I like bouncing

We are about to undertake an interesting exercise - testing our bounce reporting mechanisms.

Every email deployment results with a number of bounces, and our mailing software duly lists a variety of reasons for why messages have bounced. Do we trust them?

I guess the answer is no - which is why we are going to strip out all the bounced emails and send them a special message asking if they receive them or not. If not, then we will cull them, but I bet there will be some who say yes.

If I am right, does this throw up the whole issue of how accurate email metrics are?

My gut feel is that they are inaccurate, but consistently so - which is why we always encourage clients to look at trends rather than actual numbers when assessing campaign metrics. Trends are more valuable anyway.

We recently established a really good upwards trend for a new client by making a series of improvements over a period of weeks - opening rates, responses and orders have all grown consistently. Historically the client, with a DIY email set-up, had little faith in the medium. Now they love it!

We'll report back shortly on the bounce test results - could be interesting!