The Psychology of a Website: Optimize for Cognitive Biases, Conversion Triggers, and Google’s RankBrain, Matthew Capala, Alphametic
In this session, Matthew shared recent research regarding the impact of organic search traffic (a primary source for a majority of websites) and key ranking factors. Those ranking factors have evolved from keyword-based to performance-based metrics (direct site traffic, average pages-per-session, time on site, bounce rates, etc.). As a result, web developers and marketers need to think about performance factors more than ever.
He then talked about the importance of Google’s RankBrain, which is the new machine learning algorithm. Since Google owns Chrome, Android and properties that generate tons of data, they can analyze to determine human emotions, which is a relatively new concept. Compounding the issue is a significant attention deficit, cognitive biases and other challenges to consider. Matthew dove down into more actionable insights, like website shortcuts that leverage automatic behaviors and conversion triggers. He recommended two excellent authors, Robert Cialdini and Dain Ariely, who explain human behaviors that can inform design.
One of the first conversion optimization tactic is the zero-risk bias (free trials). The bias means we’re more likely to avoid risk or loss than seek gain. We also get desensitized when over-exposed to risk and changes perception of risk. One example of zero-risk bias is free trials, yet we tend not to want to provide a credit card. Interestingly, conversion and retention rates are higher for non-credit card required trials (according to a Totango study).
Other examples of conversion optimization tactics include:
• Zero risk bias: money-back guarantees are self-explanatory
• Reciprocity: free samples are a “gift” and triggers a sense of obligation
• Loss aversion bias: humans are more sensitive to the prospect of loss than of potential gain. This can be addressed by the ownership trigger with limited time, exclusivity or other FOMO messaging.
• Anchoring bias: latching onto key concepts and launching into new concepts. The most common tactic is showing a premium product or price point, to make the mid-level price point attractive (or vice versa).
• Framing bias: perceptions of the same thing will change based on the context in which it’s perceived. He used a funeral insurance example, which included positive and negative framing (likely to cover both bases).
• Social proof: the bandwagon effect creates a sense of comfort.
• The authority principle: people tend to look to experts for validation (celebrity or industry guru endorsements, etc.)
• In-group favoritism: this is where cult-brands leverage the community following. Examples include Harley-Davidson and Apple, but even small business can create a similar sense.
The outcome of a choice is predetermined by what we put attention on before making a decision. What’s focal is often perceived as causal. As a result, consumers need options, since price is not absolute. In a newspaper subscription example used a decoy to change decisions and increase purchase of a higher price point (web + print option).
Matthew closed his presentation with a few parting thoughts:
• Leverage in-page data (heatmaps, scrollmaps, confettis, click recordings (CrazyEgg, etc.)
• Always be testing: A/B or multi-variate to maximize conversion rates
• He recommends Ghostery and the Wayback Machine to analyze websites
Fyre Festival: 10 Marketing Lessons Your Business Can Leverage, Brian Fanzo, iSocialFanz
If you haven’t seen the documentaries about Fyre Festival, you should. In this session, Brian outlines key lessons from the Fyre Festival debacle. Here are a few of the lessons/mistakes made.
1. Millennials will pay for experiences, so Fyre had the right idea: create a memorable, unique and exclusive experience. The Fyre Festival promotional video delivered in spades on this front.
2. Unfortunately, the event creators scheduled the festival 6 months out, on an island without any real infrastructure, which was largely the undoing of the event. Lesson: give yourself time and hire experts.
3. Fyre Festival expertly leveraged social media to generate buzz and sell tickets. According to research, 64 percent of Millennials trust a YouTube Creator more than a friend at school. As a result, the Festival leveraged influencers and created attention-getting content and campaigns (like replacing content with the orange title and link to the Fyre Festival account and URL.
4. Unfortunately, the event did not live up to expectations by a long-shot and social media blew up the Festival and the organizers reacted poorly by trying to delete comments, etc. So what built their brand ultimately took them down.
5. KISS: The Fyre team made it simple for the infuencers: gave them an orange color tile image and a link to post at a specific date to promote the Festival. Easy and effective. The influencers created additional content outside of the originals scope, as they were provided a unique experience (filming the promotional video on the island).
6. Emotional Marketing = Good – Bad – Ugly. The fear of missing out (FOMO) is a key driver for Millennials (and other generations) these days. The Festival tapped this trend.
7. Tell a Story: document the experience with video. Video has higher recall and engagement than images and text. He showed a Shutterstock parody video utilizing only stock footage to recreate the Fyre Festival promotional video.
