Practical Tips To Lift Your Conversion Rate
Hey, guys! I didn’t get nearly enough liveblogging last week, so I thought I’d sit in on this morning’s (or afternoon, depending on where you live) MarketingSherpa Landing Page Handbook seminar. Something about sitting in on the Multivariate Testing and Conversion Tweaking session last week has me dreaming about landing pages. And chocolate. However, the chocolate dreaming is a pretty natural occurrence for me. I don’t think it had anything to do with last week’s session.
Okay, enough of my rambling, we’re starting.
With us today are Marketo VP of Marketing John Miller, MarketingSherpa Research Director Stefan Tornquist and MarketingSherpa Senior Analyst Tim McAtee. We’re going to be covering findings from MarketingSherpa’s recently updated Landing Page Handbook.
We’re told that the data included in the handbook was gathered from surveys of 4,213 marketers, 3500+ consumers, lab tests and partners research, ‘best of’ research from 650+ third party organizations and 800+ Sherpa case studies.
What kinds of landing pages work best? Below is a screenshot of some of the top tests:
Tim and Stefan go through examples of the different types of landing page tests you can do.
- Dynamic Search Copy: We’re shown a B2C example for a site selling music equipment. They adapted their search page to reiterate what users typed into the search box ("you searched for: Stratocaster guitar on Google") and increased conversions 48 percent. It’s so simple but it orients the user back to their search. It tells them, "hey, this is relevant content for the search you conducted".
- Registration From Tests: Your marketing doesn’t end when someone gets to the form. So many organizations do a great job getting customers into the conversion funnel, but then they abandon them with a bad form. There is no golden rule for how many questions to put on a form. Sometimes it makes sense to have a lengthy form, for other business it doesn’t work well. The more complex the question, the greater degree of invalidity.
When you ask people about budget or the number of employees in their organization, over 50 percent of people said they don’t give accurate answers. Most of the time it’s because they don’t know the information yet, not because they want to lie to you. If you require those questions you’ll get bad data and get more people abandoning the process.
Stefans shares that 22 percent of those surveyed said they still have a "reset" button on their form sitting next to the "submit" button. There is no good reason to have this button! People will hit it accidentally, get frustrated, and then leave. You find this legacy programming a lot on B2B sites.
- Creative Elements: As a creative rule of thumb, make eyeflow easier. Anything you can do to simplify and clarify a landing page will produce better results, unless you’re serving Asian countries.
- Eyeflow and Columns: Most people use 1-2 columns on email ads and search landing pages. You want as few columns as possible. The more columns you have, the harder it is for the human eye to consume your content. The panelists noted that landing pages coming from an email ad are typically simpler and easier to read than landing pages found via search. Often times what happens is that people who come from email ads are going to a very specific page that was designed to be a landing page. That’s not always the case with search. It’s not as coherent. What makes a landing page effective is its focus around a single topic.
- Eyeflow and Typeface: Conversions require more than hip graphics. Your designer may think white text on a black background works well, but your customers will be rubbing their eyes. Verdana 10pt is the most popular font/size combination on the Web even though it’s too small for most people over 40 to read. Make sure your text is black and hyperlinks are blue, otherwise people will have a hard time reading your copy.
- Eyeflow and Buttons: Don’t say "submit" say "add to shopping basket". Don’t say "click here" say "click here to buy X". The bigger the item is, the most likely it is to get clicked and to get eyeballs. If someone stands across the room and can’t read the button, it’s too small. You want it to be obvious what you want users to do.
If you can do nothing else, budget for analytics and testing:
- 48 percent of those surveyed say they don’t do any A/B testing.
- 40 percent only test at launch and leave forever.
- 16 percent don’t share test results with agency.
And the scariest data point of them all: 18 percent of those surveyed said that no one even knows their landing page results. Yikes.
Tip: Measure landing pages based on final KPIs, not interim clicks. They mention the SEOmoz landing page contest and how they went about finding their new Premium Service landing page. Found that the long, scroll forever page got less clicks but higher conversions. You have to look at the second stage metrics.
From here, Stefan and Tim take a look at some user-submitted landing pages and basically pick them apart. I’ll share the important stuff that may be applicable your site.
- Test your landing pages, but don’t over test. Having too many test versions means it will take a long time to get anything even remotely looking like accurate results.
- Don’t pass a lead to sales before it’s ready. Manage your leads:
- Score leads so you know who’s ready for sales
- Nurture leads that aren’t yet ready.
- Hand leads to sales at the right time, with value-added information.
- Modify programs as new requirements are found.
- Don’t use a lot of extraneous navigation. You don’t want to distract people from the primary action.
- Place important content, including your action items, above the fold.
- Include multiple points of contact.
- Lay off the long URLs.