Click through rates hide the true impact of marketing

Click through rates are a one sided measure that can hide the negative impact of marketing messages

Mobile marketing notification campaigns are often measured using open rates. This is not only a hugely simplistic measure, but also a one sided one. Someone opened the message - but how did their behaviour change? That’s not measured. More importantly what about all those people who saw but didn’t open the message? If their behaviour changed, we don’t know about it.

Beyond open rates, you can measure click throughs. People opened the message, but did they then click the link to the landing page, or the shopping cart? Exactly the same problem. We don’t have any visibility of those who didn’t click.

Even if you get as far as a conversion rate, tracking purchase transactions, the one sided data problem persists.

This type of measurement leaves you blissfully unaware of possibly negative outcomes. All those people who didn’t click the link. Did their behaviour change? Are they more, or maybe less attracted to your brand?

And negative reactions to marketing messages certainly do happen.

Since June 2017 we have been providing our customers with ‘R4 Scores’ - a comprehensive measurement of real-world impact. We monitor the behaviour of groups that did and didn’t receive a marketing message. And use statistical analysis to measure behaviour change.

The most important finding has been the number of negative reaction messages. These are push notifications that make people LESS likely to visit the place or participate in the event that has been suggested. The effect can sometimes be clearly measured into the following day.

Let’s go back to traditional marketing messages, tracked using open rates or click throughs. A few percent of recipients might click a link. Those that don’t click are invisible.

If we send 1,000 messages, we might get 20 clicks. If we send 1 million messages we might get 20,000 clicks. The more the better.

But what about the 98% of people who didn’t click? Without knowing how the message has affected them, we’re missing data on 980,000 people exposed to our message. We have no idea what the long term net effect of the campaign is.

It would well be that offsetting the extra 20,000 visits to a website attributed to the campaign, there’s an overall decline in traffic over the next month of 100,000 visitors.

Crowd Connected’s R4 score is different. We can measure the behavioural impact of the campaign as a whole, positive or negative.

The results we’re seeing reinforce a strategy that concentrates on service delivery and customer experience, and not marketing.

Messages that just tell the customer to do something - and they were already aware of the option - perform worst. This is pure advertising. There’s really no value to the customer. There’s no new information provided. There’s no enhancement to the customer experience or the service. Regularly we see these types of messages actually REDUCE footfall to the advertised location.

Irena Cronin from Transformation Group calls this ‘customer dumbification’, in which every marketer assumes every person within range of any given location should get suggestions and instructions.

Messages that make customers aware of something new, provide information on options, or in some way add something to the service or customer experience, work well.

This has some interesting implications for location-based targeting.

A common tactic is to target messages to people based on where they are right now. Beacon based proximity messages do this. Target messages to passers-by - on the basis that nearby is relevant, and far away is not. This model works well for search - nearby results do tend to be the ones we’re after. But it may not work so well for push notifications.

A passer-by is more likely to already be aware. They can see the shop, and if you message them suggesting they pop in, that’s just the kind of advertisement that we’re seeing can negatively affect customers.

Location targeting needs to be become much more sophisticated than that, and measurement needs to be based on real behaviour, not on misleading opens or click throughs. Otherwise a lot of mobile marketing is unwittingly damaging the very market it’s trying to grow.