Ask an Expert: Email's Vital Signs
This week’s Ask-an-Expert question is: What are email's blood-pressure-and-cholesterol metrics, ie. What are the first two or three a consultant will look at to determine the health of a client's program and the first they will move to help the client fix?
Once again, we turn to the experts on the Only Influencers discussion list, an invitation-only list of online marketing experts:
Gretchen Scheiman, partner, director, CRM, OgilvyOne Worldwide:
The first thing I ask is "what do you measure?" There are some truly oddball metrics out there, and they give a tremendous amount of insight into how the client thinks about their program.
Next I ask for a trendline of those metrics. A trendline of sent, delivered, open rate, click rate, and click-to-open rates can generally alert me to problems with deliverability, segmentation, list quality, acquisition practices and content quality.
From there, I can help the client understand why I need extra data—along the lines of some of the suggestions already mentioned. This helps when clients are often already overwhelmed with little time to spare for requests for data which they don't necessarily understand. Pointing to a trendline and saying: "I suspect the problem is X, let's look at the data to prove/disprove that theory" helps create a partnership and a willingness to spend the effort to push overworked analytics teams to cough up some data.
Jaren Angerbauer, deliverability consultant:
As far as looking at email's "blood-pressure-and-cholesterol metrics" or IMO "vital signs," there is a whole technical side that needs to be included.
1. Where does the subscriber data come from and what is the quality of the data?
2. What kind of email infrastructure is being used, and does it have all of the necessary pieces that will translate to good delivery?
3. Are *proactive* practices being done to help show transparency and a good overall reputation as an email sender?
Jessica Best, marketing manager, Emfluence:
1. Bounce rate by domain: If you're bouncing 25 percent at Yahoo!, something's up at Yahoo! If you're seeing an ongoing 8 percent bounce rate, your list is in trouble in general. Let's talk about why/how.
2. Open rate by domain: If your overall open rate is 20 percent and you're seeing 6 percent at Hotmail.com, you may want to check your seed lists to see if you think you're getting through to the inbox.
3. Unsubscribe/Complaint rate: This is cheating because I'm saying 2 different ones, but this relates to list quality. I bet you I can tell how you acquire your list (or educated guess) based on what your unsubscribe rate and complaint rate were for your last 3 large send emails.
The bonus on these is, no matter who the client is or who they send through, you know you've got this data. It's core to email marketing.
Andy Thorpe, deliverability and compliance manager, Pure 360:
The first metrics I look out for are the negative ones. Opt-outs, bounces, complaints + FBLs [feedback loops] compared to the opens.
In a similar way that that successful football manager (soccer) would make sure they don't concede lots of goals by ensuring they have a good back-four & keeper before they start investing in strikers.
The people who are already engaging and spending can be optimised later, but if the negative affects on reputation are not attended to early, the people who really want the email may not get it because the complaints and bounces send all emails to junk or worse.
"Building a good foundation of reputation maintenance is vital to be able to achieve consistent low maintenance deliverability.
This will provide the opportunity for quality: to produce creative & remarkable engagement & revenue opportunities. That is what I see marketers getting excited about so that is I want them to be able to do, without the ambiguous fear of the junk folder."
Karen Talavera, principal, consultant, Synchronicity Marketing:
I have a 5-step discovery process I use to determine the health of a client's program:
1. First, get the big picture. It's vitally important to obtain an initial understanding of the big picture for email by asking: "What is your goal for email and how does email marketing fit into your business? How do you use email and why?"
I'll want to know where in the company and how they use it. Often I'll be working with just one group or division of a multi-pronged email operation, so it's important to know the macro—and micro—context for email.
2) Step two, understand the source of the list(s). By that I mean I want to know how they've built their list(s), sources, permission practices by source. This goes for in-house and rented lists both. This is commonly overlooked but if you know it going in can explain all sorts of deliverability, reputation and engagement problems (if any) later on. There can be hidden gem sources of names that are yielding high response as well as problem children, so it's important to turn over all the rocks.
3) Step 3: Historical campaign process metrics and benchmarks. These are the first initial diagnostics that yield an understanding of how email is working for the client, so I'll be asking for trend line information on opens, clicks, conversions, deliverability, unsubs, forwards, shares to social - whatever is a relevant process measure of the objective of the campaign. As someone else pointed out, if the objective of a company's email is mostly to drive traffic for content consumption, then a click is a conversion - it's a very different scenario than for e-commerce email. There are also branding objectives, research, etc. - plenty of other things that don't require a purchase, so it's important to match relevant metrics to objectives.
4) That said, if email as a channel isn't adding economic value (producing revenue or saving money) what's the point? So finally, I want the "rubber-meets-the-road" contribution measures like average order value, cost to acquire a new name, ROI, and even LCV (lifetime email customer value), etc. I'll likely compare these between campaign if the client hasn't benchmarked over time already.
