The ‘SEND’ button can be a fickle friend during a downturn

Measure email ‘success’ with care or risk squandering your list

When a merchant struggles to meet it’s sales plan, resisting the ‘send’ button on another email broadcast can be devilishly hard to do. Relatively fast and cheap compared to catalogues, where’s the harm with frequent email offers? Let’s explore the answer with one merchant’s true story from December that should give you cause for reflection.  Watching the sales roll in, they overlooked costly consequences.

But before I begin the case study, let’s jump ahead to the key learning points:

1. Be careful what you measure when evaluating your email marketing programme. If you only measure aggregate sales driven by email to determine effectiveness, the ‘send’ button can be a tempting but fickle friend. Pressure to achieve sales targets or maximize a seasonal opportunity drives up the frequency of email offers. But remember: an “It’s working, keep sending!” mentality has consequences that must also be measured.

2. Measure email results with an eye towards maximizing the lifetime (i.e., long-term) value of each address. That presupposes that you’ll have a *relationship* via email, which in turn assumes you’ll even be able to get your emails delivered to that person in the future. If you’re not regularly measuring deliverability and assessing receptivity (both explained below), with a similar kind of attention that you give to measuring sales, then watch out for a painful surprise.

Now back to the case study:

A multi-channel retailer with a bit over £1M in website sales conceived of a ‘deal of the day’ promotion to be delivered via email in the run up to Christmas. Their list consisted of customers or opt-in subscribers who had historically received 1 to 2 email offers per month. As usual, those email addresses received two offers at the start of the month. Then over the next 22 days, an additional 12 email offers arrived.

Each offer featured a product or category prominently, with a special price that expired at midnight or within the week. Each offer’s details appeared both as text (HTML) and again as a tasteful graphic, with the text version present in case the recipients’ email settings blocked images from being displayed. Subject lines conveyed urgency and opportunity. Sometimes two or three additional items received secondary placement. All in all, a professionally planned and executed campaign.

Near the top of every message appeared an invitation to opt-out of email offers with a red “click here” text link. While this always appeared at the bottom of every message, the additional placement at the top demonstrated the retailer’s concern about the potential frequency overload.

Let’s check results:

* Email was the merchant’s only ‘source’ of e-commerce traffic and sales that showed a gain over the prior year (other sources being organic search, catalogue direct referrals, etc. as measured by Google Analytics). On the way to a +30% gain over the prior December’s sales driven by email meant smiles all around the office and a continuing confidence in hitting the “send” button.

* In fact, sales from email helped the company grow year-on-year sales by 10%. Exclude the major contribution from email and sales would have contracted by nearly an equal amount.

By these measures, the email programme showed brilliance in conception and delivery during a very tough retail period. And that’s where the retailer was focussed:  on a measure of aggregate sales driven by email offers.

So let’s look closer and identify other consequences:

* Click through rates (meaning the # of human clicks on links within each email offer) declined from 7% to 5% to 3% to 1.5%.  This demonstrates a decline in interest or unwillingness to really “open” the message.  (By the way, conventional measures of “open rates” are rubbish since the advent of the message preview window. Clicks are what matter.)

* Subscriber list size declined 5% as a result of unsubscribes. People were taking up the offer at the top of the message to ‘click here’ and end the bombardment.

* The list contracted another 5% due to ‘bounces’. This can mean that ‘soft’ bounces due to full mailboxes turned into ‘hard’ or permanent bounces due to sufficient frequency of re-sends during a certain window of time.

* ISP’s noticed an increase in the rate at which recipients (the ISP’s customers) flagged the messages as spam. Combined with the higher than normal bounce rates, some ISP’s took action. One, the 4th most important ISP on the retailer’s list, temporarily blacklisted the merchant. This meant all messages were sent to junk mail and not delivered to recipients. Another ISP, Gmail, flagged the retailer with a “watch” status, meaning it was on the verge of being blacklisted, a fate from which it’s hard to recover with that particular ISP. And those are just the ISP’s the merchant knows about.

With hindsight, a seemingly successful programme may have weakened the brand’s ability to market effectively by email over the long-term. In my opinion, I would never wish to see a decline in the size of a list.  I would prioritise a focus on maintaining click-through rates that are above average.  As a result, I believe that sales growth will naturally follow.

Put yourself in your customer’s shoes.  She’s probably signed up to 10 or 20 email lists, depending on how prodigious a shopper she is.  We’re all vying for her attention, and a share in her discretional spending.  Unless she’s an ultimate fan of your brand (yes – they do exist, and you should make the most of them), surely it’s common sense that her attention span will diminish in inverse proportion to the number of messages you send her – until eventually she doesn’t even notice what they’re emailing about?  The tragedy of this scattergun strategy is that when you finally email your customer with something she genuinely is interested in, she may well have hit the delete button before she even noticed.

The risk of too frequent *broadcasts* is email list fatigue and contraction. And you won’t know what too frequent is for your customers unless you can really measure it. Moreover, you risk the reputation of your server and business, should ISP’s decide not to deliver your messages. So you’d do well to measure ‘deliverability’, too.

Since 2006, my companies have relied on an email marketing platform from E-business Coach that measures the health of my programme in these ways. See the list nearby for important questions to ask your email service provider, or to ask when evaluating a new one.

Here’s one way in which I would have handled the campaign detailed above differently.  After the 2nd in the flurry of email offers, I would have filtered the list according to the actions of the recipients themselves. Did they click on a link, giving a true measure of whether they opened the message and paid attention? If yes, send them a subsequent offer. Measure. Repeat.

This kind of email campaign becomes an exercise in “narrowcasting” as compared with universal “broadcasting” to the whole list. An email campaign sustains and strengthens a retailer when it encourages customer relationships by narrowcasting relevant offers at an “appropriate” frequency. The current retail climate is the perfect time to use such a strategy, given that there are many retailers who will find it difficult to resist the send button over the coming year, and tax our patience in the process.

What do you think? Please write in to share your assessment or suggestion for handling this retailer’s email campaign. What would you do differently given the chance?

*********************************************************************************
Can you measure these essential aspects of your email programme?

Ask your email service provider, or the next one you evaluate:

1. Can I measure the rate at which my messages are delivered to the top (50, 100, or more) ISP’s?

2. Can I test that deliverability before I send the final version of my message.  In other words, can I test deliverability without sending to my own list?

3. Can I measure click-through rates, or the number of times a link in the message was clicked?

4. Can I identify specifically which link each recipient of the message clicked?

5. Can I easily develop a follow-up recipient list tailored to those email addresses that show having clicked on a certain link?

If you’re capable of running an email campaign with those capabilities above, you’re well on your way to making email a sustainable driver of web sales, come good times or bad.

*********************************************************************************

© Murray Kenneth 2009
First published in Catalogue & e-Business
Picture by S Migol

Tags: Email marketing