Jenny Chan 陳詠欣
Feb 4, 2015

Email marketing tips: Personalise without frightening, avoid the spam box

BEIJING / HONG KONG - Campaign sat down with Emarsys executives Stuart Barker (Hong Kong country manager) and Ofri Cohen (China general manager) and asked them to explain the insights and analytics that go on behind promotional emails—in plain English.

Emarsys's predictive recommendation engine
Emarsys's predictive recommendation engine

Emarsys clients in Greater China include PEdaily.cn, SCMP, Time Out, Flight Centre, Galeries Lafayette, Chow Sang Sang, Jiuxian, Lacoste, I.T., Sasa, Dymocks, Li Ning and Watsons. How are clients in different industries using email marketing?

Barker: I think the best place to start is actually by taking time off the clock and going back in time a little bit. About a decade ago when we were established in Europe, the email marketing industry was emerging. It didn’t change a great deal for the first five or six years. We were dealing with customers from all sorts of verticals: retailers, e-commerce, travel and tourism, hospitality, finance, publishing, printing.

As time has progressed, what’s happened is the requirements of our clients, in whatever sector they may be, have increased, and they’ve looked to us to provide them with more and more information and capability to allow them to speak to their customers. The underlying principle of what we’re here to do is to help brands have great relationships with their customers. And that hasn’t changed. 

Now, what’s interesting is that in the last couple of years, we’ve started to see a few differences in the ways some verticals are now demanding. We certainly meet the needs of probably 60 or 70 per cent of the market for technology solutions. But I can see that our company is heading to meet the needs of a couple of key verticals: retail and e-commerce, travel and tourism, media and publishing.

In e-commerce and retail, there is a lot of backlash to retargeting. Where do you cross the line between personalisation and invasion of privacy when you’re shopping online?

Barker: Actually, consumers globally have always been proven to respond well to content and communication that recognises who they are and treat them as VIPs. If you’re receiving endless emails about men’s shoes, which I’m guessing are no good to you, you’re going to be affronted by the brand, you’re not going to engage, you’re going to be disappointed, you’re going to turn away. Eventually you’ll unsubscribe. 

If that brand is consistently delivering content-like offers and product news which is not only relevant for you but feels personalised and tailored for you, then your sentiment towards that brand is going to be very much more positive, and your willingness to engage and become a customer, of course, will remain. So that’s the key challenge, which is using the best of technology to make customers think, "How marvellous. They thought of me. This brand thought of me before they sent this email."

That’s Email Marketing 101 really. But we’re facing a lot of issues in that…some people are saying loyalty programmes are dead. There are too many emails that we’re getting, needing a lot of electronic housecleaning. Some people don’t even look at their emails anymore. So there is maybe 80 per cent of wastage, or even more, for a single email-marketing campaign.

Barker: You raise a really good point, and it’s one that we totally embrace. In fact, what we’re trying to do is getting our clients to send less emails, but to make sure the emails that they send are really relevant. You’ve identified a real problem for the industry. For so long, a lot of brands have just been sending large-volume bulk emails with no personalisation, with no effort to honour the individual who’s receiving it. And because of this, consumers are becoming weary. They’re thinking 'this is of no value to me'. However, time and time again—and we know this from our work in Hong Kong and China—if we go back to that 101, it's an opportunity to improve. 

Can you give examples of how you’re trying to solve this problem? What exactly is relevance to a customer? Everybody has different preferences. So how do you gauge? 

Cohen: It is a very interesting question. In China, we found out that there are some other aspects that come into play in the wastage of email marketing.

Three or four years ago, we could send whatever content we wanted to send to whomever we wanted. Of course, the 'email epidemic' got worse and worse in China. And the major mailbox providers, to protect user privacy and user interest, initiated strict regulations or policies. Placing a message inside the inbox of a recipient should not be taken for granted now. Chinese email providers are setting daily volume limitations according to email domains, and IP reputation shortcuts that used to work before will not work today.

The most important thing that marketers need to acknowledge in China is email marketing can not be used as an acquisition tool. We try to shift the focus from acquisition to retention. But we are not saying acquisition is not important. Once we have acquired a customer, we should be able to treat him or her differently and make sure he or she is buying more, by integrating email marketing with predictive analytics and customer intelligence.

The key battleground is getting someone who’s made one purchase to make a second.  If they make a second, the likelihood of making a third, fourth, and fifth goes up exponentially. So treating first-time buyers differently is important.

There’s a statistic: 80 per cent of your revenue is coming from your retained customers. The money that you spend on acquisition get some of the poorest returns. Clearly, brands have got to keep topping up their customer base, but if they can work with their existing customers, it's much easier to make money.

Barker: In Hong Kong, the quality of the email campaign makes the difference between being first or second in the consumer’s mind. In China, the quality of the campaign actually enables you to do further business. This notion of quality is being measured by the consumer. 

Cohen: So hit the highly-engaged consumer that’s going to value your message. Hit them right on the nose with something targeted and powerful. Don’t try and just throw emails at your database left, right, and centre without any precision. Once you comply with ISP regulations, you can hit the inboxes. If you look for shortcuts, like switching domains and IPs frequently, you will be dropped.

Specifically, how are you enhancing your IP reputation? Let’s get to the nitty-gritty.

