For 25 years, Nick has been working to optimize web content for over 100 major corporations, including such names as Citibank, Apple, Chrysler, Looksmart, MSN, Yahoo!, the US Navy and America Online. Nick has been honored with fifteen major awards for his marketing work, and is internationally recognized as a leading expert on the subject of writing for the Web. He is also the author of the critically acclaimed book, “Net Words: Creating High-Impact Online Copy.”
Andrey: Thank you for taking the time to answer a few of our questions, Nick. Please tell us a little about your background and how you got involved with search engine marketing.
Nick: My own background is in direct marketing as a copywriter. That’s what I was doing from 1980 to 1997. From the first day of January, 1997 I made the decision to work exclusively in the online marketing space.
From the beginning it was clear that being a copywriter for the web would come with a whole new set of demands. Some of the skills I already had were transferable – others were not. In addition, there was a whole new learning curve to climb, as the Web was, and is, so fundamentally different from other media. One thing I had to start learning early on was the impact of search engine marketing. It wasn’t called that back then, but the challenge of getting listed high up in search results was present from the beginning.
From 1997 to the tech crash of 2001 was a fascinating time, and a time when every day brought some new learning opportunity. Soaking up everything I could about this new medium culminated in my writing my book, Net Words, which was published in early 2002.
After several years of working freelance as a writer and consultant for dozens of companies online, in mid-2005 I started working full time for MarketingExperiments.com. Why? Because it is the only place I know where I can ask questions about what works online, and what doesn’t … and then be able to run tests and find out the truth.
Andrey: MarketingExperiments.com’s guiding idea of testing to find out what works seems like a great one. So what exactly goes into “discovering what really works”, and how do you go about conducting your experiments?
Nick: Let me start with some questions.
On a given landing page, will long copy sell more products or services than short copy? If you have a three-column page design, will it convert better or worse than a two-column design? If you have a beige background on a landing page, will it convert better or worse than a page with a white background? If your subscription page asks readers to complete ten fields of information, by how much will conversions improve if you reduce the number of fields to three?
These are just a few examples of the dozens of questions we ask, and then test … for web pages, emails and e-newsletters.
Does it matter? It certainly seems to. With some of our research partners we have conducted multiple tests on increasingly optimized pages and improved conversion rates in some cases by over 1,000%.
How do we do it? First, we analyze the existing page we want to optimize. Then, based on our current body of knowledge, we create optimized versions of that page; but we never assume we have made the best possible changes to the page. We always test.
The testing may take the form of a simple sequential test, an A/B split test or a multivariable test. It all depends on the data we are looking for and the stage of the testing program.
We also pay a lot of attention to establishing the validity of each test result. All too often companies make strategic or tactical decisions based in invalid test results.
But that’s what we do. We never assume we know what works best. We always test. That’s how we set about “discovering what really works”.
Andrey: In keeping with this issue’s theme, which focuses on website and ad copy conversions, what experiments have you conducted on the topic and what were some of your findings?
Nick: We have conducted hundreds of tests on this topic. We are continuously testing the performance of both PPC ads and web pages.
With regard to PPC ads, we recently tested four different ad headlines, with the remainder of the ad remaining the same. The worst performer delivered a 2.86% click-through. The best of the four gave us 6.74%. Consider that difference and calculate what it means in terms of revenue. If we hadn’t tested, we would be leaving a lot of money on the table.
Very often, we test the combination of a PPC ad and its associated landing page. When you are using PPC ads, you get the best results by matching the ad to its landing page and optimizing both together. Quite often, for example, testing the headline in a PPC ad will give you valuable insights regarding the headline on the landing page itself.
As for optimizing web pages and sequences of pages, we have found that a program of ongoing tests and improvements can yield very impressive results.
For an online publisher of B2B newsletters we optimized their website subscription pathway and increased the number of subscribers they were getting by 816%.
We also worked on a landing page and subscription pathway for a major financial publisher and increased their subscription rate by 607%.
We did a lot of work on optimizing PPC ads and landing pages for a site focused on child safety, and achieved a 1,156% increase in sales.
Of course, some of the results we achieve are more modest. And, as with all testing, sometimes we’ll try a new page version and find that conversion rates have actually fallen. In those circumstances, we dig deeper and keep working on the page until we can see a significant improvement in its performance.
Andrey: What would be some of the most common mistakes people make when it comes to website usability, ad copy and landing page conversions?
Nick: The most common mistakes?
