It has been an excellent start at the Hero Conference 2014. It’s a great environment full of PPC professionals eager to learn and share insights and studies.
Several good presentations are in agenda, I’m sad to miss some, but we have to draw the line somewhere.
I have collected all the main takeaways from the presentations I attended on day one. Here was my agenda for the day:
1. Bryan Eisenberg’s keynote ‘Perfecting Landing Pages’.
2. ‘The Art and Science of Managing a PPC Team’ presented by Jeff Allen and Paul Jeszenszky.
3. ‘Digging Deeper with Scripts, Macros and Custom Reports’ presented by Frederick Vallaeys and David Pedersen.
4. And of course. You can’t skip a chance to see DataPop present on ‘“Feed”ing The Beast: Winning Results from PLAs’.
Without further ado, I leave you with the most important points touched on these presentations:
1) Perfecting Landing Pages with Bryan Eisenberg
Bryan Eisenberg did a good job at making 90% of the PPC professionals in the audience feel embarrassed about the small amount of landing page testing we are making.
Most advertising campaigns (including PPC) are not creating enough landing pages to back up their advertising campaigns and are leaving business on the table via this approach.
Most websites don’t have a traffic problem. There are tons of ways to bring traffic to a website. On the other hand, every single website has a conversion problem.
Everyone is using the same recipe across all the websites that are being built. The success of the final website is determined by the biases of the people who create the website/landing page.
Depending on who is creating the landing page, the success will change. Bryan mentions if a copywriting guy creates a website, the copy will be super focused, whereas an SEO guy will focus on keywords and other SEO efforts.
The Curse of Knowledge
We bring our own biases about what we know about and the service we offer and try to explain the same subject to people who don’t care as much as we do about our company. Many of us create landing pages that are written for ourselves instead of for our real customers.
Conversion Rate Optimization Is a Journey
It’s crucial that you make sure that you keep the same scent (design, copy, offer, etc.) throughout your entire conversion funnel.
He mentions an amazing example from Geico where the ad and landing page were an amazing fit, but the sign-up process was a generic, boring sign-up flow.
The Hierarchy of Optimization
You need to take your user from Functional –> Accessible –> Usable –> Intuitive –> Persuasive.
Many people buy from poor websites because their offer is so strong. Many websites are struggling with even the basics. We don’t get it right and we therefore never get to the important stages (i.e. persuasion) when our site navigation isn’t working.
Retailers lost over $44B due to transaction problems on their site. Bryan asks how many billions have been lost due to usability problems.
The Average Customer Is a Myth
Average doesn’t work and shouldn’t be applied to marketing. You need to use one on one marketing by using Personas or segments on your website. Only by advertising to single personas at a time can you maximize your conversions.
PPC Marketers Don’t Create Enough Landing Pages
On average, Unbounce (Oli Gardner was called out on the spot, so the number might vary) said their customer creates 15 landing pages. All PPCers at the conference agreed that we use more than 15 keywords per campaign, so we should definitely also create more landing pages than this.
Keywords Don’t Fail to Convert
We as PPC professionals failed to convert the keyword. We failed to understand what that keyword meant for the customer. If we understood the relevance between keyword and consumer, we could have converted the click to a customer.
Often, we bring visitors to bad landing pages. It’s not the ad, not the keyword, not the search term. It’s the landing page.
Jeff Bezos – A Long Term Investment
Google created Product Listing Ads to compete with the incredibly efficient PPC ads that Amazon was running. They were so specific that Google needed a way to let their advertiser compete with Amazon.
Amazon is focused on the customer experience. Their brand promise is ease and amazing customer experience.
Amazon did over 1,900 tests last year. How many new landing pages did YOU actually launch per month last year?
The Conversion Trinity
The biggest take-away from Bryan’s keynote was that it’s all about the Conversion Trinity and how it applies to landing page optimization:
1) You need to be relevant to that person at that time.
2) You need to express value. Why is your offer better than your competitors’?
3) Have you built enough confidence for me to take the call to action?
This formula can be applied to every single ad, landing page and website.
Furthermore, Bryan took us through his own process of how he’s tried to simplify the landing page optimization formula the last 14 years. Starting with over 100 different factors that were decreased to 36 factors in his A/B testing book. He is now down to “10 ingredients of every successful landing page”.
All landing pages need to include the following elements to be successful. Go through each one of these items and write every single one again and again.
Prioritize how you want each area to be shown.
2) Headline: Text and Graphical
4) Descriptive Copy
- List of key features
- List of key benefits
5) Product/Service presentation
- Product Image
6) Calls to Action
7) Confidence Building
- Examples of users
- 3rd party validators
8) Contact information
9) Link to more information
10) Template elements
You always want to test your headline and your offer.
2) Getting The Most from Google Shopping Ads
When a company such as DataPop talks about Comparing Search Engines, you listen. Unfortunately Dave from DataPop didn’t get quite as much time to talk as I would have liked, but his tips for managing a successful data feed were very useful.
Dave reminded me about one specific thing: Your data feed is your AdWords ad when it comes to Google Shopping campaigns.
You can’t set and forget your feed, and thereby having your internal product names define your ads. You would never put the word Sunshine Dress in an ad that is supposed to show for Yellow Dress. Dave reminds us that we’re search marketers and we should apply what made us so successful in traditional search into Product Listing Ads.
Ensuring that your feed is well structured has just become even more important now that you have Google Shopping. Google Shopping campaigns are explicitly built on your feed, so you need to start with your data feed before you create Google Shopping campaigns.
Takeaway tip: Try the following structure for building out your data feed product titles:
“Color + Brand + Category”.
