Does anyone know how to set up a split test to your e-mail list?

Thanks!

We will be working on implementing a split-testing engine, but there is no ETA on when this will be available...might not be this year. It will be your basic split-test engine...cycling subject lines and other stuff.

One workaround that I've seen is assigning contacts to several dummy user accounts via a round robin (this assumes you have spare user accounts)...and then searching for everyone who's owner is "User1" and adding them to "campaign1", then searching for everyone who is "User2" and adding them to the "campaign2"...get the idea? It's a manual process.

So this is how it works: I setup a round robin that evenly distributes contacts to my user accounts. When someone comes in via a landing page, they are assigned to one of my test users. This way I have evenly segmented my list and can run a split test by searching for the owner of the contact records and then adding them to whatever campaign or batch email I want to split test. It's not pretty, but I've had a client use this technique recently.

Ugh! What a process!

I have about 10,000 people that I'm going to be moving over to a new autoresponder sequence, but i'd like to test the first promotion that i send to them. Couldn't I take half of those names and put them in a group and then assign the other half to another group? I know it's not exactly the best solution, but I was thinking this might be a workaround for now since I only want to do one test to an existing group of people.

The problem your going to have is that there is currently no search functionality for pulling up a list with a limit of 5,000 people...it's easy for a developer to do on the back-end in SQL, but not on the front-end. What I would do is create 2 groups (group A and group I would first limit my results to only contacts who have data in their email address field, then I would just search for everyone who's last name starts with "A", and add them to group A, last name starts with "B", add them to group A, and so on until you get about 5,000, then dump everyone else (contacts who are not in group A) into group B. Not pretty, but it is doable. That's how I would do it in my database, you may have other criteria other than last name that would make more sense when segmenting your database. Hope that helps.

You could also set up a round robin and evenly distribute your list between 2 users ...that would actually be pretty fast

Hey guys, not to complicate an already complicated issue but...

Another way to accomplish this would be to have a custom field that represents which "test group" they belong to. a,b,c,d,......etc

Then you could set conditions in your campaigns so that if "test group" equals "a" then run campaign "a", if "test group" equals "b" then run campaign "b" and so on.

One good thing about this is that you could dump your contacts into excel with just their "Name" and "ContactID", then add one field (column) in excel to represent their "test group" type in whatever you want their "test group" to be and when you are done import the list as an update to the contacts, just map your new column from excel "test group" to your new custom field "test group" in the CRM.

That way you could just clone campaigns or marketing pieces, change the headline and set them to go only to people with the corresponding "test group" field you want to test.

Thanks,

Micah Mitchell
Infusion Certified Consultant

Just wanted to let you guys know we have finished our desktop application that links to your infusionsoft application and allows you to create split test groups of your people.

You can grab any percentage of any or multiple tags or your whole list. Then split them into any number of splits to test the effectiveness of your emails.

For example if you choose to take 20% of your whole data base and split them into an A/B test you can do this. It will Randomly grab 20% and then split that 20% into two groups randomly.

Once the split is complete you can find each group and send out your broadcasts. Use Infusionsofts email reporting to track the effectiveness of each. You will then know which one received the best conversions and then send the 'best' email to the other 80%.

You will grabbing that extra money you have been leaving on the table for so long from not testing. If you don't test your content will never know you could be making more money..

http://infusionpros.com/isplit/1.php

All my best,
Steve Hofstetter
Owner
Infusionpros llc

For those who haven't had time to look at Steve's good looking desktop application, here's another A/B Split Test generator.

Here are some of the features:

• This version is web-based, which gives you server-to-server connectivity (speed)


• It allows you to split into as many segments as you wish.


• It allows you to set percentage levels for each test segment you generate.


• It takes 3 clicks to create your segments (as in 1, 2 and Finish)


• The actual process takes place as a background task, freeing your screen, your computer and you to do other things - what's a Moldingsoft tool without magic?


• Notifies you by email when the background is finished.


• The price is designed to fit in everyone's budget: it's free, now and forever.

Get it here: http://absplitter.moldingsoft.com/

Enjoy!

Best,
Bob

techportal wrote:

Here are some of the features:

• It allows you to split into as many segments as you wish.


• It allows you to set percentage levels for each test segment you generate.


• It takes 3 clicks to create your segments (as in 1, 2 and Finish)

Love the originality Bob. Guess that is why I had all those ISplit opt-ins from you a few weeks ago... All you need now is for it to be able to split into custom fields like mine and you will be all set

All the best to you,
Steve

Steve,

There isn't much originality in splitting a list. Infusionsoft apps live and thrive on it.

I am glad, though, that you point out that there are some differences. Yes, these difference are crucial and form the core of our system.

Some background... This system was built for a customer (name available upon request) who graciously said that we could offer it for free.

It is true that he pointed me to your site and told me what features he wanted and which he didn't want.

Neither of us invented the concept of split testing and neither of us invented Infusionsoft's tag system.

My customer had no use for custom fields due to the 100-limit imposed by Infusionsoft and that's why we don't have them in there.

Just like you, I would love to have been the original creator of the Infusionsoft tag system, but, unfortunately, neither of us can take credit for that.

So, aside from the things that neither of us invented, the main difference between our systems is in the processing.

With our system, you can click 1-2-3 and literally turn your computer off. The task will continue to process in the background rather than on a web page or your desktop.

This particular client has a fairly large mailing list and it was important to him that the tag assignment be at least 2-3 times faster than on your system and that it run as a background task, where it could not be interrupted. His office has Mac's and PC's and he therefore wanted a platform independent application.

In summary:

Yes, we use tags to segment mailing lists.

Yes, we use percentages to limit the size of segments.

Yes, it is fast(er).

Yes, it is password protected.

No, it isn't a platform dependent solution (runs on Mac's, PC's and Un*x workstations, your iPod or Blackberry, etc.)

No, it doesn't require that a program or window remain open or that the computer even be turned on in order to run.

No, it doesn't cost a penny.

Finally, since you are giving me this opportunity to showcase my stuff, I should also point out that a simpler version of this tool was already included in our infusionWP Wordpress/Infusionsoft plugin. The only major difference there is the background processing option.

All in all, the Infusionsoft universe is fairly small. Most users have the same needs, even if they vary slightly at times. I was therefore not surprised when you offered the same identical support package as Infused Systems, only a day after they blasted their announcement.

Our tool, at least, offers some very distinct features from yours and is technologically speaking much more advanced than yours.

The better mousetrap is as American as apple pie. I'm already thinking of ways to improve it with Micah's selector routine. So, Steve, in the interest of all our friends and Infusionsoft users, seize the opportunity and come up with a yet better mousetrap as well.

I wish you continued success.

Best,
Bob

P.S. I forgot to point out, that although we do not use them, the idea to use custom fields originally came from my friend and partner, Micah Mitchell, as evidenced in this very thread.

Bob,
touche!
Well played my friend. I am proud to be a man that can admit when he was owned. I was owned on that post - as if you had it pre-written

And yes you are right about the blast. That was pure randomness in this small Infusionsoft world we are in.

You are doing great things the the Infusionsoft plug-in world. Keep it up.

Oh yeah. I will most likely be purchasing your WP plug-in for one of my new 'non-infusionsoft' ventures in the very near future.

All my best,
Steve