5) Last but far from least, deliverability, bounce and complaint metrics. Although I'm not a deliverability expert I need to know if there are serious issues. If so, I bring in or recommend a deliverability/auditing company or expert.
DJ Waldow, director of community, Blue Sky Factory:
A strong indicator of a healthy email marketing program is high inbox placement. If your emails are landing in the inbox on a consistent basis, it's an indication of a few things - all which are interrelated: a clean list, engaged subscribers, & a strong IP/domain reputation.
A clean list (few bad/unknown email addresses, low spam complaint rate, etc) helps with your IP/domain reputation. This in turn contributes to a higher inbox placement. Engaged subscribers (those who open, click, convert, share, etc) also help improve IP/domain reputation. This, again, improves the likelihood that emails will land in the inbox.
Overall, email marketing success is about sending the right email to the right subscriber at the right time. Timely, targeted, relevant, and valuable emails to those who have opted in.
Bill Kaplan, CEO, FreshAddress:
The three email metrics we look at are:
1) Deliverability (i.e. heart rate) stats: What percent of a client’s emails are being delivered and what are the average open and click-through rates? For a proper analysis of deliverability rates, one should look at a client’s entire email program across all of the ISPs, analyzing full deliverability numbers for each ISP. For inbox delivery, seed tracking can provide a very rough idea of deliverability into the inbox but understand that the number of seeds used by providers is probably not sufficient to be statistically significant within plus or minus 34.1 percent, a one standard deviation event. If any of these stats are below par, find a specialist who will help you clean your list ASAP as the vast majority of email deliverability problems are caused by hygiene problems with the underlying list (e.g. excessive bounces, excessive spam complaints, existence of spam traps).
2) Is the client experiencing any blocking or blacklisting (i.e. faintness of breath, headaches, dizziness, high cholesterol) issues? The sooner you can catch these, the sooner you can resolve them and put your client back on the road to recovery. Again, email hygiene can help solve blocking/blacklisting issues, identify your problem lead sources and channels, and prevent future problems down the road.
3) What is the revenue generated (e.g. your overall health) per email sent? This is, of course, an easier question to ask than to answer, given the difficulty of attributing sales properly across all of one’s marketing channels. But if you can’t answer this question and don’t know the value of an email address to you, you need to put your marketing and analytic teams to work. If you’ve passed metrics 1) and 2) above, then the tough work of marketing is needed to really get you in shape (e.g. customer segmentation, one-to-one marketing, drip marketing, relevant content, appealing offers, calls to action, coupon tracking, etc.). Remember that email marketing is a marathon, not a race. There are no quick fixes but the companies with the proper preparation who can maintain a consistent pace will find themselves crossing the finish line at the top of the pack year in and year out.
Given the tremendous revenues being generated by email every day—anyone heard of Groupon?—maintaining and optimizing the health of your email program should be a top priority for every company.
Loren McDonald, vice president, industry relations, Silverpop:
When looking at a client's program, I'm going to want to look at not just past performance, but what is the potential for improvement? Depending upon the industry and objective for email at the company, I'd look at different aspects, so let's take an e-retailer as an example:
1) What is the percentage of "inactives" (however defined) in the email database? All other metrics are pretty meaningless unless you have some baseline of how engaged your list is. If it is greater than, say, 50 percent the company has serious challenges ahead. I like to then see what the usual response metrics are by the remaining "actives.” This then provides a decent sense of whether there is any heart beat in the program.
2) What percentage of their email revenue is attributed to trigger-based emails? If it is small, say less than 5 percent or 10 percent (versus say greater than 30 percent) then I know the company hasn't moved their program very far along the sophistication path. This suggests what potential opportunities and obstacles might exist to take the program to the next level. Do they have the team, resources, desire and management support to move their program along the marketing automation path? Or are they content to just pound away on their list?
3) The first area then to focus on is the quality of the list. The easier things to fix are creative/design, offers, cadence, adding triggers, better segmentation, etc. But it will bear little fruit if the core email list is made up of people who opted in just for the chance to win an iPad and then never engaged.
Alexandro Romeo, email marketing manager, Houseparty.com:
Delivery is always your first priority. You have to be sure you can get the most email in front of your consumers as possible and constantly monitor/nurture that ability. Obviously, this is completely based on source of acquisition, communication relevancy and appropriate frequency. This has to be as close to 100 percent as possible. If you are not keeping your delivery as high as possible, you are playing ball handicapped.
Complaints speak to the "health" of your program. Depending on your volumes, you cannot exceed specific thresholds that are preset by ISPs. These absolutely need to be as low as possible at all times. Trending is just as important because it provides immediate insight into volatility. Volatility is essentially your "suck" meter from a content standpoint.
Unsubscribes let you know you are failing as a marketer. If you are setting an expectation and not meeting it, people will leave, plain and simple. Unsubscribes directly correlate to value. A "one stop" mail stream will always increase your overall unsubscribe rate. Providing value, being relevant and giving your consumers flexibility with frequency, will minimize your losses.
