As someone who has worked in email marketing since 2002 I remember the days when no-one did triggered based messaging. Typically, once a week you would load in a CSV of your database into the email marketing software (they weren’t called ESP’s back then), add in your hand coded email template (Drag and drop wasn’t a thing kids) and schedule a send before pulling the bounces and unsubscribes to pass to IT to update the AS400 mainframe database!
We’ve thankfully evolved now and many of the campaigns we do are automated, triggered at a specific time in the customer lifecycle. Having automated emails such as cart abandonment, welcome journeys and winbacks are the norm – but are we putting too much emphasis on them nowadays?
While its true these emails typically generate more revenue per email than your batch sends, simply adding more of them isn’t the right way forward for most marketers.
Here is why:
Triggered emails such as cart abandonment and welcome emails can look better than they really perform due to how you measure, and who you are sending to. These audiences are the most likely to be making a purchase within a short period of time so if you are measuring with a last click attribution, or simply counting those that got or opened the email within a few days you are likely to be overstating their effectiveness.
A lot of these customers would have bought anyway, and as they are engaged opened and interacted with the email. The essential test is always to measure with a hold out group and compare those not getting the email and how many sales they make in the following week.
If you find that actually there isn’t much incremental uplift by sending the emails then you are not going to generate much more with more triggers in this area,
By their nature triggered emails go to very small audiences. If you take a winback campaign that say is sent 180 days since last purchase that will go to a tiny audience every day, and over the course of the year add up to not a lot.
Instead, one email to the whole lapsed database will dwarf that triggers annual send volume. To match that with triggers you need a lot – 180 days, 210 days, 240 and so on. You’ll quickly get fed up trying to build all of those creatives.
In this instance it would be far more effective and easier to manage to simply plot on a calendar this reactivation activity every 4-6 weeks at a time, working around clashes with other promotional activity.
Each triggered message you create requires maintenance. As your business changes – whether that be products you offer or brand changes – you will need to update all of these.
Having too many triggers actually stops you from getting on with more strategic work as you end up spending hours each week fixing old templates, updating copy and debugging workflows.
The way most marketers setup triggered messaging is the selection criteria doesn’t take into account all the other possible messages that could be sent. For example, a recent customer could qualify for the welcome email, cart abandonment and your weekly newsletter all on the same day.
Adding more triggered messages simply exacerbates this problem and leads to customers feeling swamped with the volume of email and conflicting messaging.
If you were planning more automated messages but now you’re not sure where to spend your time what should you do instead?
This is where personalisation comes in. Personalisation in your main campaign sends makes them work harder, at scale. Personalisation still shares the same attribute of triggers that you setup today, and you reap the benefits on an ongoing basis.
But the difference is even a pessimistic 10% increase in revenue personalisation can add from your regular campaigns dwarfs that any new triggers could achieve. After all most brands (when using realistic revenue attribution) are only generating 10-20% of the email marketing revenue from triggers.