Landing page targeting

It’s an emotive debate this long vs short landing page one. I have read a lot of tirades against annoying sale
pages that scroll and scroll forever.

However, I have seen enough of these
long form pages (Here’s Econsultancy’s landing page) to know that people are using them
for a reason. It can’t be coincidence. 

And some of the companies using long
form are respected brands (e.g. Amazon) with digital pedigree, so why would
they contravene the basic tenets of usability and user experience?

This blog looks at the approaches and tools you can use to optimise your landing pages and take the emotion out of design and decision making.

There is no right answer

Let’s start with me putting my balls
on the block and stating my professional opinion. There isn’t a definitive
answer. The optimum design and length of a landing page is dependent on several
factors:

1. The audience you are trying to reach.

2. The products/services you are trying
to promote.

3. The goals for that page.

4. Your brand and its
reputation/heritage.

OK, that’s not an exhaustive list but
it covers the key influences. You should never start a landing page project
with a fixed idea about what to build. Use your experience, intuition and
opinion but be open to change because it should be an iterative process.

Why do long landing pages sometimes
work?

 There are many reasons but the most
logical are:

1. You need to tackle potential
objections that can’t be done by face-to-face communication.

2. Some products and services are
complicated sells, you need to give all relevant content to cater for a
diverse audience.

3. Some people like detail and have time
on their hands, leave them satisfied.

4. Your brand will be unknown to many
visitors, you need to convey credibility and reliability and content can
help achieve this by giving reassurances.

Getting the basics right – logical
hypotheses

To help devise logical hypotheses (i.e.
assumptions for what content you need to create to satisfy the needs of
visitors
) for your landing page, use the following indicators:

1. If it is product based, what are the
most popular/fast-selling products in your catalogue that can be highlighted?

2. What are the current site search
patterns related to the subject of the landing page?

3. What are the market search patterns
related to the subject of the landing page.

4. What questions does your Customer
Service team frequently get asked related to this subject (helps tackle
barriers)?

5. What are your competitors doing in
this area?

For example, re: 3. above, if you are
creating a landing page for home fitness machines such as exercise bikes and
rowers, use the free Google External Keyword Tool (or access it via your Adword
account) and Google Insights to identify the most relevant trending keyword
combinations. This will help you match content to popular searches to increase
relevance.

This also helps identify keyword targets
for SEO optimisation.

Google Insights for search

Visualising where people scroll to

Web analytics data will tell you which
buttons and links visitors click on and the % of visits clicking on each
on-page element, but it won’t tell you how many people are viewing each
section of the landing page.

This additional information is really
helpful as it demonstrates the propensity to scroll and absorb your content.

If
people are scrolling and seeing everything but engagement and clicks are low,
your data is telling you that your landing page is not strong enough at driving
action.

This can help you modify content to encourage greater engagement, such
as testing stronger calls to action and more prominent visual signs.

I love the Crazy Egg scrollmap and confetti
tool for visualising landing page activity. You can integrate with your GA
account to refine the display based on traffic source, great for working out
how landing pages cater for different audiences.

You can also get this sort of info from companies like Clicktale.

Crazy Egg scrollmap
Crazy Egg confetti

Testing short v long form

And this is the crux of it. Not even
Smarties have the answer. If you want the best results, or at least to know
that you have stuck with the optimum solution (that is the landing page that
gives you the best set of results based on your KPIs), you need to test.

Testing often makes people look away to
a far off happy place – its implied complexity hurts. However, A/B testing
doesn’t need to be complex or prohibitively expensive.

You can use tools like Google
Website Optimizer to do this and create a few creative treatments for the landing
page. Depending on traffic volumes, you can have readable data within a few
days.

Testing doesn’t have to be finite.
Each time you find a winning page version, set-up another test for a specific
component of the page
(e.g. style of buttons for calls to action) and refine
further
. It’s an iterative process reflecting the changing needs of your
visitors.

There’s a great case study from
Conversion Rate Experts on Sunshine Holidays: “How we made an extra £14m a
year for a travel company”
.

Detail via web analytics

The challenge with your testing is to
find relevance. If you simply compare top-level KPIs for landing page versions
based on all visits, you miss out on the golden nuggets.

For example, you may
find that your search visitors behave differently to email visitors and within
search, brand behaves differently to generic.

This type of analysis can reveal
interesting patterns that might lead you to creating multiple versions of a
landing page to suit different audiences. The marginal cost of doing this can
help drive your ROI and give you pay-back.

If you’re opting for the more
straightforward A/B testing route, you can use web analytics tools like Google
Analytics to measure this.

  • Compare KPIs like page engagement
    and conversion based on traffic source.
  • Compare performance for new v
    returning customers.
  • For search traffic, drill down to
    keyword level to see how landing page performance varies at keyword level.

As far as I know, it’s also possible
to use GA to do this for MVT testing but it’s more complicated and you would
need to use custom variables. That’s not my expertise so would welcome
comments from those in the know.

Drilling down into the data will give
you insight from which you can make decisions. You might find that short tail
keyword traffic responds better to a long form page because the in-depth
content satisfies the research needs.

You might not. But if you don’t do the
analysis, you might be missing something important and compromising results.

Simple techniques to encourage visitors
to scroll

If you’re opting for a long form
landing page there are some cool tips for getting visitors to get past first
base and go deep:

1. Ensure key selling points and calls
to action are above the fold – yep obvious but not always adhered to.

2. Use links to other sections of the
page to fast track people to content lower down the page.

3. Avoid horizontal blank space on or
near the fold – this gives the impression there is nothing below and impacts
scrolling.

4. Use levers in the content to
encourage scrolling (Wiltshire Farm Foods found a neat solution for this – see bottom right of the page).

5. Use technology to surface deeper
content without the need to scroll or open new pages e.g. product carousels.

The best article I’ve read on
scrolling tips comes from Conversion Rate Experts: “How to make users scroll
down your page”
and its thanks to them that I found the Wiltshire Farm Foods
example.

To check what content appears above the
fold for different browsers, you can use the Foldtester.com free online tool
(see below).

Via your web analytics you can identify what % of traffic comes
from different browsers/resolutions (including mobile browsers) to determine
how best to optimise display.

Foldtester.com fold visualisation 

Still don’t think long form landing
pages can work?

I love a sceptic. If it really didn’t
work, then why would some of the most successful digital brands use oh so long
product pages?

Now don’t get me wrong, I’m not saying lets all rip up
the concise approach and go nuts. There’s a time and place for everything,
just don’t exclude long form because you have a personal issue against it.

It
really can work and your landing page design should be based on what visitors
want; let the data make the decision.

There’s a neat visualisation of one of
Amazon’s product pages on the Conversion Rate Experts blog (see above link) using the Hoff as a scroll barometer. It’s
mighty big.

So what do you think? I’m pretty much
agnostic when it comes to the question, “Should my landing page by short or
long?” I don’t really care. All I care about is finding the best solution
to drive KPIs, provide good user experience and achieve the goals for that
page.

Please drop by with comments to fuel the
discussion.