Part 2 - Improve deliverability: How to glide through spam filters

We dealt with validating customer email addresses last week in this special deliverability series and today we are going to tackle the issue of spam filters.

One of the biggest concerns when sending out an email, is whether or not it will be marked as spam by any of the filters it passes through on its way to your customer’s inbox. Even when adhering to authentication  standards such as SPF & DKIM, using "warm" IP addresses and ensuring that your email lists are clean; the spam filters have the last say on your deliverability and their decision-making logic is a well kept secret.

There are no well documented algorithms, keywords or methods used to filter emails. What makes the challenge even harder, is that spam filters constantly change their rules to keep up with new methods used by unscrupulous spammers.

Just as illegitimate mass emailers learn the rules to stay one step ahead of the filter software, so too do spam filter rules change and we have to understand the parameters which cause some emails to be cast away.

Don’t get labelled as spam: structure your email correctly

If your email "looks" like spam, it’s likely to be marked as spam by the receiver. You may also find that your campaign starts off with a good  deliverability rate, but this quickly deteriorates as the filters "learn" from their users. So how do you structure your email?

The underlying HTML needs to be impeccable. There are rules that must be adhered to when building an email and simply converting a word document to HTML format will not conform to the strict standards.

• Make sure you remove misunderstood "spammy" words and avoid using excessive punctuation or CAPS.
• Identify yourself in the FROM 
• Make sure your subject accurately describes the content.
• Have the correct balance of images and test. You’ll want to send it as a Multi-part MIME email (HTML & Text versions).
• Make your email short and sweet so that it is easy to read. You don’t want it being mistaken for spam just because at a glance you could not tell.
• Make sure you have an unsubscribe link so people don’t click the spam button instead. 
• Unfortunately, following the above ‘rules of thumb’ will still not guarantee that an email will go straight to the inbox.

The solution is to test, test, test

Start with your content -  test your email copy using the many free tools available, then test the entire email as a package by monitoring inbox open rates and previewing emails in multiple email clients. This is essential to an ongoing, successful email send.

Adjust and take action

Many of the top ISPs provide a mechanism to report back to the sender whenever a mail is reported as “spam” or “junk” by the customer. Called Feedback Loops, a standardized report is returned as an email which provides the sender with the information needed to remove the email address from the list.

ISPs provide a subscription method via a web form, usually on the same postmaster site that is used to register for whitelisting. By entering into the agreement that the sender will monitor and take action on complained email addresses, the reputation is increased and the emails being sent are legitimized.

In summary:

• Create a well structured email
• Check the content
• Test before you send
• Adjust if you are getting complaints or see the open rates dropping

We hope you are enjoying our email deliverability series. If you haven’t yet downloaded the white paper, why don’t you do so now? It’s full of easy to understand advice on how to improve deliverability of your emails. Plus it’s free!

Keep following the series, next week I will be covering the big topic of how you get your customers to OPEN your email!

Alex Papadopulos 
striata.com

Email Deliverability Part 1 - Improve deliverability: validate customer email addresses

Getting the email into the inbox is so much more than just creating something that looks great, passes spam tests and having your database of addresses ready. Before hitting send it’s vital to validate the email addresses in the customer database that you are sending to. Our special deliverability blog series starts here!

This is an important step in your deliverability efforts because continually sending communications to a failed email address can result in your emails being blocked by the receiving mail server. Email addresses are captured incorrectly because of user typing errors, but all too often email addresses have been gathered over the phone or via hand written correspondence, which can then lead to misreading and input errors. The good news however is that a large percentage of these can be corrected before the first email is sent.

4 Easy steps you can take to ensure proper email validation

1. Check email addresses for correct formatting
   This first step can be performed at its most basic level by checking for the basic structure of an email address, i.e. name@domain and excluding obvious errors such as name@nodomain.


2. Measure emails against common rules
   By running through each email and validating against common rules, you can eliminate the incorrect email addresses. This helps your deliverability rates and also gives you the opportunity to reconnect with your customer and verify the information you have on file. This is not an intensive process, so it can be performed every time an email is sent out to avoid badly formatted emails from even attempting to enter the delivery queue.


3. Scan for common mistakes
   Take the process one step further by searching for common mistakes. For example, a common manual input error would be recording the email domain yahooo.com instead of yahoo.com by mistake. By using a database of the most common errors, a report can be generated showing the suspected incorrect email address and suggesting the correction. Unfortunately, this method is open to mistakes and as such should always be validated manually – (for example, while Yahooo.com is almost always an input error, it is not guaranteed that this domain does not actually exist). It is also advised to remove placeholder email addresses such as no@email.com.


4. Run every email address through a server, which does a 'live check'
   The last method to verify email addresses is the most accurate, but unfortunately, also the most intensive. It requires running every email through an application which makes a connection to the SMTP server of the domain and "asks" whether or not the email address is valid. Due to the fact that this is a live check, results can be regarded as almost 100% accurate. For larger lists, this can take a lot of time and should only be performed once (or periodically) rather than checking emails individually as they are being sent. Also use your delivery report information to remove any bounces and undeliverables from your database, as this will help you avoid spam traps.

These methods should be applied to any email list or customer database, whether marketing based emails or those that are transactional in nature, to ensure optimal deliverability.

It is important to note that, as each of these techniques has its merits and pitfalls, you’ll need to determine which one applies best to the type of email, source of the email list and number of emails you are sending out. Email is a powerful tool, so make sure it works for you.

