Hi there!

Welcome to the Unbounce #CTAConf notes. We’ll be here for all 2 days, 17 sessions, and 400+ attendees’ worth of conversion and digital marketing magic. We hope you find these notes truly mesmerizing and, of course, you can tweet at us if you find something you’d like us to include!

Notes brought to you by Unbouncers:
Mark John Hiemstra,

Brad Tiller, @bradtiller

Enjoy the rest of the conference!





[Tuesday, September 15th]

Don’t Do Anything on Your Landing Page You Wouldn’t Do on a First Date (Slides)

How Can You Foster a Better Culture Around Optimization?

Video for Conversion & Beyond

Full Stack Panel: Actionable, Practical A/B Testing

From Conversion to Manipulation: How Successful Marketing Practices Persuade

How to Generate Those Leads All Day, Everyday

How to Conduct Solid, Data-Driven Conversion Research

Death to Fear and Laziness! How to Push Yourself to Write Sticky Landing Page Copy

[Monday, September 14th]

Brand & CRO: Trick or Treat, the Choice is Yours

3 Impossible-to-Please Copywriters Tear Down and Rewrite Your Headlines Live

Pulling the Emotional Trigger: Psychological Tactics to Convert Your Leads into Customers

Full Stack Panel: Send Emails That Actually Convert

A Framework for Mobile Conversion (Slides)

Persuasive & Perverse Psychology: How to Win Customers and Influence People (Slides)

Grow your scientific knowledge
Brand-Infused PPC for Better Results (Slides)

The Four Corners of Conversion: Understanding the Intersection of Copy, Design, Interaction & Psychology (Slides)


[Tuesday, September 15th]

Don’t Do Anything on Your Landing Page You Wouldn’t Do on a First Date (Slides)karl-gillis.jpeg

Karl Gillis, @AGConsult

1. Don’t start with a French kiss

2. Be on time  - Fast loading times are important

3. Don’t tell your life story in the 1st minute

4. Don’t shout for attention. Give attention.

5. Make sure you get the attention you deserve.

6. Show interest in your date. Don’t make it all about you.

Some of Karl’s fav questions to ask:

On website

Existing clients

7. The fun factor: Make your date laugh and you'll win their heart.

8. Be yourself on your first date

9. Your first date isn’t a police interrogation

10. Once again: Be patient

11. Initiate a next step (Handshake, kiss…)

Part 2: Don’t take my advice for granted

Please remember - don’t believe what I said. A lot of my tips and best practices do work on most pages.

But sometimes they will fail. Because your website, your product, your audience, your design … is unique. So please: start testing instead of copying.

How Can You Foster a Better Culture Around Optimization?joanna-lord.png

Joanna Lord, @JoannaLord

        There are three ways your testing culture exists today

1. We think testing is a really good idea

2. We do it just enough to cover ourselves

3. Testing is the gateway to our future. << This is the only way to a testing culture. The others are just not quite there

Everyone should have the ability to recommend a “big swing.”

Are you getting good ideas from a meeting? Stop the meeting, put it on the whiteboard, and start talking about the next steps.

Go work at a place where you can have an impact.

Video for Conversion & Beyondphil-nottingham.png

Phil Nottingham, @philnottingham

Does video help increase conversions?

Copy and image, all at once.

Where can video help increase conversions?

Video = form, it’s a media type.

What kind of videos are we making for what context and why?

Explainer videos, live action tutorials, webinars.

How can we use video to drive more leads and increase conversions?

Landing page videos

Social videos - contextually broad

The best videos are always goal-driven

When you see video marketing that doesn’t perform well, it’s typically because it’s created with an unclear goal in mind.

What sort of goal should we be creating video for, then?

If you always try to create content that is there to be closing, it comes across as salesy and you fail to create videos that resonate.

MVC: Minimal Viable Conversion: What’s the next minimum action that someone could take after consuming this piece of content that would be valuable to you in your overall conversion funnel.

When someone watches the video, what do you want them to do next? In reality, whenever you’re watching a video, it’s not usually conversion-focused. You’re not thinking, “I need to go and buy something!”

More realistically, you’re introspective/thinking - which is where Actioning comes in: expressing a goal as a transitive verb.

Use this as a framework for a brief for a video: “The purpose of this video is to ACTON [compel/excite/amuse] the audience into completing the MVC.