1. Lack of clarity as to the objective of a web page. If the page doesn’t have a clear and simple objective, how can you expect it to succeed?
2. Lack of clear Unique Value Proposition in the headline. If people aren’t captivated by the value you communicate in the headline, how can you expect them to read the rest of the page?
3. Lack of clear sales path. Too many web pages, or sequence of pages, lack a clearly structured pathway for the reader to follow. Whatever the objective of a page, the text needs to drive the reader forward in a simple, compelling manner.
4. Poor page design. A single column vertical sales and copy path will, in our experience, always outperform any other kind of page layout. Web pages with too many columns, graphics and other elements split the readers’ attention and dilute the power of the sales path.
5. Poor copywriting. This is endemic online. Too much attention is paid to making pages look good (and the best “looking” pages often convert the worst), and too little attention is paid to the quality of the content and the copywriting.
6. Asking for too much personal information too soon. Why decimate your conversion rates by presenting people with a three-page, 50-field application, registration or sales sequence? Never ask for more information than you absolutely need, particularly if you are addressing a first-time visitor to your site.
Andrey: Are there any specific industries that you have had more luck with than with others? In other words, are any of your findings industry-specific or could they be applied to every website out there?
Nick: We work with partners from all kind of industries. We work with some of the largest publishers in the U.S. We work with large and small e-retailers. We work with medium-sized newsletter publishers. We work with companies selling products, services and subscriptions.
In other words, we don’t focus on any one type or size of company or industry.
Looking back at all the tests we have conducted, while some may have been very specific to a particular company or circumstance, what we learn from each test can generally be applied to any kind of company of any size.
So, yes, our findings can usually be applied to pretty much any website.
That said, we never assume that what we have learned with one test can simply be applied elsewhere. We use our learning as a starting point, and then design tests to either confirm or refute our assumption.
Andrey: What test tactics could a company implement if it has a small budget and/or limited amount of time?
Nick: If you look back to my answers regarding the most common mistakes people make, getting it right doesn’t come with a big price tag. It really needn’t cost a lot of money to create pages that convert well.
As for testing, even a small company can outsource that at a very reasonable cost. If you have internal IT people who can set up simple sequential or A/B split tests for you, so much the better. Multivariable tests are a lot more complicated, and most companies will need outside help with those.
Optimizing websites, emails and e-newsletters is not a big-budget undertaking. The qualities you need most are clarity of thinking and brutality. Be clear about the objective of every page and every element on each page. Be brutal about removing everything that does not directly support that objective.
Andrey: Considering the fact that MarketingExperiments.com has processed so much data, there is no doubt you have a comprehensive picture of what the search marketing industry looks like and what can be expected tomorrow. What are your thoughts on the new trends in the industry and where it is headed?
Nick: Search marketing continues to grow in two directions. You can pay for search traffic through PPC advertising, or you can optimize your pages to achieve higher organic listings in search results.
There is plenty of future growth in both areas.
The challenge is not really in where the search industry is heading. The challenge lies in pushing online marketers up a very steep learning curve.
In the case of PPC advertising on search page results, too many marketers believe they know enough on this topic to do a good job. They don’t. Most know enough to do an acceptable job. This is becoming a very competitive area and will become more so. “Acceptable” won’t cut it for very long.
Marketers responsible for PPC work need to quickly increase their knowledge and expertise in this area. As the competition heats up, only those who do an excellent job will be able to deliver a strong ROI.
As for organic search, the vast majority of companies do a very poor job in this area. The larger the company, the harder it seems to be for them to build pages and sites that address the priorities of the major search engines.
There are a vast and growing number of “trash-content” sites on the web. For online marketers to lift legitimate sites above all the garbage, they need to invest a lot more effort and resources in search engine optimization.
Andrey: I know you publish your findings in a newsletter. Tell us a little bit about its format and how one can sign up for it.
Nick: We do this in two formats. Every two weeks we conduct a live teleconference call during which we address a particular topic and share our most recent test results. For instance, our last call was on the topic of Landing Page Optimization.
A few days later we then publish the audio recording of the call, and a written version in the form of a research brief. The written brief is sent out as a newsletter and also added to the archive area of our site.
Our calls and accompanying newsletters are not about our opinions; they all contain new and valid test data on a particular topic.
There is no charge to attend the calls or to receive and access the briefs. It’s all free. All you need to do is get yourself on our subscription list. To do that, sign up on our home page at MarketingExperiments.com. All we ask for is your name and email address; and, as I mentioned, there is no charge.