Episode 58 - How to fast track CRM Marketing performance

Did someone forward this to you? If so, click to subscribe

📰 CRM Marketing Results, Monitoring & Testing

The problem with open rates 

Its cool to say open rates rates dont matter and that focus should be put on deeper indicators of behaviour, like clicks, conversions or engagement metrics. 

While deeper indicators of behaviour are the north star, the reality is if your email isn’t opened your message wont have the same impact.*

*Occasionally very clear subject lines can positively impact behaviour even without being opened. 

The reality is tracking open rates on iOS became significantly less reliable due to Apple’s Mail Privacy Protection (MPP), introduced with iOS 15 in September 2021. 

Here’s a clear breakdown of what the issues are and why they matter for CRM marketers:

• Preloads tracking pixels: When a user opts into MPP (which most do), Apple Mail downloads all remote email content, including tracking pixels, regardless of whether the user actually opens the email.

• Masks IP addresses: Apple hides the recipient’s IP address, preventing location and device data from being captured.

• Spoofs open times: Opens may appear as happening at odd times (e.g. 3am), making behavioral triggers unreliable.

MPP inflates open rates (emails appear opened even when they werent), reduces segmentation accuracy (e.g. engaged users can include iOS users who havent opened) and it can negatively impact send time optimisation tests and automated triggers e.g. if they open email 1, send email 2. 

While this is challenging, Android is unaffected (especially gmail users) and for that reason, where possible you should segment Android users from iOS and base your open rate metrics on Android customers, because open rates and subject line testing does matter. 

It matters because an open leads to the next step like clicks, engagement, purchase, referrals or signing up for a loyalty program. 

Aligning measurement to communication goals 

Every CRM Marketing initiative needs a goal and you need to know if the goal was impacted by the message.

The goal varies based on the lifestage or moment of truth being addressed. 

Typically your goal should be single minded. 

Examples could be 

• Conversion e.g. Upsell, Crossell

• In-app or on-site engagement 

• Usage of a product already owned  

• Referrals  

• Gather feedback  

• Signing up for a loyalty program 

• Taking up discount following an apology offered after a bad experience 

The effectiveness should be determined by the proportion of your customers that took the action vs those that did not. 

All customers have baseline propensity to take an action or a baseline inherent value that they are born to you with, this is based on product / brand / customer fit and how they realise value with your product or service. 

So to determine the effectiveness of your communications on behaviour, you need to examine the proportion of the audience that would have took the desired action anyway vs the proportion of the audience whose behaviour was incrementally impacted by the CRM marketing initiative. 

Baseline customer behaviour can be approximated using control groups. The impact of CRM Marketing is determined by the incremental behaviour generated within the treatment group and above the baseline identified via the control group. 

Calculating incrementality 

CRM Marketing success hinges on structured experimentation. 

Every test - win or fail drives learning, and learning drives uplift. 

Success is a function of the number of tests you run and how fast you learn from those tests and execute. 

But there’s a catch: if you wait for perfect results, you’ll move too slowly to make an impact.

Big tech companies like Meta and Amazon demand 95% statistical confidence because they run experiments on millions of users. 

But for most brands, especially when testing small cohorts or niche journeys, that threshold is far too high. 

It leads to analysis paralysis: slow rollouts, rejected ideas, and missed retention improvement, because on smaller audience sizes you need bigger differences in outcomes to hit the magic 95% confidence. 

To get over this adopt an 80% confidence threshold for CRM Marketing experiments. 

This faster approach gets you to optimisation far quicker, while still being right 8 times out of 10. If something doesn’t work at roll out, you can always, roll it back and move on.

Retention is a game of iteration. 

The faster you learn, the faster you grow. 

And the payoff is huge if you get it right, as we covered in episode 13. 

Final Thoughts 

Effective CRM marketing lives and dies on your ability to measure what matters, test what works, and move at speed. 

Yes, open rates are less reliable on iOS - but they still matter because every subsequent action starts with an open. 

But opens alone aren’t enough. You need a clear goal for each message and a method to isolate its true impact, through control groups and confidence thresholds. Adopt 80% confidence because CRM Marketing impact is directly correlated to testing velocity. 

Until next week, 

Tom  

Do you need help with Customer Retention?

When you are ready, contact me to discuss consulting, my fast-track retention accelerator, courses, and training. Or if you are interested in sponsoring this newsletter, get in touch via [email protected]