How New Zealand’s classifieds giant is tapping automation for member marketing
- 01 May, 2017 06:00
Finding a balance between dynamic, personalised communication that is subtle and effective and content that crosses the creepy line is the ongoing ambition for Trade Me head of marketing automation, Anna McLauchlan.
The Wellington-based online classified site has spent the past two years developing its marketing automation strategy to improve interactions with members. McLauchlan was brought on-board two years ago to set up the new platform, a task she admits was challenging and daunting initially because of the company’s culture of building digital systems internally.
“It took a while to navigate through that, and the fact that a big change was coming, but the reality was the internally built email platform wasn’t going to scale, couldn’t do data-driven marketing, could barely send responsive emails, and was incredibly inefficient,” she told CMO.
Key to selling Marketo’s marketing automation platform internally was focusing on efficiency improvements. At the same time, McLauchlan needed to educate the business that the technology wouldn’t be successful without the right staff and processes supporting it.
A first step was deciding what roles were required. “We have put in an analyst, plus our own designer, and we have campaign managers that run the platform and all the best practices that go along with that,” McLauchlanexplained. “Quite often, 90 per cent of thinking and work is done outside the platform.”
The other key is the data driving the platform. “If there’s no data, there’s no automation,” McLauchlan said. “And not just any data, it’s got to be the right data to push the right message at the right time. This is the role the analyst plays – they do a lot of pre-campaign analytics, combined with post-campaign analytics to ensure we’re targeting members at the right time.”
Trade Me has huge volumes of consumer traffic and had been capturing all information in the back end. However, it historically hadn’t tapped these data sets for insights because it didn’t need to, McLauchlan said.
“Everyone naturally came to Trade Me. But it’s pretty easy to get stuff into NZ now and when other international players hit the market like eBay and Amazon, Trade Me has to step up,” she said. “Its competitive advantage was the fact that it was sitting on this awesome set of the most amazing digital data in a beautiful data warehouse. But it’s not exactly in the right format for the analyst to dive in and mine that data.”
In addition, because Trade Me hadn’t really undertaken any marketing before, there was a need to test and learn to see what did and didn’t resonate with members, McLauchlan continued.
“We have very loyal, engaged members, so we had to be careful about how we were going to do email marketing. We didn’t want them all to unsubscribe or hit them with creepy marketing messages,” she said. “We try to be subtle but effective at the same time.”
Getting quick wins
One lesson McLauchlan’s learnt with marketing automation is that you don’t try to tackle all marketing campaigns and emails straight off the bat. With an ambition to get its first campaigns up and running in eight weeks to secure some quick wins, her team instead started small.
The team came up with 10 ‘use cases’, targeting different campaigns, and aimed at reduce the time it took to create an email campaign from up to 40 hours, to less than a day. That meant building up email deliverability processes, standardised templates, and a unified brand style.
“Our email templates hadn’t been touched in 15 years pretty much, and a whole new design was needed, making them responsive, testing them, and so on,” McLauchlan said.
The first campaign was delivered in six weeks, and promoted sale offers from one store on the Trade Me site. “It was reasonably generic but a good first one to trial,” McLauchlan said, adding challenges with ISP blocking also made it an obvious choice.
The first year was then about testing the platform and seeing what would work with customers.
“In the last 18 months in our testing, we’ve seen that generic messaging of ‘hey you need to buy this’ doesn’t work nearly as well as some of the more refined, data-driven marketing,” McLauchlan said. “We can see open rates and clickthrough rates double when we have a specific, tailored marketing message for you. We can do that because we have our analytics team diving into the data. It may look like one version of an email, but behind the scenes, all the imagery changes.”
For example, over Christmas, if Trade Me knew a consumer had looked for Lego the previous year, it’d potentially move Lego up higher in the email sent to that member.
“It’s customised using data to be dynamic, and we can have thousands of versions of that one email,” McLauchlan added. “It’s about mining data and serving you up relevant content that you’re going to become engaged in.”