What kind of MVCs are suitable for video?

Landing page video:

Social videos:

You’ve got your goal - the next step is optimization.

Three things you should optimize for:

  1. Click/Play, AKA increasing play rate:
  1. Customize your video player. At Wistia, lots of data around the way people are using videos. They’re going to share them with you :)
  1. Branded player colour =19% increase in play rate
  2. Custom thumbnail = 35% increase in playrate (Moz example) (Think of thumbnails like movie posters - it doesn’t necessarily have to be a frame from the video.)
  1. Hack: Go into photoshop and add an overlay which explains the value proposition of the video itself
  2. Wistia video example, paid photographer to take photos of the shoot. Allowed them to integrate the story throughout the landing page, blog page and supporting media. Show an element of the parade without giving the story away completely, encouraging them to watch the video. And possibly makes for a great thumbnail!
  3. B-roll footage: “announcement and video coming soon” - tiny snippet used on FB and Instagram. Building demand for the content before they even released it.
  4. Optimize social video for clicks, not keywords.

Optimizing videos for Youtube

Optimizing videos for Facebook

  1. Watch video

  1. Completing the MVC (call to action)
  1. Annotations (links on video)
  1. Align annotations to existing points where people are rewatching the video - go to your engagement graph in wista and see where people are rewinding to watch again - that part of the video is interesting to them. Ask why it’s interesting and then create an annotation that aligns.
  2. Consider annotations in post-production - in YouTube, that roll at the end of the video that is super branded. They design that in After Effects and then add links in YouTube. Think about including them in the post-production stage - designing them.
  1. Full-screen CTAs
  1. ie Whiteboard Fridays - Getting people in the funnel to watch more content, consume more of it and eventually become a lead.
  2. In YouTube, “cards” allows you to do a similar thing.
  1. Turnstyle
  1. CTAs or keys that are overlaid to collect leads or unlock further content.
  2. Data:
  1. 25% put turnstiles at the start = 16% conversion rate
  2. 25% put turnstiles in the middle = 24% conversion rate
  3. 49% at the end = 3.4% conversion rate

[graph “Turnstile Conversion Rate by Location”]

5-10 seconds into the video seems to be the sweet spot.

What about seeding and distribution?

To make sure you’re doing this properly, run lean tests before investing.

<Caveat>: Good video helps increase conversions.

Only 0.25% of landing pages have video - if Lenny the dog can do it, then you can too!


Full Stack Panel: Actionable, Practical A/B Testing

Michael Aagaard, @ContentVerve
Peep Laja, @peeplaja

Ton Wesseling, @tonw

Ryan Engley is moderating and will be asking the panel questions.

Q: Should all the attendees be A/B testing?

The panel all say no.

Ton asks the audience to stand. Less than 100 conversions per month, he says sit down. Not very many people sit. Less than 500 conversions per month now. Still about 60% standing. Less than 1,000 conversions per month still about half. Now he asks that those who are running A/B tests stay standing and most do. And now we’re down to fewer than 10,000 conversions.

Here’s why: If you have fewer than 1,000 conversions per month, there’s no point in A/B testing.

Peep: Math is not magic. There is no magic number for how many conversions you need.

Michael: You might need softer conversion rates.

Peep: A/B testing is not optimizing.

Ryan: Some people think A/B testing is THE way to optimize. They should still be optimizing, but what is it?

Michael: If you don’t have enough traffic, A/B testing is not helpful, but one thing that always is helpful is doing your research on your customer.

Ton: It takes a lot of effort to get an increase in conversions of 15%. That’s huge. But you have to take risks.

Michael: You have to realize what facilitates change. You need a hypothesis to understand what is driving change.

Peep: You have to remember we are not in the business of science. We are in the business of making money. Change things, it works, who cares what it is, as long as it works.

Michael: If your research says there are 10 things wrong with your landing page, then change everything instead of trying three or four changes.

Ryan: If someone is getting started with A/B testing, what kind of research should they be doing?

Ton: Most small things make a small impact. You have to take bigger risks to get bigger rewards.

Michael: To answer the original question: Basic stuff is a step problem analysis. Where is your website leaking money? It really helps to find out where the biggest leak is. Then figure out what device people are using, what browser, a lot of technical bugs you may have that you can find out and change. You see with ecommerce that checkout funnels are often flawed, and that’s something to optimize.

