CRO Conversion Rate Optimization PDF

CRO: Conversion Rate Optimization CRO Is the process of increasing the percentage of conversions from a website or mobile app. CRO typically involves generating ideas for elements on your site or app that can be improved and then validating those hypotheses through different types of testi...

CRO: Conversion Rate Optimization CRO Is the process of increasing the percentage of conversions from a website or mobile app. CRO typically involves generating ideas for elements on your site or app that can be improved and then validating those hypotheses through different types of testing such as A/B testing and multivariate testing. Digital Marketing - Alba Ramírez Pagès  PERSONALIZATION CAPABILITY Personalized marketing Personalized experiences (Generar demanda) 3 (Compras y Servicios) SEO / SEM Onsite Search Personalization Programmatic Personalized Recommendations & Targeting Email Opt imiza ti on Personalized Co nt e nt La ndi ng Page Opt imiza ti on Design Opt imiz a ti on Data DMP Custo me r Geno me Measurement MMM / At t ri b u ti o n Dig ital Anal ytics Omni-channel UNIVERSAL ID Enablement 8 Where? AWARENESS INTEREST DESIRE ACTION Source: Elias St. Elmo Lewis (1989)  DMP MAIN FUNCTIONALITIES WHAT IS A DMP? Copyright © 2020 Accenture. Todos los 6 Centralized technology derechos reservados. Información Confidencial platform that allows advertisers to create target audiences, combining 1st, 2nd and 3rd party data for subsequent activation through digital channels. 11  DMP MAIN FUNCTIONALITIES WHAT CAN WE DO WITH IT 7 1. Anonymize data to locate target 1. Store customer PII audiences and deliver a (Personal Identifiable personalized message or Information) experience. 2. Store campaign creatives 2. Segment by combining different 3. Activate touchpoint data sources (1st, 2nd and 3rd marketing party data). 4. Provide reports on 3. Use identity management tools to marketing performance resolve cross-device challenges. 4. Protect data by choosing which data is sent to which platform through the DMP. 11 IT ENABLES COMPANIES TO… Improve customer service Helps maintain a single holistic view of the customer between online and offline environments. Efficiency in media buying DMPs provide us with information on campaign performance with a global view. This allows us to identify opportunities for higher ROI. Cross-channel and cross-device personalization Thanks to the vision of people instead of devices, it allows us, through a deep understanding of the customer, to increase conversions. deep understanding of the customer, increase conversions and sales ratios. Improve analytics and segmentation By collecting information from all digital touchpoints in a single system, DMPs help us make better decisions. Cop yrig ht © 2 02 0 Accenture. Todos los derechos reservados. Info rmación Confidencial 8 14 HOW COMPANIES OPERATIONALIZE AN EXPERIMENT In consideration of the volume of requests for experiments from the various products/teams, we are going to share a template to evangelize and unify the format of, in this case, Test Briefs. This brief does not constitute a Test Plan in itself, but it is the first step towards the unification of the capacity and maturity between the different teams and a more precise way of communicating the actions that are proposed. As key points to highlight, it is composed of: Experiment sheet Description of the experiment Scope of the experiment Experiment proposal Risks and dependencies. Cop yrig ht © 2 02 0 Accenture. Todos los derechos reservados. Info rmación Confidencial 9  Guidelines for the CRO BRIEF 1. Experiment Sheet 3. Scope of the experiment Experiment Name: Submission Date: Objectives (primary and secondary): Formulation Place/Date: Primary Objective: Sector/Field/Product: Secondary Objectives: Team: 2. Experiment Description Details of the activity: Problem/Opportunity identified: Type of activity: Description of the problem or opportunity the experiment aims to Target audience: address. Metrics/Indicators of success: Target audience definition: Age, interests, habits, consumption barriers, key drivers for activating 4. Experiment proposal the identified problem or opportunity, or initial hypothesis. Control version (describe and quantify): Experiment hypothesis: Variation version (describe and quantify): Clear statement. 5. Risks and dependencies Describe possible risks and dependencies: Example of a CRO Test for Zalando Website Hypothesis: Changing the design of the product detail page (PDP) by adding a "low stock" urgency message will increase conversions by creating a sense of scarcity. 1. Problem/Opportunity Identified: Problem: Users often browse multiple products but delay purchasing, leading to cart abandonment. Opportunity: Introducing urgency cues could motivate users to complete purchases faster. 2. Target Audience Definition: Age: 18-35 Interests: Online shopping, fashion, convenience. Habits: Frequently compare multiple products, indecisive during checkout. Drivers: High-quality visuals, detailed product descriptions, and limited-time offers. 3. Scope of the Experiment: Objectives: Primary: Increase the conversion rate on PDPs. Secondary: Increase add-to-cart rate and reduce cart abandonment. Details of the Activity: Type of Activity: A/B test on the product detail page. Target Audience: All website visitors viewing a specific product category (e.g., women's shoes). Metrics/Indicators of Success: Conversion rate, click-through rate to checkout, and average time on PDP. 4. Experiment Proposal: Version A (Control): Standard product detail page without any urgency messages. Version B (Variation): Product detail page with a “Hurry! Only 3 left in stock” message displayed prominently under the price and next to the "Add to Cart" button. 