Landing Page Optimisation is a PPC centric session, first thing in the morning of day one and right after the keynote speech. It was moderated by Jon Myers and presented by Dr Karl Blanks of Conversion Rate Experts and Nathan Richter of Monetate
Nathan Richter of Monetate
It’ll be easy to write up Nathan’s presentation as it’s already online. I think it’s the same presentation as he used for SES San Francisco but I’ll embed the SES London upload below just to be safe.
I don’t think it’s appropriate to embed every presentation. That’s unfair to the for-profit conference but where speakers have made the slides available outside the SES infrastructure then I’ll use those.
Check out the Q&A section below for more of Nathan’s insight.
Dr Karl Blanks of Conversion Rate Experts
Okay, okay – I admit it – Dr Blanks may have lingered for a few minutes too long on his own company at the start of the presentation, triggered by pathological hatred of any sales pitch that dares to appear in a skills conference and may have triggered a few grumpy tweets.
I was won around by the end – perhaps not by the presentation but from Karl’s clear abundance of knowledge in the area. For example, as a throw away comment he recommended Voices.com as a good place to go to get hassle free voice overs.
The Conversion Rate Experts presentation can be summarised as a “method acting” based approach to creating landing pages. They encourage us to become our clients’ customers in order to design the landing page.
There are two key skill sets;
- Be able to sell the product in a face to face situation
- Be able to write that pitch down
Become a customer – go through the process of buying something from your client and use your own money. Blanks wryly and rightly points out something odd starts to happen to purchase decisions when you’re using company money.
Why use fictional customer personas if you can use real ones. If you’ve really gotten to know the customers and company well enough then this should be possible.
Blanks recommends Kiss Insights as a great web-enabled survey collection solution.
Writing your sales pitch down and turning it into a landing page seems quite a balancing act. Blanks suggests you use as many words in your landing page as you would use verbally when selling the product. However, he also argues to be concise and to edit all the time.
He uses a template to help construct the verbal argument in a landing page. Very roughly it begins with an opening sentence that the reader relates to (the ‘that’s me’ moment), a catchy headline, a joining on paragraph followed by some carefully composed built points. That’s finished off by the appropriate call to action and techniques like urgency creation often work.
Landing Page Optimisation Question & Answers
Question: What’s the most important part of landing page?
Answer: That’s a bit like asking a mechanic what the most important part of a car is, responds Blanks. One way to think about it is to find out why your users aren’t converting at this minute.
Question: What’s a good recommendation for a multi-variant testing tool that can cope with different user agents – ie, mobile
Answer: Nathan Richter from Monetate recommends Monetate.
Question: In the keynote session Avinash Kaushik of Google and analytics expert urged the audience to go beyond simply measuring conversions. How does the panel respond to that?
Answers: Blanks argues that you should not just optimise for the sale – optimise for other successes too. Richter encourages developing a list of KPIs and assigning a monetary value to them
Question: In May the UK will fully adopt the EU’s ePrivacy Directive. This means that web visitors cannot be cookied without their consent and without explaining what the cookie is for. Does this not destroy multi-variant question as we know it today?
Disclaimer: This was my question. It was evil.
Answer: Richter responses by agreeing that the industry faces a huge challenge but that it’s not the only one. Hopefully we won’t end up in the situation where effective multi-variant testing becomes illegal. He points out there is possible scope for multi-variant testing the most effective form of permission prompt.