Thread

I ran a couple of tests today and yesterday and got pretty decent results. 0 discrepancies. Only a couple of spam filter fails. All results indicated that they were complete. Nothing was pending.

Now I go back tonight and look at my previous results and both my tests failed miserably. Hundreds of discrepancies. Spam failure in all but a few clients. What has changed in the interim??

Simplelives,

When you say “previous results” - are you referring to a test that you viewed from your archive?  If that is the case, nothing has changed.

Or are you saying that you ran a new test using the same email?  Did you make any changes to the email since yesterday?

I viewed two tests from the recent activity box on my dashboard which I guess corresponds to my archive right now. I clicked View Test Result and the results were **very** changed from what they had been about 24 hours previously in one case and 6 hours previously with another test.

In fact, when I first saw the Spam Analysis in the archive/recent activity box, each test only showed 2 red dots indicating spam failures. After viewing the tests and returning to the archive, both tests failed in all but 3 email clients.

I was going to run another test to see if this was just a glitch. But in the archives box, it says “You may re-run any of your acid tests, free of charge.” but when I click on the Re-Run Test button when viewing test results, it says “
NOTE: By submitting this form 1 test credit will be deducted from your account.” and I wasn’t sure which to believe.

Thanks for your help.

Yes, your “last five tests” on your dashboard and your archive are the same.  You will not be charged a credit if you click on “view test result” from either location.

The “re-run” test button when viewing a test result is a different story.  That allows developers to run a new test without having to fill out the entire form over and over.  Yes, it does cost one credit each time you re-run a test because it presumes that you have made changes to the code. 

I would recommend that you login, go to your archive and let me know the subject line for each of the tests for which you say you are getting completely different results.  We will investigate further from there.

Thanks!

Thanks, Michelle. The two tests in question are:  “February 2011 from Simple Lives - Preview” and “February 2011 - Preview.” The Preview in the subject lines were added by Campaign Monitor which sent the emails I used for the tests.

And just so you don’t think I was imagining things , please see another thread I started at http://www.emailonacid.com/forum/viewthread/63/#121 . I posted a question regarding spam failures about 28 hours after I ran the test “February 2011 - Preview” and note that I said that I only had two spam failures at the time of that post. That was all I was seeing at that time.

Appreciate your help with this.

OK, I will compare the “discrepancies” from each test but I can tell you right now, that the reason you are getting different SPAM results is due to the subject line being different. 

The three, most important factors for spam filters are: Subject Line, From Address and the Contents of the email. 

This discussion brings up an interesting point - it’s really hard to decipher which of the factors I just listed are actually causing an email to fail.  We are considering adding a few more tools that help people troubleshoot. Until then, if all you changed was the subject line, that’s probably what is causing it to pass/fail in different filters.

And thanks again. Just so I’m clear, it’s not the difference between the results of the two tests that bothers me. It is that my initial results for both tests were so different from what they are now reporting. Originally, both tests reported zero discrepancies and only two spam failures. A day later, both tests had hundreds of discrepancies and many spam failures.

And as I look now at the code analysis for one of these tests, it says “1080 Total Discrepancies Found 360 Critical 0 Moderate 0 Minor” which obviously doesn’t add up at all. The total found seems to be tripled.

??

OK, we ran your two tests and have confirmed that there is an issue with our initial reporting of discrepancies - we found this issue in two locations: in the top right corner and in the “Code Analysis Summary.” 

If you look at the “Code Analysis” for each individual email client, you will see accurate results. 

We are very sorry for this inconvenience and will resolve the issue right away. These updates will be included in our next release scheduled for 2/6/11.

Cheers simplelives, and happy testing!

Outstanding! Thanks for all your help.