For more strategies and tools on how to ensure your emails reach customers’ inboxes, be sure to download our new Deliverability White Paper "7 Steps to better email deliverability"

Look out for next week’s installment of this special deliverability blog series; I will be focusing on spam filters and how to avoid them.

Alex Papadopulos 
striata.com

How well do you deliver? Follow our deliverability blog series and make your emails count

The discipline of email distribution is riddled with acronyms and industry terms. DMARC, SPF, DKIM, CTR, pre-headers, engagement rates... are just some of the expressions that you are faced with on a daily basis when involved in sending bulk or triggered emails.

We’re making it easier for you to get in with the lingo

When it comes to sending emails, the one area that is full of the lingo that every person involved needs to know and understand is around deliverability – the art of getting your emails into your customers’ inbox. And as fantastic a tool as email is to send bills, statements, policies, marketing, promotions, lifecycle communications, letters, etc., it has to reach the recipient’s inbox in order for it to be effective in any way.

Bearing this in mind, we thought we’d share some of that knowledge with the launch of our Deliverability White Paper and Deliverability Blog Series that will give you the information you need to make your emails count!

You can download our White Paper '7 steps to better email deliverability' and make sure you’re glued to your inbox over the coming weeks as we delve into the following topics in our blog series to help you understand the subject of deliverability:

1. Validate customer email addresses
We look at why data hygiene is important in getting your emails delivered and give you some tips on fixing common errors and purging those bounces
2. How to avoid spam filters
We will give you the tips on structuring your emails correctly, appraising your content and testing the email to see the results before you send
3. Getting customers to take action
Your email has been delivered. Now what? Your customer engaging with your email is also important in deliverability, so we take you through the details and advise on how to increase engagement
4. Behind the scenes technical stuff you need to know
This blog tackles all that jargon by giving you the details you need to know and action in a manner that doesn’t make your eyes glaze over.

So why is it so important to understand the ins and outs of email?

There is usually a separation of tasks and activities between business and IT/ ESP when it comes to email distribution, with the focus on deliverability falling into IT / ESP remit. However, the processes and decisions on implementation need to be understood by everyone involved, because getting an email delivered and getting customers to engage, involves a number of moving parts, all of which have to work in unison.

Without a focus on deliverability, all efforts in sending emails to customers, is diminished. Systems, people and processes are all crucial in the quest to get emails into the inbox.


Keen to get some expert advice on deliverability? Then be sure to sign up for our deliverability blog series

Check out our new Deliverability White Paper '7 steps to better email deliverability' as well - you can download it here

We hope that you find the information useful and if there is anything more you would like details on or if you would like to chat to an email specialist, please contact us.

 Mia Papanicolaou
striata.com

What do they do with my data?

I recently purchased my first residential property, so I have had firsthand experience of the rather harrowing process of applying for a mortgage through a mortgage originator. The application process required handing over a plethora of personal information, including my personal identity number, bank details and salary slips. I understand that the financial institution needs this information to register and validate me, but if a criminal got hold of it, I could lose everything.
This got me wondering (in a slightly panicky fashion): Can I trust my suppliers to keep my personal information secure?

I assume that because I’m dealing with a reputable company, my information will remain safe. But sadly, thanks to Google, my panic level rose as my search result revealed numerous data breaches leading to millions of personal records landing in malicious hands.

What companies should be doing:

Given the rise in these malicious attacks and the potential impact, companies should consider a holistic approach to protecting data wherever it is – at rest, in motion or in use.

I recently read an article quoting Tim Matthews, Senior Director of Product Marketing at Symantec: “The good news is that there is a straightforward regimen to help stop these kinds of risks.”

Matthews recommended that companies look at the following security measures:

1. Installing device control: where no copying onto a disk of any type is possible, or via Bluetooth or Wi-Fi data transfer.
2. Endpoint data loss prevention methods: which prevents sensitive data from being copied - copying of data or files is blocked based on content.
3. Encryption: these types of programs render some or all data unreadable by anyone who does not have proper authorisation, should they bypass the access control to these documents. End-to-end encryption is preferable because at every stage whether data is in use, in motion or at rest, is it encrypted and secure, and it is never ‘in the clear’.

Furthermore, documents that contain sensitive information, such as my electronic statements, should always be encrypted and password protected. In an online environment, companies can have firewalls and monitoring facilities where they can detect a breach. Whereas once a document is emailed out, the biller has no control over who can get access to that customer’s information. The sensitive information in these documents can include credit card information and personal identification details like ID number and date of birth. The only way to prevent unsolicited access is to password protect the document.

I wholeheartedly agree with Matthews' statement: “With such well understood defences available, companies really have no excuse for not putting them in place.”

What should I be doing?

Simply put, I need to be cautious and vigilant. Always find out who is receiving my personal information and ensure that they implement the appropriate defences.

The cost of a data breach

The 2012 Verizon Data Breach Investigations report says that in 2011 there were 855 data breach incidents involving 174 million compromised records.

The personal cost of a security breach to each compromised individual is potentially massive: a sudden zero bank balance or complete identity theft. But the cost to companies can be catastrophic. From Symantec’s Cost of Data Breach study, United States (2011), the average cost per data breach for an organisation is $5.5 million, or $194 per record.

While customers need to be cautious and vigilant when sharing personal information, organisations collecting customer data must ensure that they implement the necessary defenses against data breaches - don’t take the risk. Increase your security and reduce risk with 'Push' eDocument delivery, call us for a free consultation.

Simon Johnston
striata.com