Comparing statistics

Clients often ask about their statistics.

Why is my open rate so low? Why have so few clicked through to our offer? Why did 9 people unsubscribe?

Quite often you sense that they are disappointed by the statistical measurements of their send out, and it is not unusual for them to blame you!

Problems of this nature should never occur. The responsible email publisher will always brief their clients at the start of their relationship.

Email is different to other media - recipient statistics are based on the behaviour of the entire audience - not on a supposedly representative sample and then extrapolated.

So, assuming (for the moment) there is no margin for error, email statistics provide you with a true and global assessment of audience behaviour. You simply do not get that with other media, where stats can be manipulated to give you the answers you want, rather than the real picture.

So yes, Mr or Mrs Client, sometimes you may not like what you see with email statistics, but there is good news too - you have a quantitive level on which you can base future performance targets. So if open rates are, say, 27%, you can look closely at why you think that might be, and work to improve that.

Taken in isolation, a single set of stats from one send out are useless.

To make stats work for you, compare and contrast from semd out to send out. Adjust aspects of content, delivery timing, subject line and other behavioural factors. Analyse their impact.

So what about that margin for error? Yes, it is true that email stats are not 100% accurate. Email publishers simply cannoy develop technology rapidly enought to kep pace with browsers, recipient hardware etc, so accept up to a 10% error factor. But.....the margin of error is reasonably constant, thus ensuring that comparisons of stats on stats are valid.

So explain to your client; it is not about the absolute numbers, it is about trends and interpretation.