Compass Designs .net

At Joomlashack, we know that as on online business, email deliverability is very important. Whether sending newsletters or transactional emails from systems, they won't do much good if they don't reach people's inbox.

We use iContact for our newsletters, and have had great success with it. Recently though, we have been looking at supplementing it with another system for our transactional emails/ecommerce etc.

Once interesting alternative is Infusionsoft, which has a rich API, and ecommerce functionality built right in.

But what about deliverability?

Mr Tester.... start your engines!

Testing Newsletters

The first thing we wanted to do was test a newsletter to get a basic sense of deliverability. We simply split our email list into two and them imported one of them into Infusionsoft.We then sent an identical message.

As Infusionsoft won't track actual clicks by users from the email, we actually measured this through Google Analytics by seeing who was actually landing on the site.

Here is seems there is a clear edge to iContact in terms of deliverability, but that's not quite the whole picture.

Infusionsoft has a system where they use different IP addresses to deliver their email. When you first import users, they are set to single opt-in. You'll then have to get them to click on a link (that you can send in an email) that will have them double opt-in. These two different statuses are sent from different IP's which different (and undisclosed) deliverability rates.

This is one aspect where iContact has an edge. Imported emails can be set to double opt-in, and there seems to be no difference in their IP/deliverability rates anyway.

Testing Infusionsoft Opt-ins

Having established a comparison between imported contacts in iContact and Infusionsoft, we then wanted to dig deeper and see how much Infusionsoft's deliverability could be increased.

We took our imported users in Infusionsoft and asked them to opt-in. A couple of weeks later, we sent another email (again identical) separately to the single opt-ins (those that never clicked the offered link to opt-in) and the double opt-ins.

Here we can see a very large improvement in deliverability once users opt-in. Not that surprising, but the amount of improvement was interesting. We can also see that when the two emails are considered together, the rates are roughly similar to the ones above on first import.

Conclusions

Its pretty difficult to make too many conclusions from these tests. Several I might make are:

• iContact and Infusionsoft have similar deliverability rates, with a slight edge perhaps going to iContact
• Getting people to opt-in is good (duh!)
• The bigger the list, the more full of dead emails gets - mailinator inboxes, people left companies or changed email addresses
• Neither Infusionsoft or iContact handle dead emails very well. They process bounces, but can't recognize an autoresponder saying Bob doesn't work here any more (with revenue based on list size, is it in their interests to?)
• Only a small portion of your list are your "tribe"

One last thing we like about Infusionsoft is regarding that last conclusion. You can easily track who is clicking on your emails and target them with more refined messages. Its the sort of thing enterprise systems like Amazon have been doing for years.

We'd love to hear from you if you have done any email testing between different systems please tell us in the comments, and you can join our testing of email marketing services project.