Hello email marketing fans! Quick recap… back in October, I subscribed to the email blasts of 50 of the Top 100 Internet Retailers, and monitored the email sends over the next seven weeks. The purpose of this was twofold:
1. To find good ideas that can be translated across any email marketing campaign.
2. As email typically shows the strongest return on investment of any online marketing channel, to see if it was really true that even the biggest companies still made basic mistakes.
With that said, welcome to Phase 3, the last blog post in the Email Experiment trilogy. This post can be helpful on its own, but if you are interested, feel free to check out the previous two phases, to see how we set up the experiment and what we’ve learned so far:
- Phase 1 – Sign Up, List Growth, & Lifecycle Emails
- Phase 2 – Social Integration, Mix Of Messaging, Remailing, & Unsubscribing
This final post lays out the big picture stats for the whole experiment, reviewing overall email distribution, summarizing my five golden rules for email marketing, and divulging my blooper reel from the experiment!
Big Picture Stats
I received 1,614 emails in total during the seven weeks of the experiment: 85 of which were triggered emails and 1,529 were regular email blasts.
That represented an average 30.6 emails per brand (or 31.9 if you ignore the two brands that failed to send any emails – more on that later) and an average of around 4.5 emails per week.
I also kept track of how many emails I received per day and what time of the day they were sent. For the charts below, I excluded lifecycle or triggered emails, as these should be responsive to an action and therefore not set for a specific day or time. I also excluded Week 1, as that week was low in email sends, most likely due to some brands experiencing a delay between opt in and ESP (Email Service Provider) system updates.
The peaks were commonly seen between 4am and 6am, which is pretty standard. The strategy behind this timing is for the email to appear in a prominent position in the subscribers email box when they first check their emails in the morning. However, depending on your brand, this information may also be helpful to identify the quieter times, when there is less competition.
Evenings have traditionally been slow for email sends, as we saw in our experiment. However, there are current trends in the online marketing industry, which might cause this to change, particularly for retail brands. Mobile search activity is on the verge of overtaking desktop search numbers. When you consider that alongside the fact that people are also much more likely to use their mobile phones in the evenings, as well as search across multiple devices, it might not be such a bad idea to test a few evening email sends. Furthermore, one ESP states the most active email open hours are between 2pm and 5pm, so why not strategically segment your email blasts to test the best time of the day for your audience?
In terms of volume by week, unsurprisingly, email sends really ramped up towards Black Friday and Cyber Monday (which fell in Week 7 of the experiment), with that week seeing a 60% increase in email volume compared to the next closest week (Week 6).
Quality Control a.k.a. The Blooper Reel
There are glitches in every system, but I spotted many during the course of the experiment that are pretty basic, particularly for such an impressive set of brands. No naming and shaming here (Wpromote is way too nice for that!) but perhaps this can give you a few ideas of things that you might like to avoid in your own email marketing program…
Sign-Up Nightmares: For one company, the email sign-up box on the homepage was broken, with the “Submit” button being nothing more than an image. When I did manage to sign-up via another route, the first email asked me to confirm my subscription by clicking on a link. When I did so, it told me my email address wasn’t in their system, so I had to go through the whole process again. It is important to test the system regularly, to avoid missing out on would-be subscribers.
Spelling Errors: Make sure that there aren’t any typos – I saw “RECIEVE” twice during the sign-up process alone, from two different brands.
Broken Links: Whilst I was happy to see social media profile buttons being used on so many emails, not all of the links worked. Another element that is good to test regularly.
Win-Back Emails: A win-back email is a lifecycle email that you send to try to re-engage a subscriber who hasn’t interacted with you in a while. One company insisted on sending me a win-back email only 5 days after my welcome email. It was along the lines of “So this is awkward, but are we breaking up?”, which would be cute except that I received it once a week for the first 3 weeks of my subscription. So yes we are definitely breaking up if this is the only thing you’re going to send to me – you are so needy!
SPAM Protection: One particularly prominent retailer used its regular domain for the email blast sender email address. For domain-protection issues related to SPAM, we always advise sending emails from a sub-domain, as most of the other brands did. That way if the domain gets flagged for SPAM for any reason, it will not affect the email addresses of corporate employees. A sub-domain is also preferable to a new domain (i.e. SampleBrand.com should use something like “e.samplebrand.com” as opposed to “samplebrandemails.com”) so that your subscribers can be confident that you are the official email program.
No Emails: You would think that sending emails to your subscriber base was a fairly fundamental part of an email campaign, but even the big boys seem to find this tricky! Two brands, after the initial welcome email, had a delay of nearly 7 weeks before sending any further emails. Two other brands failed to send me any emails at all, aside from the welcome email that proved they had the correct address. Testing is very important to make sure that interested subscribers are added quickly and receive their emails.
Have you seen any other howlers? We’d love to hear from you in the comments section below!
My Five Golden Rules: The Big Takeaways From The Experiment
So, what have we learned? Here are my top recommendations, which I have rather modestly called my five golden rules of email marketing:
RULE #1: Let people know that you have an email program and make it easy to find and sign-up! (Plus, test the sign-up process to make sure that they are added to the database quickly and start receiving your emails.)
RULE #2: Send a prompt welcome email to everyone who signs up. Use it to set expectations, gather more demographic details for segmentation and strike whilst the iron is hot for a sale!
RULE #3: Use the data at your disposal. This means integrating email with your social media and other marketing channels. Try remailing tactics and create segments and funnels to send targeted messages.
RULE #4: Don’t be boring! Entertain your subscribers with fun content and helpful information, as well as transactional messaging to secure a sale.
RULE #5: Give someone options when they want to unsubscribe. Don’t cut them off completely if you don’t have to!
Are there any other elements or results you’d be interested in learning from this experiment? There are obviously many more elements that we could have explored. Please let us know what you’d like to see. We’d love to hear from you!
Also, if you love these ideas but don’t have the resources to implement them – please get in touch! We’d be happy to chat about your email needs and see how we could be of help.