Peep: If you’re asking what should I test next, then you’re doing it wrong. A/B test is a solution to identify a problem. You need a testing roadmap. Never ask, you have to know and you learn that through research.

Ton: If you can’t do A/B testing because not enough traffic, try out heat maps. they’ll tell you a lot about interaction. Also, talk to your customers. They’ll give you great answers on what they’re looking for that can help you a lot.

Michael: Talked earlier today about how important it is to do homework. If you don’t have enough traffic for A/B testing, it’s even more important to do that research.

Peep: You need to be able to measure how good your testing program is. There are three metrics to pay attention to: 1. The number of tests you run. 2. Percentage of winning tests. 3. The impact of the testing.

Ton: You want to learn about your user. You don’t want to just change the color of the button. Testing for the sake of testing is not real testing.

Peep: Testing tools make it so easy to do testing that anyone can do it… and they do.

Ryan: What should someone do to get started with research that they have when they’ve identified a problem area?

Michael: One of the things we saw at Unbounce with the signup process was that they needed four pieces of information and then after they click the CTA button all of a sudden they get a bunch more fields plus credit card and many abandoned it. User testing can really help you optimize your website.

Peep: If you don’t know what people are doing on the page, you’re in the dark. You need to record what’s happening on your page in order to identify connections between certain behaviors and conversion rate.

Ton: Check to see what elements of the page people are using. Sometimes he eliminates  one element of the page just to see what happens and then he can discover what matters.

Peep: If you don’t know what matters to your customers, you have to figure it out, or you can’t  optimize.

Michael: Been surprised many times to learn that what customer service says and what’s on the page are often two very different things. And it’s always easier if you have just one product as opposed to many.

Ryan: We’ve found something to optimize and then set up a test. What’s next?

Ton: The data research is very important up front.

Peep: You need to calculate your sample size up front. You don’t stop at 95% confidence, you stop when you have enough traffic. You can use tools to accomplish this.

Ton: That’s why we say a limit of 1,000 conversions per month at least. You have to test for six or seven weeks to get enough traffic to really understand what happens. In that time, people delete cookies. So you have sample pollution. Within 2 weeks you can get a 10% dropout of people deleting cookies and that can really affect your sample.

Peep: When do I end A/B test. 1. Enough sample size. 2. Test duration, or how long a test should run. You need a representative sample, not a convenient sample....

Ton: You have to calculate up front and don’t stop until you get the results you’re after.

Peep: Statistical significance only makes sense once you have sample duration. Before that it’s a meaningless number.

Ryan: What happens if cookies are expiring, but you haven’t reached significance?

Ton: Ton runs A/A tests (completely the same page) against each other. 77% of the test throw out a winner after one day or four days or seven days. Those are false positives. That’s why you have to calculate up front.

Michael: If you know how long to test, then a false positive is something you expect to see.

Ryan: How do you document all the tests you run to make sure you’re constantly learning?

Michael: There are so many ways to do it, but the main thing is that you actually share it. A big part of this is documentation. One thing Michael uses is Evernote and he puts all his screen dumps in there to keep track. If you get too complicated you don’t do it because it’s too hard.

Ton: When we started we would just use PowerPoint presentations, but then there were so many that they wouldn’t be able to find it. Then they switched to Excel, but they found it difficult to share with the rest of the company. Now they use Trello to keep track and the DB is searchable, so it’s out there for them to use.

Ryan: What kind of a staff do you need to optimize?

Peep: An optimizer is someone who needs to know a lot about a lot.

Now Questions from the audience:

Q: My question is f you were looking at outsourcing, how would you select a marketing firm?

Michael: I would ask them about the process. Ask them what their testing process is.

Ton: If they have no data and no process, then don't’ bother.

Q: I just doubled my signup rate which I thought was awesome. But boss said that those ones aren’t spending any money.

Ton: You have to get the right KPI. What can I learn about the lifetime value? Once they’ve signed up, now a whole new CRO stage starts.

Michael: Don’t forget to integrate test data with analytics. Don’t just look at averages. If you’re only looking at two different creatives, you’re just getting averages and it’s not enough to understand.