3) Digging Deeper with Scripts, Macros and Custom Reports
The most accredited source on AdWords Scripts is without a doubt Frederick Vallaeys. His presentation on useful tips for proper AdWords Scripts usage was one of the most popular sessions and I walked away with several takeaways.
Working with AdWords Scripts is as simple as copy-pasting pre-made scripts. However, you should always aim to preview the scripts that you intend to run. There is no undo button, so don’t just run it without checking that it won’t ruin your account.
Frederick provided a good tip about using the AdWords Editor as a quick, manual way of taking backups.
He provided a list of resources where you can find free and paid scripts:
4) The Art and Science of Managing a PPC Team
The head of Search in Periscopix, Emma Welland shared a very nice presentation about training of new team members for PPC. It was funny to see how much it coincides with our own initial training program (which runs for 5 weeks rather than her 4 weeks), but overall the content was very similar.
Emma talked about training being one of the most important aspects to keeping employees happy. Employees are not given enough training to keep growing as professionals.
- Recruitment (the right people)
- The most important skill: Wants to be in PPC. At Periscopix they ask all potential employees to do a presentation on PPC. This weeds out a lot of people who are not interested in PPC.
- New Joiner Training.
- Continual Development
Breaking Down the New Joiner Training
Compulsory Structured Program: Train them in the way that you do things. Everyone, no matter his or her background, should go through the same basic training.
Presented by Experts: Don’t let your Junior people do your training. Junior people training new hires results in you ending up with Junior people. Senior staff also better pick up when their training is not working on what needs to be trained further.
Wide Company Picture: Train people on the entire spectrum of your field. Both creation/ optimization and specific skills if you’re splitting up the tasks.
Mixed Approach: Allow for various personalities pick up the field. Get different types of material (audio, video, text, visual, etc.).
Feedback: Ask for Feedback from the new staff. What do they think we can do better? what did we do well?
Everyone passes the AdWords Certification Exams after 4 weeks.
5) Ad Testing in a Multi-Device World by Brad Geddes
Brad Geddes finished the day off with a heck of a presentation. Brad is known for being one of the most interesting guys on stage in PPC, so my hopes were quite high.
However, when he started talking about Rotate Indefinitely vs Google’s Conversion Optimizer, I started to worry about whether he was going to do a basic presentation. Boy was I wrong.
Unfortunately, my computer ran out of battery throughout his presentation, so I didn’t get to type it all up, but the parts I actually covered below were gold. He shared a ton of golden nuggets to ad testing in a multi-device world.
The main take-away from Brad is that you no longer can run campaigns using general ads. Always, always, always include mobile-preferred ads and if you don’t, you need to look at data segments when determining test-winners. The best ad for mobile will most likely not be the best one for desktop.
Below are my notes from the session:
You can’t make a decision when it comes to ad testing without using segments for mobile/desktop at least.
Many ads work best on mobile, but don’t work well on desktop. Therefore always use mobile-preferred ads, so you can control your ads per device. Way too few advertisers use mobile-preferred ads. Brad mentioned numbers that were below 3-4% of all ads are mobile (graph).
Start with the ad groups with the highest amount of impressions. This will allow you to get started quickly and see the results of your increased labor.
Step one is the first step to gaining control of your ad serving.
Ad Testing Metrics
You often don’t have enough data in the AdWords interface to make a decision about what ad is the best. Brad shows an example of 5 ads, where all 5 are best in their own way.
Each individual Ad Testing Metric (Conversion Rate, CPA, ROI, ROAS) is not good if you’re only looking at them individually.
“Every time you have an impression, you have the possibility of a conversion”.
Brad shared an example where the difference between the lowest CPC ad and the highest CPC ad were more than 600%. Quality Score is immensely important when it comes to ad testing.
Brad recommends Conversion per Impression if you’re in lead generation. This will help you get the biggest amount of leads possibly regardless of the individual CPA, CPC and CTR metrics.
For ecommerce retailers, you want to use Profit per Impression.
Remember About Lifetime Value
Brad shares an example about web hosting companies paying a $5 CPC for a $5 product. The web hosting companies of course know that this isn’t true. It’s $5 CPCs for a $5 monthly product.
Minimum Viable Data
With very few 97 impressions, even though you have statistical significance, you still need to wait until you have minimum viable data. Brad shared an example where the worst performing ads were shown as winners after 97 impressions, but after 3,215 impressions the results were turned around.
There were numerous examples where Google’s way of picking winning ads is inconsistent and based on very little data.
Minimum data recommendations:
- Min: 1 week. Each day is different. Allow ad tests to run at least 7 days.
- Better: 1 month. Each week has variants (pay weeks)
- Min: 300 clicks per ad
- Better: 500 clicks per ad.
- Best: 1000 clicks per ad.
- Minimum: 7 conversions per ad
- Ideally: 15 conversions per ad
You can make decisions faster if the results are incredibly different (15% CTR vs 1%).
Multi Ad Group Testing
This is what you can do across several ad groups if you have a low amount of clicks per ad group.
If your campaigns are huge you will run into the scenario that you need to built out ad formulas (needed with huge accounts) or if you have very few clicks, then you can test ad lines, headlines across the board.
Amazing tip for better results: Look for specific sentences (call us, book today, etc.) and test across the board.
Good Day One at The Hero Conference
I’m still excited about the Hero Conference, but I must say that I’m eagerly anticipating today’s schedule. With the likes of Oli Gardner, Brian Massey, Matt Van Wagner and Aaron Levy taking the stage, we’re up for another great day.
Now off to Bing’s Networking Reception to see who can actually play ping-pong!