e-newsletter quits after first bad email address

So I got e-newsletter all set up, and imported 13,000 email addresses. This was surprisingly fast. However, once I started a mailing, it stopped on the third email, saying the email address was bad.

Now, there will be bad email addresses. That's the nature of user-generated lists. I can't imagine this thing has been in production and it's been usable if it stops running on the first bad email address, so... what am I doing wrong? How can I fix this?

  • aecnu

    Greetings Everyone,

    The biggest difference with the paid services is that they would not have allowed the bad email syntax to begin with and that is probably the solution here as well.

    Therefore I will see if I can get the lead developer @Cole in here with his invaluable insight into this plugin for his advice/advise for us.

    Though this may take a bit longer then a normal ticket, I will try to get him in here asap.

    Cheers, Joe

  • David

    Yeah, you need to work on this product. I have to move on, because I need to make my system work. But after having done this for years, the one thing you need to know is that you can never be absolutely sure the email addresses you're sending are clean. So the sending routine has to be able to drop or ignore ones that are bad and keep on going.

    To fail mid-send is pretty bad, especially since you then have to start over. The mailer must be resilient enough to handle some glitches.

    Honestly, and you guys know I like you, but this product is not ready for prime time. At best, it's a beta product. You're missing critical functionality -- email address correction, export, etc. and you have unacceptable resiliency weaknesses.

    It has potential, but it's not yet a solid product. Sorry 'bout that.

  • aecnu

    Greetings David,

    Thank you for your input, it is appreciated.

    Absolutely their is bad syntax in email and that is everything you are describing.

    It never got a chance to send, so it is not undeliverable's.

    Bad syntax joe@, joe@domain, joe@.com

    So this has to be a bad syntax issue no doubt and it locks up when it hits this type of stuff. Funny how that works.

    Am I on the right track or what?

    Please advise.

    Cheers, Joe

  • David

    Yep. Remember that a list mailer is supposed to be a machine. It's supposed to keep on chugging. Otherwise, you could just go in and individually email everyone. The whole point of list mailing automation is it just keeps churning away.

    Also, remember that not only are you going to have bad names or bad addresses there, but users are going to regularly enter search terms in the email signup form, spam operators are going to try to inject trash, etc. So your form -- even if it has entry-time validation -- is going to be assaulted by crap and that crap will enter the database.

    In your case, however, not only does it fail, but as far as I can tell from other user reports (I didn't dig into it because I need to move on to another solution), but as far as I can tell, you not only fail on a bad address, you have no way -- outside of phpMyAdmin, to fix that bad address.

    So, there's a run-on sentence for you, but you get the idea.

  • aecnu

    Greetings David,

    Thank you for your response as always.

    It is back to syntax again to keep it chugging even if the address is undeliverable.

    I use Get Response and one cannot add a an email address that does not pass the syntax test - live and upfront.

    And if it is undeliverable three times they are automatically burned from the list.

    This is what needs to be happening to get this up to speed in my opinion.

    I know for a fact that Cole is working on a heavy duty update, that he inherited this plugin from the used to be with us Andrey who bailed leaving Dev stuck with a bunch of his .... sub standard plugins. i.e. Directory, Classifieds, and yes including enewsletters.

    He was handed a bag of garbage and if anything like the other two plugins it is almost a complete rewrite to get them up to par and usable.

    I know Cole has indeed added a couple of band-aids, but this plugin has gaping wounds as did the others. But this is certainly not the only plugin Cole has to deal with by far.

    Hopefully Cole will have his new release out soon with some of these issues resolved and of course hopefully all of them.

    Cheers, Joe

  • David

    Joe, thanks for the honest discussion. I had a feeling that's what's happening. Best practice is to never assume data to be pristine, so while you might be lucky, consider this scenario:

    Let's say you're doing production mailings of, say, 30,000 pieces. Your site gets hacked, and someone inserts a batch of email addresses. You could recover from backup, or you could hand-trim. But if you missed just one email address, you don't want the thing to stop -- especially if it's a client site. I've had clients with absolutely no tolerance or sense of humor. What if, for example, there's an even you're promoting tomorrow morning and everything stops. That's resilience.

    As for your plugins, you guys are great, but you probably need to save your customers some heartache by giving them a warning. For example, had I known this plugin wasn't ready for prime-time, I would have tried it anyway. But I'd have been prepared for some hacky time, rather than wondering if WPMUDEV doesn't understand product quality (okay, I know you guys do, but you get the idea).

    You could mark these questionable products as either beta, or even be more honest and simply post that you're not yet satisfied with their quality, but they're available if you want to take a chance.

    The big thing you want to avoid is sucking down someone's time. I generally get one day (if I'm lucky) a week for development. If I lose that because I was misled, I'm very cranky. That's incredibly valuable time that a vendor didn't respect. But if I know I'm walking down a potentially troublesome path, then that's my decision.

    Don't sweat it too much, but think about quality vs. quantity, best practices in a production environment, the insanely incredible value of your customers' time, and how you want your products to be perceived and how that impacts brand equity.

    You know. The easy stuff.

  • aecnu

    Greetings David,

    Thank you so much for the great discussion and I totally agree with you 110% on all the points in your last post especially the quality vs. quantity.

    When I think back to that time when all this hit the fan with the previous developer and we actually pulled the Directory plugin because it was such a mess, the back lash, even though it was a mess, proved worse then letting it ride and dealing with helping each person - some of those threads from taking it down made it into the hundreds of replies and posts - horrible.

    In this case here though with the eNewsletter plugin, Cole coming into it green so to speak and by letting it ride, he truly has direction by Members like yourself and the others that offer their opinions about what is wrong with or missing from the plugin which is truly valuable.

    I do so look forward to Cole's real first release and hope he has incorporated some of the features and "repaired" the bugs that members like yourself have mentioned.

    And I truly hope that is sooner then later.

    Sincerely, Joe

  • aecnu

    Greetings hhart,

    Thank you for the great question.

    This is a question with no answer until the lead developer either releases it for beta or production.

    As put out long ago by the Big Boss, there is to be no disclosure of possible release dates nor road maps until the release happens due to member complaints in the past when dates were given and were not met.

    Sorry that I cannot give you any more information then that except it has not been released for support staff beta as well.

    Thank you for your understanding and sorry for the inconvenience.

    Have a GREAT upcoming weekend!

    Cheers, Joe

Thank NAME, for their help.

Let NAME know exactly why they deserved these points.

Gift a custom amount of points.