MarketingExperiments Blog: Research-driven optimization, testing, and marketing ideas

Email testing produces some of the most interesting results I see here at MarketingExperiments. The cause for this is a combination of constantly changing variables.

For one, content within email tends to change more often than your typical landing page. This makes optimizing for content more challenging as different topics are likely to garner different levels of interest from the segments within your email list. So results will change each month based on the content alone – making A/B testing the only reliable method for measuring progress.

In addition, email lists themselves prove to be a challenge, as what works for one list may not work for another list. Even within lists, especially aggregated lists, you will see different results based on the value proposition, content, layout, and calls-to-action (CTAs) used in your email.

And to further complicate matters, you are still dealing with a funnel process in which your email must first reach a user (avoiding spam filters, personal filters, etc.), your subject line must interest the user enough to open the email, your email must display properly (with images on and off) and be compelling enough to achieve a click-through to your landing page where the battle for a conversion wages on.

In today’s world of overloaded email boxes, people declaring email bankruptcy, spam filters and everything else, this game is only getting more difficult – for marketers and users alike.

With that said, I’d like to offer up my own favorite email testing tricks and tips. It’s important to understand that what works for one segment, list, or industry will not necessarily work for another. In fact what works one month for a list may not work next month. It’s an ever-evolving process in which you must always challenge your own best practices to maximize your results.

1. Text-only email

This is my silver bullet of email optimization. In a world where everyone wants their logo, business cards, websites, and emails to be as shiny and pretty as possible, it’s easy to forget that email is a text-based medium.

Out of all the emails you actually read each day, how many are HTML vs. text-only? The important emails you receive each day – the ones from your family, friends, and co-workers – are all likely text-based. This means the HTML emails you receive are most likely not from one these aforementioned groups and thus likely less important to you on a personal level.

Obviously there are caveats to this – such as an email from your bank or a Facebook friend request – but the truth is, when you send an HTML email you are already fighting a certain level of banner blindness. If you currently only send out an HTML email, I’d challenge you to A/B split test against a text-only version of your email and measure the results.

2. Story format

Emails are a form of value exchange. In exchange for someone’s time and interest, you must first provide something of value.

One of the easiest, most interesting ways to provide value in an email is to tell a relevant story. Not only can telling a story create interest in your topic, but it also can provide you with a natural sounding CTA of “Continue Reading” or “Read More” that requires less commitment from the user before clicking through to your landing page.

The downside to the story format is it requires you to have an interesting story to tell as well as an audience willing to read through the text.

3. Big button

Sometimes if your offer is compelling enough, all you need to do is give people a place to click. It helps if your email list is already familiar with your brand, is interested and familiar with what you’re offering and needs little explanation before clicking through.

For example, our Marketing Director is drawn to J. Crew’s buttons like a moth to a lightbulb…

The focus should be on the value proposition and the CTA copy as this will determine the commitment level required to achieve a click-through. A “Learn More” CTA will bring more clicks, but less qualified traffic than a “Buy Now” CTA. Experiment with this CTA copy to see what works best for your particular audience and dial in the right amount of click-through vs. conversion.

4. Multiple CTAs

I’ve found that when dealing with large, aggregated email lists, including multiple, different CTAs can help increase click-through on the basis that different people will be interested in different aspects of a topic.

For instance, if you are offering a free PDF download on a topic that you hope will encourage people to provide their email address, and you also have a webinar that does the same thing, go ahead and include a link to “View the PDF” and another CTA to “Sign-up for the Webinar.”

What you’ll find is that offering multiple contrasting calls to action will result in a higher click-through rate as some people will be interested in downloading the PDF, while others will prefer the webinar – two different mediums that attract two different groups of people but achieve the same objective.

5. Digest

If you currently send out a longer email newsletter, I’d encourage you to try a digest format. In the digest format you offer several links at the top of the email that either direct users to a particular article on your website or anchor down to an excerpt within the email itself, followed by a CTA to “read more.”

I’ve found digest emails work well when you have a lot of content to offer and you are able to effectively arrange this content so as to attract a wide variety of clicks. It’s a combination of the story format and multiple CTA emails mentioned previously.

Related Resources

The Five Best Ways to Optimize Email Response: How to craft effective email messages that drive customers to action

Optimize your Email in Three Steps: How one marketer tripled revenue from their house list

Order your custom Email Response Optimization Package

Email Marketing, Research Topics

Earlier this week, Google formally announced the completion of its new web indexing system cleverly named Caffeine. According to Google, Caffeine provides 50% fresher results for web searches than its last index and is the largest collection of web content the search giant has ever offered.

Our old index had several layers, some of which were refreshed at a faster rate than others; the main layer would update every couple of weeks. To refresh a layer of the old index, we would analyze the entire web, which meant there was a significant delay between when we found a page and made it available to you.

With Caffeine, we analyze the web in small portions and update our search index on a continuous basis, globally. As we find new pages, or new information on existing pages, we can add these straight to the index. That means you can find fresher information than ever before—no matter when or where it was published.