Knowing how personalised to get
One of the big questions is just how personalised campaigns should be. McLauchlan suggested personalisation is a trade-off with time.
“If that relevancy generated 100 different emails and it took six weeks to get a campaign out the door, then never,” she claimed. “We should be able to generate that file of dynamic content pretty easily and for me, you need to be able to read an email, and know the company got that right, but not be creepy. It’s got to be subtle – that’s what we’re looking to achieve.”
An example of what has worked well to date is Trade Me’s dynamic property price emails, triggered to members when the price of a particular property they’ve viewed online drops.
“We send you an email about the actual house, with a dynamic link to it online, that is has dropped in price, and encourage you to check it out,” she said. The emails have open rates of more than 60 per cent and clickthrough rates double the average.
“It’s simple, relevant, timely and useful,” McLauchlan said. “We’re still trialling a bunch of those. I don’t think we would have known a year ago that would have worked.”
Mining the data
What’s helping the marketing teams work out which members to target with what offer or email is a comprehensive data analytics approach based around a business objective. Trade Me senior campaign analyst, Rachel Wilson, said that in most cases, things start with a kick-off meeting, where the analyst, designer, copyrighter and marketing managers get together and brainstorm an idea.
“From there, we get ideas on who we might want to target - often marketers do want to hit a certain amount of people,” she said. “We’ll do some pre-analysis, look at numbers and see where cut-off points may be.”
McLauchlan said it’s about having a business problem, rather than a request for people who have blue eyes or walk to work. “That’s where the analyst can add such value, and that’s taken a while to get to that point,” she said.
It’s also important to be savvy with data and think beyond one campaign. “How that data could potentially go elsewhere, how you can scale and what other campaigns could use that data are all key here,” McLachlan said.
Putting the emphasis on process
Through all of this, creating best practice around marketing automation and campaigns has been vital. Trade Me not only has a 55-point checklist campaign managers must check campaigns against, it also has created its own automation playbook.
“We try and put design at the front of what we do, plus job numbers,” McLauchlan said. She attributed these processes to why the organisation won ‘The Performer’ award at this year’s Marketo Revvie awards.
While there wasn’t a lot of cultural resistance to adopting marketing automation, McLauchlan said some pushback came in the early stages around trust.
“Teams needed to know we weren’t going to be sending out 10 million emails when we shouldn’t be,” she said. “That’s why the checks that were going to be put in place around campaigns was key.”
McLauchlan also undertook international presentations, and worked closely with the customer experience team managing the front line.
“As we bring more people into the team, we continue do a lot of on-the-ground training around things we need people to do – such as the types of text and style of our emails,” she said. “We need to have people that can deal with that level of detail. When we hire, it’s also people who can write, because that’s almost more important than people with Marketo experience.”
Next priority: The content challenge
McLauchlan’s next priority is content, and she noted Trade Me lacks a centralised content hub. There are also a number of businesses running under the Trade Me umbrella brand, including jobs, property and automotive, often operating as siloed teams. This presents opportunities for more integrated member communication.
“We’re working through customer lifecycle journeys for each business unit, so that we all work in unison,” McLauchlan said. “You could be looking for a job, buying a house, buying or selling a car. Understanding what those customer lifecycles are so we get to you right before you buy that house, and know to send you relevant content at that point, is key. We’re not there yet but we’re getting closer.”
Read more of CMO's in-depth case studies on marketing automation success:
- How HBF rolled out marketing automation in eight months
- Cronulla Sharks to expand marketing automation efforts
- 3 brands using marketing automation for more than just email
- 5 lessons from AMP on deploying marketing automation
- How Flight Centre’s B2B teams harnessed marketing automation
- Why Melbourne Victory Football Club is rolling out marketing automation
- How marketing automation is helping Xero achieve real-time and personalised communication
- Nadia Cameron travelled to Marketo Marketing Nation summit as a guest of Marketo.