8. Digital Empathy: Brian left us with the idea of connecting with a demographic and evoking emotion through social media. He reminded us that the local businesses negatively impacted by the fraudulent festival raised $220K for the Exuma Point restaurant thanks to a Netflix GoFundMe campaign. Create an experience that delivers on expectations: learn from Fyre Festival.
Sound, Search and Semantics, Upasna Gautam, Ziff Davis
In her presentation, Upasna outlined the evolution of search, from keyword-focused to semantic-centric. Effectively, SEO no longer stands for search engine optimization, it stands for search experience optimization. She outlined various elements of the new SEO:
• Entity Optimization: understanding the intent of the query (including locations, people, abbreviations, synonyms and context). Examples include GM (general motors or genetically modified).
• Knowledge Graph: the ecosystem of relationships between words and where intelligence is stored to understand the meaning of various entities. This typically powers the boxes on the right side, outside of local business listings.
• Structured Data: the language the helps Google understand relevance and authority and is how entities are indexed. This is particularly important when it comes voice search. Speakable Schema is here and will grow rapidly in the near future.
• Information Architecture: this is the most important component, especially when factoring in rankings. The hierarchal structure of the site provides essential context for the content on the site. A good architecture helps humans AND crawlers. She recommends reading Information Architecture for the Web and Beyond by Peter Morville.
• Co-occurrence and Clustering: The two components include topics generated by word frequency and weighted biograph clustering, which is more powerful than latent semantic indexing (LSI).
Upasna took moment to explain some of the more complex concepts. Learning mathematical relevance helps us understand search on a functional level. LSI uses singular value decomposition (SVD) to retrieve text on the web and understands how terms and concepts are related. These multi-dimensional structure projects down into a smaller number of dimensions. The benefit of LSI is to reduce noise, reveal similarities that were latest and similar terms become more similar while dissimilar things remain distinct. She outlined the challenges with short texts, such as search queries and tweets, which do not provide sufficient context.
The next section of the presentation was one of my favorite topics: voice search optimization. She pointed out that optimizing for search results is the same as optimizing for voice search, for the most part. At the core of voice search is automatic speech recognition (ASR), fueled by a deep learning AI engine. According to Upasna, ASR is the form behind the voice search function.
For Google’s voice search efforts, they developed quality metrics. They use the metrics to drive research and improve the search experience. Those voice search metrics are:
• Word Error Rate (WER)
• Semantic Quality (Webscore)
• Perplexity (PPL)
• Out-of-Vocabulary Rate (OOV)
As a result, the search engine results page (SERP) has evolved into a dynamic, purchase-driven environment with the integration of products, carousels, featured snippets with product rankings, research and shopping carousels. Note that Google believes a high-quality user experience is a fast user experience. In effect, Google will prioritize speed over quality, according to their own published guidelines. Parting thoughts from the presentation include:
• Craft and optimize content for topics and concepts, not just keywords
• Use structured data to help search engines understand context
• Align the information architecture of your website to the customer journey (navigation, sitemaps, page structure)
• Invest in speed optimization
• Provide answers to specific questions about your products and services (featured snippets)
The Era of the Social Brand Troll: How We Got Here and What You Should Do, Obele Brown West, Weber Shandwick
Obele shared examples of “brand snark” and trolling by brands (like Wendys, Oreo and Verizon). While companies have built on value-based messaging, times are changing. As channels expand and messages become fragmented, it’s more difficult than ever to break through. Note: this session was the third I attended with the goldfish attention span metaphor, due to increased smart phone usage. According to AdWeek, brands need to be younger humans, not just human. Thus the evolution of snark and trolling by brands. Even the US Air Force got in on the Laurel & Yanny conversation, in reference to an A10 post, for which they later apologized. According to research, brands that “troll” have been successful. Obele cited the KFC example of following 11 Herbs & Spices. Good reminders to be authentic, yet engaging. Be out there and be memorable.
Kent Lewis is a co-founder and the first President of SEMpdx, a professional trade organization based in Portland, Oregon that provides educational opportunities for local businesses and individuals interested in a career in search engine marketing. To pay the bills, he plays the role of President of Anvil Media, Inc., where he is responsible for managing operations, marketing and business development to achieve the company’s mission: to build Anvil into the one of the most respected search engine marketing agencies in the world. In 2001, Lewis created pdxMindShare to build businesses and careers through networking and weekly emails featuring jobs and industry events. He is also an adjunct professor at Portland State University.