– Carrie Grimes, Software Engineer, Google

This is great for those of us who use Google to search and find relevant results to our most common inquiries. Results will become timelier, more social and rely more heavily on keyword strings, ultimately providing more useful results as newer content can be indexed much quicker and from a much larger base of sites.

Is your SERP spot threatened?

When Google says “fresher” results, what they’re saying is that ranking principles have not changed, but rather rankings are (and will become) more dynamic, shifting to display the latest and greatest (and, therefore, hopefully best) information as a result of being able to reach deeper and more frequently into the Web.

But what about website owners who have come to rely on the steady ebb and flow of organic traffic that a high Search Engine Results Page (SERP) position provides?

Many website owners who have long enjoyed a top spot, or even a high spot, have suddenly found their sites displaced, resulting in a massive dip in organic traffic. And to make matters even more vexing, the position your site is in this month will likely be different from where you find yourself next month.

Which is not entirely new, right? Anyone well-versed in search engine optimization (SEO) knows that it is a never-ending battle. The difference is, ranking improvements and demotions may happen even quicker than before because content that you and your competitors are creating will have a more immediate impact within the results. So if you thought SEO was a wild ride before, hang on.

Of course, your main goal should be to deliver value to your customers and audience. After all, ranking is only a means to an end. And since Caffeine should do a better job of measuring that value, it might start putting some distance between those who do provide quality content and those who are merely gaming the system.

Caffeine makes it more difficult, although not impossible, for sites using black hat SEO tactics to reach and/or maintain a position at the top of the rankings for long periods of time. And while I believe SEOs will always find new ways to game the system, I think Google has made a step forward in terms of providing better-quality results.

How to get a boost from Caffeine

So if you can’t rule Google SERPs by just throwing up an automated page with repurposed content, what should you do? Here’s my advice to website owners who rely heavily on organic traffic:

• Continually look for opportunities to expand or update the content on your site for improved keyword targeting
• Re-evaluate your current keywords and always look for opportunities to expand and capture more long-tail keywords
• Build a site that contains clean code and  a clear site structure
• Look for opportunities to capitalize on social media as real-time results become more integrated with search results
• Monitoring your competitors will be paramount as new content brought in by them will be indexed quicker than ever.

So fresh and so clean

I think this transition to providing “fresher” results was inevitable as competition from Bing and the massive growth of “real-time” information from social networking sites like Twitter and Facebook have created a need for better, faster, (stronger), search engine results. And from a conversion standpoint, I’d argue that this move could ultimately prove to be more beneficial to ecommerce sites that provide high-quality content, because “fresher,” more relevant results also means more qualified traffic.

Of course, cleaner, less manipulated results will have huge benefits to searchers and real, quality sites alike. Remember that both you and Google are on the same mission: provide the right page to the right user. Oh, and don’t be evil.

Related Resources

Search Marketing: Tips on mastering the latest innovations in this mature category

PPC Innovation: How will Google’s new lead capture extension affect your pay-per-click campaigns?

Optimizing PPC campaigns to boost conversions, ROI

Photo attribution: The Official Google Blog

Research Topics, Search Engine Optimization (SEO)

Search for “social media marketing” on Google (or Bing) and you’ll likely end up in a black hole of Twitter guides, bit.ly retweets, social media “mavens” and Top 7 Tips for Creating a Facebook Fanpage blog posts. And it’s no wonder everyone sees the need to discuss and “explain” how social media works for the 270 billionth time. With so much discussion on the topic, you need to be a social media expert just to navigate it all.

Yet, one social media topic manages to slip through the cracks. And it’s often the first obstacle companies encounter when they decide “social media” is the answer to all of their problems:

Who’s going to carry out all of these social media initiatives?

It’s the Human Factor – who is going to create the content for that blog? Who is going to reply to all those tweets? Who is going to make sure your Facebook page doesn’t turn into a hate fest? In fact, who decides if you should have a Facebook page at all?

If you’re like many companies, you might think outsourcing social marketing is your best bet.

But more than ever companies are working to keep social media in-house because it requires such an intimate knowledge of the brand and because of the personal nature of social media interactions in general. Customers want to talk to you, not an “outsourced” spammy twitter account. And when you leave them with no choice, they happily take their discussion to your nearest competitor.

Based on the MarketingSherpa Social Media Marketing Benchmark Survey of more than 2,300 respondents in November of 2009, social marketing responsibilities are outsourced less often than traditional marketing responsibilities– meaning you’re more than likely going to need to look internally to find your resources for a social marketing team.

And here’s where I believe many companies get it wrong. Instead of hiring or tasking the best person for the job, whether that’s managing the Twitter account or actively engaging in forum discussions, many companies put their least experienced, least qualified people on an overwhelming number of social media initiatives. Usually this person is in marketing and may be tasked to cover topics or areas of social media they have little or no experience in. But this isn’t the most efficient and certainly not the most effective method for achieving social media success.

