We do a lot of work with clients to help them improve reputation to improve delivery issues at top ISPs. In doing so, we find ourselves doing the same sort of analyses over and over. This is the kind of thing that every marketer should keep an eye on, even if they are not currently having a delivery problem.
These three analyses are crucial for getting to the root of an existing deliverability problem and for preventing future problems from occurring:
1. Are there pockets of complaints or unsubscribes in my mailing program? Most every marketer knows that complaints are a key factor that drives blocking by ISPs. While being on feedback loops and removing complainers from your file is important, it is better for your long-term deliverability prospects to find and eliminate the source of complaints.
This is where an analysis of where your complaints come from can be a huge help. For example, an analysis of complaint data from a company that sends sales notices to both frequent and non-frequent buyers may reveal that the non-frequent buyers complain 10 times more often than frequent buyers. Or the reverse could be true.
Source of the subscription can also be an important factor. So, customers who are opted into the program assumptively through the buying process may complain more often than prospects who subscribe to your email by proactively seeking out the sign-up form. And of course you are very likely to find that messages that are highly relevant -- either because of behavioral data or subscriber preferences or both -- are less likely to generate complaints than more generic messages.
2. Are there pockets of open/click in my mailing program? This is the same type of analysis as above, but now looking at the open and click data. This data can actually help you predict problems. Subscribers who don't click or open are likely to be on a path to complaining eventually, especially if you send at a high frequency. As with the complaint and unsubscribe data, you can use this information to make changes to improve your program. You can also use this information to engage in win-back strategies with the inactive subscribers.
3. What's changed? Ideally a company should never make any significant change to an email program (e.g., frequency, data source, content types) without a detailed plan for testing before, during and after the change to check for deliverability impacts. The reality --especially for big companies with multifaceted email campaigns -- is that changes both big and small are happening all the time. So this means you need to dig into your deliverability data and map any changes in your inbox placement rates and other metrics to changes in the program.
For example, a change from three emails per week for a given segment to five emails per week may map to a decrease in inbox placement at many top ISPs. Your analysis will probably show a progression of effects: the increase in frequency will lead to an increase in complaints, which leads to a decrease in your inbox percentages. This effect can be masked in a big program if the change was only made to a small part of the file. This is why analysis of the segment that experienced the change is crucial.
The same is true for new data sources. When you start to collect emails in a new way or from a new partner, the deliverability failures are going to come down the line after the bad data and spam traps become a big enough percentage to ruin your sender reputation.
| About us
| Pricing
| Privacy Policy
| Anti-spam policy
