What email marketers can learn from newspaper headlines
16 Oct 2013
Many years ago I was chatting with a newspaper editor about how he picked the ‘splash’ (lead story) in each edition. His response offered as good a definition of relevance as any I’ve heard, and one that works equally well for email marketing.
I’d been surprised to see the headline “So Close to Tragedy” in January 1994, the day after an earthquake had hit Los Angeles, killing 57 people. The headline, however, referred to Queen Elizabeth falling off a horse and breaking her wrist. The earthquake made just a few column inches, tucked away in the corner. I wanted to know ‘why’.
He used the analogy of ripples from a pebble in a pond to explain it. The further away you are from where the pebble hits, the weaker the ripple when it reaches you. Events in LA, being over 5,400 miles away, are simply less relevant to people in London than events in Sandringham.
In other words, location matters. The earthquake, devastating as it was in California, just wasn’t relevant enough to dominate the day’s news here.
Location is becoming the new touchstone for marketers trying to define the relevance of their customer conversations. As an industry, we’ve invested a lot of time and energy in modelling declared data (such as preferences and demographics), observed data (web and email-related behaviour) and transactional data (RFM), but we’ve been slower to capitalise on context.
Context can be harder to get right – but it’s potentially much more rewarding. And much more engaging.
Local content=relevant content
I was at an Adobe conference last week where Andy Gray, email marketing manager for the Post Office, threw away a remark about getting local content into his digital conversations as a way of trying to recreate community involvement. As sponsors of Channel 4 weather, there’s an obvious link to exploit – and of course, the Post Office has historically been a focal point of local community, and the Postmaster a key figure in community life.
So incorporating your customer’s location in your targeting strategy just makes sense. We’ve been putting addresses of local branches based on the recipient's home address in the footer of retailers’ emails for years – just ask the likes of Jigsaw.
But that kind of fixed, offline location is only the tip of the iceberg. As email has now become a primarily mobile-consumed channel, we also need to consider a couple of different contexts for location:
1. “I’m here and I want (to do) this, now”
2. “I will be there and I will want (to do) that, later.”
We can see – and act on – the first scenario very directly. I’ve now got quite specific geographic data (admittedly with a margin of error) about every opened email my agency has sent over the past year or so. I can now dynamically or programmatically target specific content based on that information, either as a single targeting point, or as one factor in a more complex decision.
Dynamically, by varying an image in the email based on location; programmatically by mapping previous opens back to an email address and making assumptions.
For the likes of Screwfix, this means we can target based on tightly defined weather forecasts (such as selling ice chains and de-icer products when frosts and snow are forecast). For Lastminute.com it might mean focusing hotel offers in York to customers in the broader Yorkshire area.
Combined with the investment we’ve already made into modelled evidence of intent (such as declared data, or observed search or browse data), location matters.
Mike Weston, Managing Director, Profusion
I’d been surprised to see the headline “So Close to Tragedy” in January 1994, the day after an earthquake had hit Los Angeles, killing 57 people. The headline, however, referred to Queen Elizabeth falling off a horse and breaking her wrist. The earthquake made just a few column inches, tucked away in the corner. I wanted to know ‘why’.
He used the analogy of ripples from a pebble in a pond to explain it. The further away you are from where the pebble hits, the weaker the ripple when it reaches you. Events in LA, being over 5,400 miles away, are simply less relevant to people in London than events in Sandringham.
In other words, location matters. The earthquake, devastating as it was in California, just wasn’t relevant enough to dominate the day’s news here.
Location is becoming the new touchstone for marketers trying to define the relevance of their customer conversations. As an industry, we’ve invested a lot of time and energy in modelling declared data (such as preferences and demographics), observed data (web and email-related behaviour) and transactional data (RFM), but we’ve been slower to capitalise on context.
Context can be harder to get right – but it’s potentially much more rewarding. And much more engaging.
Local content=relevant content
I was at an Adobe conference last week where Andy Gray, email marketing manager for the Post Office, threw away a remark about getting local content into his digital conversations as a way of trying to recreate community involvement. As sponsors of Channel 4 weather, there’s an obvious link to exploit – and of course, the Post Office has historically been a focal point of local community, and the Postmaster a key figure in community life.
So incorporating your customer’s location in your targeting strategy just makes sense. We’ve been putting addresses of local branches based on the recipient's home address in the footer of retailers’ emails for years – just ask the likes of Jigsaw.
But that kind of fixed, offline location is only the tip of the iceberg. As email has now become a primarily mobile-consumed channel, we also need to consider a couple of different contexts for location:
1. “I’m here and I want (to do) this, now”
2. “I will be there and I will want (to do) that, later.”
We can see – and act on – the first scenario very directly. I’ve now got quite specific geographic data (admittedly with a margin of error) about every opened email my agency has sent over the past year or so. I can now dynamically or programmatically target specific content based on that information, either as a single targeting point, or as one factor in a more complex decision.
Dynamically, by varying an image in the email based on location; programmatically by mapping previous opens back to an email address and making assumptions.
For the likes of Screwfix, this means we can target based on tightly defined weather forecasts (such as selling ice chains and de-icer products when frosts and snow are forecast). For Lastminute.com it might mean focusing hotel offers in York to customers in the broader Yorkshire area.
Combined with the investment we’ve already made into modelled evidence of intent (such as declared data, or observed search or browse data), location matters.
Mike Weston, Managing Director, Profusion
Please login to comment.
Comments