the subscribers receiving emails are engaging with us.

The big unsub

Email subscriber management in my org has never been anyone’s job, so it has been “neglected” for ………. 10 years? Not anymore.

One part of my email marketing project is to clean our email subscriber database of unengaged subscribers. I estimate we will save 20% on our cost to send email just by “breaking up” with users who don’t engage with us.

Even though we haven’t “proactively” managed the data, we do follow the rules. The data is secure, we don’t spam, we don’t sell user data, etc… But one problem is the bulk of our user data was collected between 5-7 years ago, when our largest site was registration/subscription-based. Since that business model was changed, we haven’t required registration or login for regular, day-to-day users. Lists have slowly declined throughout the years.

So, where does that leave us? Registrant numbers in the six-figures, but thousands who have long since forgotten about us and our email. Our messages are just ignored or hitting their spam folder, never opened. But, we’re still sending them emails day and night if they’ve never clicked an unsubscribe link.

That’s where I come in. Over the past two months, I’ve worked on a re-engagement plan to reach out to the users ignoring our email with a simple, “Are you in or are you out?”

First, we determined the criteria for what we considered “unengaged.” We chose: Past 90 days, 11 emails sent, 0 opens, 0 clicks. I was able to pull this report in Exact Target, our email sending software provider.

Then, with the help of our marketing department, I crafted an email to send to those users to ask them if they still want to receive email from us. The users had 3 weeks to respond. Less than 2% responded at all, and only half of those said they still wanted our emails.

Today, I used link-click tracking in the email to sort all of the responses into two tables: no action and unsubscribe. Then, I will provide the “unsubscribe” list to our web development team and have them unsubscribe all of those users from all emails we send.

The expected result?

The negatives:
– List decline. Large declines for our biggest lists, I’m sure. But if they aren’t ever reading, who cares?
– Possibility of user complaints for being unsubscribed unknowingly.  

The positives:
-I already mentioned the operational cost savings, up to 20%.
– Higher open and clickthrough rates. Only the engaged will remain, making our emails more sellable to advertisers.
– The start of true subscriber management. I plan to perform this re-engagement effort every three months.

This is the first time my email product management  has had a direct, measurable financial impact this year. There are plenty more things to come with this project, but I’m crossing my fingers that everything first goes well with the unsubscribes.

Great read: Flying Monkeys

This post from Jim Holland at On Product Management has some great advice for product managers to keep a level head in the midst of chaos: “you must focus on what matters most and be grounded in data, artifacts and evidence.”

So true! It’s easy to let the “out-of-nowhere” requests from management and stakeholders and corporate politics get in the way of managing your products effectively, but arming yourself with data, artifacts and evidence is the best way to battle what Holland calls “Flying Monkeys.”

I think I’m blessed to be the analytics manager in my org, as I’m constantly pouring over data and trying to identify trends. I’m often called upon to provide statistics to my management, and those requests give me an opportunity to influence strategy through my analysis.

We’re also putting a lot of effort into goal-setting and benchmarking for each of our Web sites, which will help everyone focus on what they themselves deem as important and eliminate the noise of outside, unimportant data. I’ll write a longer post on what we deem “Success Measurement” soon.

Until then, head on over to OnPM to learn how to combat the Flying Monkeys.

Product Manager for Product Management

I have kind of dubbed myself the “Product Manager for Product Management.”

“Why?” you ask.

Well, a few reasons. Like I’ve said before, I’m building product management from scratch in my org. I’m the “senior” of the product managers and know our org inside and out. I was here when our top website went to a paid-subscription model. I was here when we went back to advertising model. I’ve seen us grow to a network of 40 websites, and I’ve led the content-side launch of a new content management system for those 40 sites. If I’m not the one to implement product management, I don’t know who should (maybe an experienced product manager… but I guess I’ll do.)

So, what have I done so far? Read. read, read, read, read, read. Then read some more. I’ve read lots and lots of blog posts from great product managers. My favorite product management blogs can be found to the right in my blogroll. I need to add more. I’ve slowly eased my way into the #prodmgmt Twitter community. I’ve participated in Product Management Talks (#prodmgmttalk ) on Monday nights through Twitter. Most importantly, I’m understanding it and starting to evangelise it in my org.

I’ve talked previously about the things I’ve been working on, but a good presentation of my Email marketing roadmap a few weeks ago led to an invititation to present my vision the Product Management role in our org.

Last year, we established a true “project” process. Yes, I’ve worked in a department for 10 years that didn’t follow a project process until late last year. Yes, we paid for it dearly over those years. It’s a wonder we got anything done, right? My new product management process (managing the product lifecycle) inserts our project process as the development portion of the lifecycle.

The presentation was a hit. The sales managers asked great questions and I think everyone is on board with the vision. Just like the project process, having the vision is a great first step. The implimentation of product management has gone well so far, and I don’t plan on decommissioning it anytime soon.