But how do you go about identifying who in your organization is best suited to carry out social media objectives? The answer is surprisingly simple. You pick the right person for the job.

For example, don’t send your marketing team to engage users in a developer forum. They’ll stand out like a script kiddie at a Def Con Conference. Instead, encourage your developers to actively engage users in forums related to their industry. Task these same developers to contribute technical content to the company blog. It’s surprising the level of involvement you’ll get from your team when you place a little responsibility in their hands.

Customer service can provide you with your Twitter recruits. In many ways they are trained for the role, just in a different medium. And your Sales department is a great place to find outgoing personalities to run the company Facebook page or handle group discussions on LinkedIn.

The point is that your social media team should be composed of individuals from various departments who can each provide a certain level of expertise by contributing just a few hours a week to social marketing initiatives each week.

To learn more about how you can keep your employees accountable for social media initiatives, check out the new MarketingSherpa Social Marketing Handbook.

Related Resources

Social Media for the COO: How to become the Michael Phelps of implementing social media in your organization

The MarketingExperiments Quarterly Research Journal

MarketingSherpa Social Marketing Training

Photo Credit: Intersection Consulting

Research Topics, Social Media

As you may recall from my last post, I spent nearly six days in Austin, TX attending SXSW Interactive, a conference which Advertising Age once famously referred to as the bellwether for what lies ahead for digital culture.

The event was a fantastic gathering of ideas, technology and core conversations – the kind of conversations that tend to happen when the Internet takes a moment to meet in person and exchange fancy business cards.

Test everything?

And speaking of fancy business cards, for this event MarketingExperiments ran an A/B split test version of our business cards. For those keeping count, Side B won, although the results are far from being statistically significant.

(click below to see the two treatments)

But what the cards were able to prove is that the future of online testing is brighter than ever! While panels focused on a wide-range of topics from content development and management to social media and design standards, I spoke to a number of Web 2.0 designers, business owners and service providers all fascinated with the prospect of testing their pages, registration paths and email.

Test more?

Even more exciting is that many of these people were already testing on their own sites. One such tester, FanBridge co-founder Noah Dinkin, was especially interested in discussing testing strategies as he has attended several of our webinars here at MarketingExperiments.

FanBridge is a site that allows someone (such as a musician) to manage a list of their fans. In this way, FanBridge must focus not only on attracting new signups to its own service, but to help their current subscribers effectively communicate to their own fans. Noah, who said he tests using Google Website Optimizer, confided in me what many of us already know. He is testing some, but he needs to be testing more.

Test the right thing

This was a common theme in many of my discussions as I met people who were interested in testing and knew they should be testing, but were often unsure where to start. And I think, like any problem, the answer starts with research.

Resources such as this blog or our webinars are a great resource for anyone looking to see what testing is all about and see some real-world case studies.

I think we can all agree that the benefits of testing are such that we cannot continue to implement untested best practices or design standards without measuring the impact of our changes to the bottom line. The question becomes how many conversions can you afford to give away?

Because whether you run a Web 2.0 service application that receives five visits per day or manage an ecommerce site that receives 50,000 visits, your landing pages, cart processes and emails will determine whether or not you end up with a customer or a bounce.

Thanks to everyone I met at SXSW, and good luck testing!

Analytics & Testing

Every year, for the past 16 years, something amazing has happened in Austin, TX around this time. And this year hopes to be no different as Austin gears up for its annual South by Southwest Interactive technology conference – an event which can only be described as one of the largest, most exciting, most comprehensive collections of marketing and social web entertainment and technology this side of the Internet.

In what amounts to five days of epic conversations, presentations and social media events disguised as a “technology conference,” the world (and all of Twitter) tunes in to see the Internet trends of the last year scrutinized and the trends for the next year laid out in incredible, Apple-Keynote-quality detail.

How far can testing go?

This year, MarketingExperiments will not only be tuning in, but will take part in this conversation to discover not just what works in email and landing page optimization, but what is working in social media and other areas of the Web as well. In addition to small and large business owners, how are bloggers, designers and the social media crowd testing, measuring and collecting results? What’s working, what isn’t, and what’s to come in 2010?

If there’s ever been an ongoing topic of conversation here at MarketingExperiments, it’s the ever-present question of how far we can go in testing. How do we improve and build upon the tools we have to increase ROI, discover new insights and push the boundaries of online testing to give us more accurate, more actionable results? How will social media continue to change the way we hold conversations and gain trust with our audience online? How will we measure and apply what we learn?

See you in Austin

What better way to seek out the answers to these questions than by employing SXSWi as my backdrop for a discussion on how the Internet will continue to evolve to the tune of testing and optimization, as well as hear first-hand how other marketers, bloggers, and Internet fans in general see online testing influencing the design and execution of ideas on the Web.

And if you happen to see me there, be sure to stop me and tell me your thoughts on these issues as well.  Also, ask me for one of my “split test” business cards. More on that later…

Internet Marketing News