Ton: I never use the analytics part of the A/B testing tool because they want you to be successful and keep using it. I do my own math and I use my own GA data.

Q: As marketers we get in the onslaught of CRO, SEO, etc, so we don’t always know where to spend resources. As CRO experts, can you tell then that the product just isn’t good and there’s a new plan needed?

Peep: We have discovered through research that sometimes people just don’t like the product.

Michael: That goes back to the research. Sometimes marketing has the burden of promoting a product that just isn’t good.

Ton: A/B testing is not to optimize conversion, but to grow your business. Do whatever works to grow your business. A/B testing is not a solution, it’s a way of thinking.

Final thoughts:

Michael: Split testing is not an excuse to skip homework.

Peep: If you don’t know what to test you’re doing it wrong.

Ton: Use testing to learn from your user.

From Conversion to Manipulation: How Successful Marketing Practices Persuadenatalie-nahai.jpeg

Nathalie Nahai, @NathalieNahai

Ask three questions:

How to Generate Those Leads All Day, Everyday mike-king.png

Mike King, @iPullRank


Data-driven personas

Qualifying leads

Personas should dictate qualification

Nurturing leads is about not being a “one night stand”

Low effort vs. High effort lead gen

NATURAL DEMAND VS. ARTIFICIAL DEMAND: Mike wanted to determine what works better, a campaign where we just throw up a landing page and capture existing demand or where we do in-depth research and create lots of assets.

For high-effort lead gen

Here are tactics that usually work.

How to Conduct Solid, Data-Driven Conversion Researchmichael-aagaard.png

Michael Aagaard, @ContentVerve

Michael Aagaard, legendary optimizer and now our very own CRO! He was also the first person I ever interviewed for the blog. I made this lovely image of him:

Aagard has build a custom landing page optimization report for Google Analytics! Get it at bit.ly/1Mpg3lf << Aagaard’s report shows you the conversion rate and bounce rate by device at a glance. This thing is killer.

(Running theme of the conference seems to be looking at CRO as a framework for business rather than a mere digital marketing optimization tactic.)

Death to Fear and Laziness! How to Push Yourself to Write Sticky Landing Page Copyjoanna-wiebe.png

Joanna Wiebe, @copyhackers

  1.  “You fill in the words.” Copy works best when copy leads design. If you’re squeezing copy into your design, you’re doing it wrong.
  1. You default to easy wins. You might jump to incentives or discounts in order to get numbers up. Sometimes things like this work, but they don’t always work
  1. You say what your competitors say. We say things that are very similar to what our competitors are saying.
  1. Use tools. Google Quality Score will help with message match. The Hemingway app will help you make sure you’re being clear. Headline generators are also fantastic. Use them to drum up new ideas.
  2. Use headline formulas. Example: [Product name] is [product category] that [what it does that people want.]
  3. Be deadly. We’re often told to write emotionally. Her example is the headline “They laughed when I sat down at the piano. But when I started to play!~
  1. Swipe from customers. Let your customers write your copy, or, at least, help inform your copy. They’re not sitting around a boardroom -- they’re using your product. Joanna wrote some copy for Crazy Egg that was entirely directed or informed by what customers had said and it performed very well.


[Monday, September 14th]

Brand & CRO: Trick or Treat, the Choice is Yourswil-reynolds.png

Wil Reynolds, @wilreynolds

Wil has spent the last 11 years figuring out how to get more traffic or analyzing that traffic’s impact on business.

Brand and its role in CRO

Wil was always a direct marketer, who became a reluctant “brand guy.” But bringing a talented designer in, who started tweaking the brand, made him feel more proud about the company – it’s like how you feel when you look in the mirror.

Who is the best CRO? Wil knows some great ones, but …

CRO understands friction and removes it. That is a business function, not a web function.

Winning with why

Ask for permission to win the war. 

Inserting an CRO at a non CRO table

The impact of external factors 

3 Impossible-to-Please Copywriters Tear Down and Rewrite Your Headlines Livedemian-farnworth.pngamy-harrison.pngjoanna-wiebe.png

Joanna Wiebe, @copyhackers
Demian Farnworth,
Amy Harrison, @harrisonamy

  1. Buttons and headlines work really well together. You can repeat the headline in your cta.
  2. Think of your button not just as a cta, but more as a call to value. “I want to __________.” That can become your button copy.

