RisMedia.com disseminates news and information through daily e-mails to more than 300,000 real estate and relocation specialists around the U.S. Last year, that business was in jeopardy; each day, more than 100 people opted out of the service to which they had double opted in. Â¶ Peter Featherston, RisMedia.comâs CEO, said he tried looking at e-mail analy-tics, but the process was difficult and results were muddy. Knowing that access to the right metrics might make a difference, Featherston switched e-mail service providers, choosing one that had a robust analytics offering. Within days he could see which stories people were most interested in and tailor his offering to meet those needs.
"Before we switched to StreamSend, we didnât know who had opened what story. We could track some thingsâopen rate, for exampleâbut it was very time-consuming, and our staff had other things to do than go back and see who opened which piece of mail," Featherston said.
The company made the switch last fall. Today, only a handfulâfewer than five per dayâopt out.
RisMedia.com took advantage of a piece of e-mail marketing metricsâclick-through ratesâto figure out how to give its customers more of what they were looking for, but that statistic is only one of many that marketers should be looking at. For example, comparing open rates with delivered rates can help you uncover a problem with your e-mail server or service provider, said Al DiGuido, CEO of Bigfoot Interactive.
Messages lost on the ISP side
"AOL has said publicly that they block 3 billion messages a day," he said. "It happens. You send a message, and according to your reports, the mail was delivered. That mail is being held on the ISP side. Youâd never know that unless youâre analyzing your data or working with a provider that could do that for you."
Youâll also want to compare your relative read rate with your relative click-through rates, said Lars Helgeson, co-founder and director of CoolerEmail.
"[To get the relative read rate,] youâll want to compare the total number of readsânot just the first time someone opens a message, [but] the total times a message was openedâand divide it by the total number of recipients. That rate is a pretty good indicator of how well a campaign on a whole is received. The more itâs opened, the better it is," he said. "Relative rate is a better indicator of acceptance of content."
Still, you will want to track unique opens separately, too, said Jordan Ayan, CEO of e-mail service provider SubscriberMail. "If someone is running Outlook and has the open pane open, every time they scroll past your message it counts as another open. Unless you can compare total open to unique opens, you arenât getting the entire picture."
Forward rate key
Forward rate is important for list-building. Most programs will tell you who forwarded your message and provide the e-mail address of the person who received it. While you canât e-mail that person directly, you can forward a new message to the original recipient list asking them to forward on your behalf. If you pair this request with a special offer for both the forwarder and the new recipient, you can create a strong lead-generation opportunity, said Dan Forootan, president-CEO of EZ Publishing, which offers StreamSend.
"Percentages are normally not very high with forwards, but you can approach those people and say, âThanks for forwarding our message; hereâs a discount offer for you and your friend.â It really ups the viral marketing rate," he said.
Bouncesâboth soft and hardâcan provide clues about your listâs health as well as the quality of your e-mail service provider. Too many hard bounces can mean your list is outdated or old. It could also mean your e-mail is being blocked at the server level. Youâll find out by looking deeper into your bounce report and analyzing the domain information of mail that received a hard bounce. Too many soft bounces, which happen when a userâs mailbox is temporarily full, might mean your message was too large or that youâre sending your message on a high-traffic day when itâs less likely to be read, said David Daniels, research director at Jupiter Research.
B-to-b marketers are less likely to track purchase metrics because sales cycles are typically so long; however, some marketers are bringing that data into e-mail tracking by providing codes in e-mails that stay with customers as they move through the sales process. This lets you see how long it takes for a lead to become a sale and match that sale back to a specific campaign.
Conversely, unsubscribe numbers, as in the case of RisMedia.com, should be a red flag. Even more important, Ayan said, is the reason people are unsubscribing. "About 75% of people who unsubscribe give you a reason if you let them. That reason could be âno longer in industry,â but it could also be something that you are doing wrong," he said.
Another perceived slight can provide a warning about your e-mail program. If you have a high subscription rate but a much lower confirm rate, it could mean several things, Ayan said. Your confirming e-mail might contain something offensive orâmore likelyâyou might not be sending out that e-mail soon enough after the person subscribes.
You can test this by setting up dummy accounts and seeing how fast you receive a confirmation. Experts suggest setting up dummy accounts on 20 to 25 of the biggest ISPs for general message hygiene. This helps you determine if your message got through, how quickly it was received and which mailboxâ"in" or "junk"âit was delivered to.
Where to get analytics
There are two main sources of
e-mail analytics: those that come from stand-alone e-mail or analytic programs, and those that are delivered by a third-party e-mail service provider. The main difference is fluidity. Outsourced products are more likely to change more quickly over time. Off-the-shelf programs follow a traditional software development process, and it can take six to 12 months to see an upgrade.
Before you make the move to a new product or service, Daniels suggests taking a try-before-you-buy approach.
"We recommend to clients that they use an e-mail service providerâs tools in a lab setting," he said. "Sit the marketing professionals down and go through a mock-up of what it would take to execute a typical campaign and retrieve analytics. See if they can do it in 30 minutes or less. See how usable a tool it is and how easy it would be to do in a self-service mode rather than relying on a vendor."