Rory Carlyle, director of marketing, BombBomb:
1. Email Address Acquisition Practices - "Where are your contacts from? How often are they acquired?"
This factor addresses myriad issues: Deliverability, contact activity, list fatigue, list rot, roi, contact value, etc. This single question can highlight how seriously a client considers their email marketing in the grand scheme of their marketing plans. It shows their current investment (not just monetary) into the channel as well as what type of dedication they have towards making it successful.
A client who actively markets to consumers to garner interest and email signups will do far greater with all aspects of email marketing that someone who solely purchases lists. They've also typically taken more steps towards architecting an email marketing system instead of a batch-and-blast solution via their ESP.
2. Cadence vs. Engagement - "How many times are you reaching out to them monthly/weekly/daily? Are there large fluctuations in your lists engagement per campaign?"
A lot of email marketing programs suffer from "broadcasting" or lack of thought within their email channel; broadcasting tires and disconnects recipients due to the lack of focus or the overabundance of mailings. My tendency is to chart cadency against engagement metrics and then optimize the content/design against the prime cadence. This relieves issues with consistency and allows the marketer to plan content creation in parallel to the email schedule - optimizing one thing at a time to dial in the appropriate design, voice and overall approach to their lists.
One of the biggest issues in commercial email today is "relevancy". Which is just a cover-up word for, "stop regurgitating the same info to your customers and prospects". Sending the same information via email once a week for six weeks doesn't make it any more effective, get new info.
3. Data/List Segmentation Practices - "Over segmented? Under segmented? Are you cross-contaminating your segmentation?"
This can literally change the resources and overhead needed to run the email marketing for the company. It also can pin point segmentation as an attrition factor rather than a positive attribute. A small company doesn't need 19 lists, a large company might not need 19 lists, it's all relative to the success and growth of the channel. Managing the creation, testing, deployment and measurement of 19 campaigns against 19 lists on a weekly/monthly basis is just unheard of for most interactive marketing departments. It can be done, sure, but is it done with effectiveness? Under-segmentation is just as bad. Cross-contamination is when the segmentation minions take over the list creation and you get a single contact across multiple lists within one campaign or multiple campaigns within a short timeframe. This results in multiples in the inbox or a barrage of offers with no personalization.
Dela Quist, CEO, Alchemy Works:
You make some very good points, but I REALLY hate it when people equate the word broadcast with lack of thought, stupidity or antediluvian behaviour. Please answer this what bit of Madmen, I love Lucy, the Olympics, Super Bowl, The Masters, The Simpsons, Friends, The Office, CBS News, Star Trek, Happy Days, Glee, American Idol Harry Potter, Star Wars - I could go on for ever - are down to lack of thought or the fact that they tire and disconnect viewers due to a lack of focus or overabundance?
Sure it's harder today than it was 50 years ago to build a huge following, but it is possible and a large number of people who want the same thing beats a large number of people who want completely different things, EVERY time. So it is better to make more people want more of one thing than to make more things in the hope of being relevant.
It is called (email) marketing
Hey Dela, good point but I'm assuming Rory was referring to "broadcasting" to mean email campaigns lacking segmentation and targeting when possible.
I'm a firm believer that some emails should be broadcasts and reach the entire customer list/universe (newsletters, big announcements). More and more, however, there are opportunities for segmentation and customization of email that are ignored or wasted. While a campaign (annual sale, new event) might be applicable to the majority of a marketer's list or customer base, how it is packaged creatively can and should differ by segment. Plus, there's trigger-based email which gets even more micro-targeted; dialog tracks, short-term series, etc. which can and should be used more often.
I say there is still too much "one size fits all email" and repetitive messaging going out the door when better options are available. Broadcast has its place, yes, but too often is the lazy-marketer's solution to getting email done.
Now, just as long as no one is using the word "blast" we're okay . . .
Carlyle: I attempted to use the word broadcast to show the overall approach to the channel, not correlate it to television. Too often marketers use email as a one way street with no real interaction expected from the recipient. Too often I see marketers sending emails at random times, with hodge-podge content creation (no real theme), inconsistent ad or CTA [call-to-action] placement, etc. Broadcasting in my mind equates to a broad sweeping approach that may work for TV shows, but isn't really tailored for email. Email has the capability to do more and has shown to be generally more effective when used with a bit more specificity towards the recipients.
I agree with you and Karen here. There are many companies using a broadcast-type approach that work well, but the clincher in those approaches is consistency within the campaign. While they email everyone the same things - the approach is constant and the expectations are typically set for the recipients. My comment was more focused to the companies that have an inconsistent schedule, content matter, focus and goal for each newsletter/email and still send these mailings to the entire list.
Karen had a good summation: " I say there is still too much "one size fits all email" and repetitive messaging going out the door when better options are available. Broadcast has its place, yes, but too often is the lazy-marketer's solution to getting email done."
I don't know about you all, but I see this practice a lot.