Pulling the Emotional Trigger: Psychological Tactics to Convert Your Leads into Customerstalia-wolf.png

Talia Wolf, @TaliaGw 

Full Stack Panel: Send Emails That Actually Convertcolin-nederkoorn.pngjustine-jordan.png

Colin Nederkoorn, @alphacolin

Justine Jordan, @meladorri


Justine picked up right where Colin left off, talking about an email lifecycle campaign that she uses at Litmus.

Lifecycle Campaign Emails

The Six Stages to Subscriber Experience

  1. From name.
  1. You could use a person’s name, brand name, combination of the both -- test and find out what works best.
  2. Respond to no-reply@companyname.com? No. That sucks. Don’t do it
  1. Subject line. 
  1. There is not about what you think is the perfect subject line!
  2. It’s about what’s best for your audience.
  3. “Free” is okay. Shorter is better. Be relevant. Be specific.
  1. Preview text.
  1. It’s the text that shows up in your inbox after the subject line. Optimize the crap out of that text. It can improve open rates significantly.
  2. Make sure it doesn’t say “Email not displaying correctly.”
  1. Open. 
  1. 43% of emails are going to be viewed without images. Optimize for that.
  2. Optimize your images!
  1. Click.
  1. “I like big buttons and I cannot lie.” Make it easy to click. Especially on mobile. Remember iOS is trying to help you do things like map to your address, etc.
  2. Bigger is better. Big copy, headlines. Make everything tappable.
  3. Don’t forget to test your CTA copy! Low commitment CTAs work well on email
  1. Page/Site. 
  1. Don’t forget that you have to send someone to a… landing page! If you’re sending an email that doesn’t direct anyone to a landing page, you’ve wasted their time and yours.

A Framework for Mobile Conversion (Slides)tim-ash.png

Tim Ash, @timash

The big picture of mobile conversion

The Unique Contexts of Mobile Use:

The mobile visitor is...

It’s an uncontrolled, unpredictable environment.

The Mobile Conversion Framework:

“It depends. If I am to speak ten minutes, I need a week for preparation; if fifteen minutes, three days; if half an hour, two days; if an hour, I am ready now.”

Make Phone Contact Easy

Persuasive & Perverse Psychology: How to Win Customers and Influence People (Slides)bart-schutz.png

Bart Schutz, @BartS

Four categories in the SQUARE OF CONVERRRRRSION

  1. Wake up ration
  1. When you need a ratio, but it’s asleep. So how do you wake it up?
  2. Start by doing something that is completely unexpected. Go against habits. Those popups at the bottom of the screen are interrupting your rational system (1).
  3. We remember and like things more when we thought of it ourselves

  1. Keep ration awake
  1. Now to the Keep Ration awake principle. This is the proof that your product is best.
  2. “Pure rationality does not exist.”
  3. Ration needs emotions to add value to the argument.” - Descartes Error
  1. Deplete ratio:  
  1. When you need emotions, get rid of rationality.
  2. You might actually want to add a little friction
  3. You might give them more choices. Instead of ordering now, they could just add to wishlist. No one does. They buy or they don’t: this is the rational brain.
  4. Bart has thoughts on ethics. He wants everyone to A/B test. He wants you to make sure that you’re doing it without sacrificing your own set of ethics.
  1. Keep ratio asleep.
  1. When you need emotions, make sure rationality does not wake up.
  2. We believe we buy because we thought about it. The reality is in the online world, we’re being convinced.
  3. Our brains feel uncomfortable for not having reasons for making a decision.

Dana DiTomaso, @danaditomaso

The Four Corners of Conversion: Understanding the Intersection of Copy, Design, Interaction & Psychology (Slides)oli-gardner.png

Oli Gardner, @oligardner

Oli has taken the stage, resplendent in a pair of orange shoes, reminiscent of the beautiful petals of the tiger lily, otherwise known as the Hermmerocallis fulva (#fact).

Oli hates bad marketing. A LOT. He provided an example of bad marketing: Google search for “Vancouver whale watching tours.” Search results (and corresponding landing pages) include pictures of bears, restaurants.

This is a poor experience. Marketers should make their visitors feel like they’re “Walking on Sunshine!” How is this done?

The Four Corners of Marketing: