Research: ESPs might get you blacklisted

Guest author Joshua Geake shares his concerns about the alleged super deliverability all ESPs seem to have. And with good reason, his short research shows.

Which ESP is best?

There are hundreds of companies worldwide that offer email services but how can the average company that doesn’t understand (nor wants to understand) SMTP, DKIM and FBLs know which one to go for?

Analyzing Deliverability

I analyzed a random selection of marketing emails that landed in my inbox over the past few days against SenderScore, Sender Base, Trusted Source and MxToolbox’s blacklist checker. The scores gathered from these services will give a representation of how an ESP operates.

This includes whether it follows best principles and sends authenticated emails, unsubscribes Feed Back loop (FBL) complaints etc. It will also give an indication of how an ESP’s clients behave on their shared IPs. Remember if an ESP shares its SMTP servers between its clients, your reputation and hence deliverability is shared.

Company Iceland Spotify Currys Overclockers Pizza Hut Play Phones4u
SenderScore 99 95 93 88 78 96 78
Sender Base Neutral Neutral Good Good Good Good Good
Trusted Source Minimal Risk Minimal Risk Minimal Risk Minimal Risk Minimal Risk Minimal Risk Minimal Risk
# Blacklists 0 2 0 1 3 0 0

They are even on blacklists!

From this relatively fast analysis I wouldn’t want to share an IP with Spotify, Phones4U or Pizza Hut. Pizza Hut is listed on 3 separate blacklists and Spotify 2, understandable given how poorly engaging their emails are. Being blacklisted most likely suggests that sender IP either sends out to a great deal of unknown or angry recipients.

That doesn’t seem like something i’d like to pay for

Phones4U meanwhile has a relatively low SenderScore which is explained by their low accepted mail rate of only 62.6%. In other words if you share this IP with Phones4U then only 62% of your emails will be accepted by a recipient’s email server. That’s before they hit the junk filter. I wouldn’t want to pay for that.

Do-it-yourself e-mail sending

My personal preference for avoiding ESPs and doing it yourself seems to be what Play do. It’s no surprise given how many millions of emails they send out every day that they do handle everything in-house. As can be seen they achieve as good if not occasionally better stats than the ESPs. Remember, ESPs will not do things necessarily better than you could yourself.

Check, double check, reputation check

So, in summary, if you think you need an ESP, check them out before hand. If they say Pizza Hut is one of their clients (and they no doubt mention how their conversion’s increased 450%) then sign up to Pizza Hut’s newsletter and check out their IP reputation for yourself. Unsubscribe and see how long it takes for you to be unsubscribed. Being locked into a long term contract with a careless or massively shared ESP must be really crap.

Does your ESP get you blacklisted? What are you doing to prevent reputation damage? Please share your comments.

About joshuageake

Joshua Geake is IT & Web Manager of BrightMinds. Specializing in E-mailmarketing, Datamining and IT. He has his own blog Geak It where he shares helpful solutions he has developed or researched.

  • Andy T

    I’d be intregued to see the kinds of volumes being sent by each brand too as well as the unknown user count.
    There is obviously an correclation between unknown users (hard bounces) and poor deliverability. Also High volumes and hard bounces (and spam traps) often point towards list purchase.

    • Anonymous

      Good point Andy. Lets try and get the e-mailmarketers and ESPs from Iceland, Spotify, Currys, Overclockers, Pizza Hut, Play and Phones4u here.

  • Greg

    You failed to mention, when using an ESP you typically have a dedicated IP(often many dedicated IPs) and are not sharing it. So PizzaHut’s blacklist is unlikely to affect your mailings.

    This article is bizarrely inaccurate. I am not saying it’s not possible to send mail in house, but sending mail from an ESP will not get you blacklisted.

    • Joshua Geake

      It has to be said that I did not title this post and (despite this) although the title does seem to evoke negativity towards the ESPs – this was not my intention.

      In my experience an ESP may well place you on a dedicated IP, however equally it may also push your send to a shared SMTP cluster. All I’m doing is brining it to your attention that an ESP may well not be the guaranteed inbox delivery agent they sell themselves as.

  • Andrew Barrett

    Senders get themselves blacklisted, not the ESPs that provide the platform. The senders themselves are far more responsible for the success of their own e-mail programs than most folks (including you, apparently) typically understand.

    And ESPs typically send from a range of IP addresses, so a block on one customer for sending spam is unlikely to impact deliverability for another customer on a different IP address.

    You really ought to get your facts and methodology straight before inflicting more disinformation.

    • Anonymous

      Hey Andrew, great that you join in on the conversation. Its good to see that a critical professional like you tries to clarify thing a bit. I’m no deliverability expert, but I see the point Josh is trying to make.

      If you are with an ESP and sending with shared IP’s you have a shared IP reputation.
      As an ESP might not gét you blacklisted, all senders might be falsely or incidentally blacklisted in the course of the year. ESPs usually manage the deliverability and are responsible to try and get you and the rest of the people on your shared IP un-blacklisted.

      In conclusion, if you get put on a badly managed shared IP, you might be getting a blacklisting for free. So what Josh says us true, checking your ESPs shared IPs deliverability before getting into a long term contract might be very smart.

      • David Romerstein

        “If you are with an ESP and sending with shared IP’s you have a shared IP reputation. An ESP might not gét you blacklisted, but all senders might be falsely or incidentally blacklisted in the course of the year.”

        The sending reputation of a given IP address is the result of ALL traffic from that IP. If you’re on a shared IP address, you have to expect to share the reputation of that IP address. If it’s bad enough to warrant blocking traffic from that shared IP, you haven’t been “falsely” blacklisted – at the IP level, a receiver can’t distinguish your mail from the bad guy that’s also using that IP address.

        The obvious solution here is, “don’t use shared IP addresses”. If your ESP demands that you be on a shared IP, spend a little money and move somewhere that gives you a dedicated IP.

    • Joshua Geake

      Andrew, as I’ve mentioned below – I did not title this post and (despite this) although the title does seem to evoke negativity towards the ESPs – this was not my intention.

      I’m not inflicting disinformation – only spreading thought.

      I have developed email marketing databases and software with levels of personalisation that are simply not available commercially and my SMTP clusters achieve better deliverability than our old (and very well known) ESP. So don’t accuse me of poor facts, poor methodology and deliberately inflicting disinformation, especially when it’s quite clear from your ‘blog’ that you copy and paste from Wikipedia and have barely any readers or comments – it’s even hosted on fucking

      • Stefano Bagnara

        I don’t know Andrew, but your arguments on his blog and the hosting doens’t sounds professional words (why hosting on wordpress should be an indicator of a poor blog?)

        You say you have an SMTP cluster with better deliverability. Would you mind sharing the IPs so we can, at least, check their scores and put them in the table, too? I don’t care too much of that scores, but if you point to them you should do it transparently.

        If you are on the “I will setup your email sending solutions, just pay me instead of another ESP” or you should explain what is the difference between you and an ESP. I don’t think you are proposing that DIY email servers have bettere deliverability and that you don’t need any expertise to setup an email server and and email sending solution, so I’d like to better understand what’s your point.

  • Loren McDonald

    Joshua, when you say “Play” do you mean If so, they do in fact use an ESP.


  • Jordie van Rijn

    I also recommend reading the article Laura Atkins wrote as a response to this article:

  • Stefano Bagnara

    You say: “I analyzed a random selection of marketing emails that landed in my inbox”.
    The fact that they landed in your inbox means they did a good job. You should analyze emails that didn’t land your inbox if you want to prove there is a deliverability issue.
    Who cares of the scores and the blacklists if the emails are delivered to inbox?

  • Mikael Ryd

    I agree with previous comment, plese read Lauras respose to this as this report is to short, general och full of assumptions and errors.