Cohen: In China in particular, each of the different mailbox providers has their own set of guidelines. So, it’s not about just one national layer of legislation about email marketing. It is about being in good compliance. And it is not only about having lunch or dinner with them.

But lunch and dinner always helps, right?

Cohen: No, not in this case, I have to say. Because I can tell you that I came across a few industry experts who are saying that using relationships with mailbox providers does not open more channels. 

Okay. Now that you’ve got the regulations out of the way, there’re still the customers to win over.  Can you tell me what goes behind an email? What are the predictive technology and analytics?

Barker: It’s important that you ask this. The four questions that we help our clients to answer are: Who’s the customer? What are they interested in? When do they want to receive news about it? How do we get to them? And that 'who, what, when, and how' is actually four separate technology challenges. 

The 'who' challenge is the challenge of identity. It’s about understanding who the consumer is across multiple touch points. Is the phone, computer, browser the same person? 

The 'what' challenge is looking at the content that they need to receive. And that’s where we use our predictive recommendation algorithms. We have made significant investments in creating the mathematics and the framework that allow us to understand behaviour across touch points and channels, and make recommendations around products or content that they’ll be interested in. 

The 'when' challenge is around automation. The challenge for marketers has always been to find enough hours in the day to execute the high-quality campaigns they want to execute, using business rules rather than just building campaigns. 

The 'how' challenge used to be, for us, about email. In the last couple of years, we’ve diversified from email to on-site experiences, landing webpages, social messaging, mobile. Here’s a question for you. You have received six emails from a particular brand, and you haven’t opened any of them. How likely is it that another email is going to be the thing that finally makes you decide that you’re interested again? Pretty low, right? So perhaps email is no longer a channel you’re engaged with. Perhaps the brand can switch to a new channel. Why not try a SMS? A social channel anchored around the email address?

What if a customer has two email addresses? One is for ‘proper’ communications and one is for marketing spam? The second is what I would submit whenever I sign up for a loyalty programme or a newsletter, or when I only want to have a short-term commercial relationship with that brand. Ultimately, there are some customers who have that attitude. How do you get around this? A unified profile doesn’t work in this case.

Barker: You have this sacrificial email address because you’re tired of receiving a gazillion email messages. What we’re about, fundamentally, is changing the marketer mindsets and saying to them, people have invented email addresses that they throw away just so you don’t bombard them with rubbish. 

If that brand changes their game and says, we’re going to be relevant and timely and precise, we’re not going to bombard you with a million meaningless messages, we’re actually going to respect you as an individual and try to delight you with things that we use analytical science to prove, then over time, we think there will be less sacrificial email addresses.

How do you define a high-quality email campaign?

Barker: The first thing we do when we start working with a new client is we ask them for about 18 months’ worth of their historical transaction data. And what this allows us to do is understand precisely - not guess, but understand precisely - the things that really count in their business. 

Often, when we speak to marketers, they will say to you that they have a campaign targeted at lapsed customers. And we ask, well, how does that work? And they say, 90 days after the customers haven’t shopped, we send them an email with a coupon. And the question we ask next is, why 90 days? And everyone looks at each other, and they go, “That’s just a standard…”

And the point is, if you’re selling socks, 90 days might be about right. If you’re selling Rolex watches, I’d say probably 90 months is more appropriate. And there needs to be some science behind this.

What is the toughest question that a client has ever asked you?

Cohen: I find that in China, a lot of clients want to dig into the way we perform predictive analytic calculations and mathematical correlations. Those, as you can imagine, are tough questions. 

Barker: And it becomes incredibly boring. In a previous role, I was involved with something called choice modelling. And that was an algorithm that was looking at conjoined analysis in customer purchase decisions. So you can disappear right into science. Look, I think clients have the right to ask questions around how their data is being used. With regards to predictive analytics, it’s just mathematics that creates accurate algorithms. You know, there’s nothing sinister about it, per se. The algorithms use some kind of fractional factorials to look at the behaviour of other shoppers who are similar to one another. It aggregates that with previous shopper behaviour, basket composition and the product catalogue but with different weightings for different variables.

Cohen: We are happy to show about 90 per cent of our technology to clients, and keep a little. The secret sauce. You know how they make the batter at KFC?  You never know the last 10 per cent of ingredients.

 

Source:
Campaign Asia

Related Articles

Just Published

7 hours ago

OpenAI marks arrival in Asia Pacific with opening ...

To mark the opening of its Japan office, OpenAI has released a GPT-4 custom model optimised for the Japanese language.

8 hours ago

40 Under 40 2023: Shu Wu, McCann Worldgroup

Shu Wu's transformative leadership led McCann Worldgroup China through unprecedented success, securing major clients and empowering diverse talents for the company

9 hours ago

Canva makes an unbelievably good presentation...and ...

The work from UltraSuperNew taps iconic Japanese actors Takeshi Kitano and Gekidan Hitori in an suspenseful, funny and memorable hit that reinforces how Canva allows one to take care of business on their own.

13 hours ago

Mars Wrigley India CMO on evolving snack tastes and ...

Nikhil Rao explains how Mars Wrigley brands are adapting to health and social trends, how premiumsation factors-in for older demographics and how personalised content is key for Gen Z.