5. Risks and Dependencies: Risks: Overuse of urgency messages may reduce credibility if users frequently see "low stock" notices. Risk of technical issues with dynamic inventory messaging. Dependencies: Real-time inventory data integration to ensure accurate messaging. Proper segmentation to avoid showing urgency messages for out-of-stock products. EXPERITMENT DESCRIPTION TIPs: Problem/opportunity encountered. The more we understand the problem the easier it will be to find the solution, so the more defined the problem the better. To do this we can answer the following questions: What is the problem? Why is it a problem? Who is affected? Where does it occur? When does it occur? Sources of analysis Qualitative Analysis/Quantitative Analysis Hy pothesis of the expe riment It is important to defin e in the best possi ble wa y the hypothesis of the experime nt, and for this reaso n we recommend that you follo w the follo wing structu re: “We bel ieve that if xxxxx, [wil l] xxxxxx, [gene rating ] xxxxxx. If we are right, [we wi ll get] xxxx.” Cop yrig ht © 2 02 0 Accenture. Todos los derechos reservados. Info rmación Confidencial 12 WHEN DOES THE EXPERIMENT ENDS? To explain the statistical feasibility of the proposed experiments, we specifically analyze the statistical confidence to “predict” the future conversion value, which is the percentage that indicates the probability that the result of the experiment will hold when repeated. Is statistical confidence the only factor we should consider before declaring a winner? No, as a general rule we have to consider the following factors to declare a winner: Duration of the test is at least 3 weeks (better if 4). Scope of the minimum pre-calculated sample size (using different tools). Statistical confidence of at least 95%. IT I S P OSSIBLE TO CALCULATE FUTU RE CON VERSION STATISTICAL CON FIDEN CE IT I S N OT POSSI BLE TO CALCU LATE FUT URE CON VE RSION Cop yrig ht © 2 02 0 Accenture. Todos los derechos reservados. Info rmación Confidencial 13  A/B testing Multivariate testing 1. Audience/traf fic 2. fif fif 25 25 25 25 Swinging ty % ty % % % % % 3. a b a b c d Testing Hell Hell Hell Hell Hell Hell o o o o o o 4. 12% 36% 12% 16% 25% 36% Measurement Conversion Conversion Conversion Conversion Conversion Conversion Description A/B testing (also known as Split testing) is the Multivariate testing uses the same basic mechanism as A/B testing, but compares a larger number process of comparing two versions of a web of variables and reveals more information about how these variables interact with each other. As page, email, or other marketing asset and in an A/B test, traffic to a page is split between different versions of the design. measuring the difference in performance by balancing traffic.  Assets digital > Web / App / IoT Digital assets or the windows of interaction with customers in the digital space Web desktop/mobile app IoT McKinsey & Company 16  A/B and multivariate testing: optimization of conversion into digital assets What is it for? Evaluate website and App functionalities and optimize their design to improve conversion Measure the response of users and clients to new functionalities and changes in the website and App Optimize web and app design in real time to improve conversion Improve campaigns conversion rates Generate reports and insights to improve conversion on the web and App  Testing tools allow you to optimize assets based on data Real case: In the early days of launch, the Booking.com team realized that although many owners completed the first registration step, they got stuck on the next steps. At this time, the pages created for paid search from your native campaigns were used for the registration process. Both teams decided to work together and created three +25 versions of the landing page for Booking.com. Additional details such as ratings, awards, recognitions, user rewards, % CR etc. have been added. to the variations. The test lasted two weeks and produced a 25% increase in homeowner registration. The test results also showed a significant decrease in the cost of each registration. Optimizely report example A/B testing tools ensure the solidity of the results through the “statistical significance” that is achieved once the test has achieved a relevant volume of traffic. The best global practices direct THOUSANDS of tests simultaneously The big pure players in the digital market (booking, Amazon, Netflix...) launch, on a continuous basis, thousands of tests seeking to optimize internal rates of return and performance …they obsessively focus on A/B testing — today, they may be the most sophisticated A/B testing company in the world, with 1,000 experiments running at any time https://medium.com/@aaronglazer/how-booking-com-ab-tests-ten-novenonagintillion-versions-of-its-site-25fc3a9e875b Where? AWARENESS INTEREST DESIRE ACTION Source: Elias St. Elmo Lewis (1989) FUNNEL 100.000 IMPRESSIONS 2.000 LANDING VIEWS 500 CLICKS ON THE LANDING 200 CLICKS ON OTHER PARTS 200 ADD TO THE CART 50 SALES Digital Marketing - Alba Ramírez Pagès Digital Marketing - Alba Ramírez Pagès  Generate Digital Marketing - Alba Ramírez Pagès AWARENESS VIDEO AD SEO ADWORDS AD Digital Marketing - Alba Ramírez Pagès INTEREST Digital Marketing - Alba Ramírez Pagès CONSIDERATION Digital Marketing - Alba Ramírez Pagès PURCHASING FUNNEL - INTENT Digital Marketing - Alba Ramírez Pagès PURCHASING FUNNEL - INTENT Digital Marketing - Alba Ramírez Pagès PURCHASING FUNNEL Digital Marketing - Alba Ramírez Pagès Digital Marketing - Alba Ramírez Pagès FUNNEL 100.000 IMPRESSIONS 2.000 LANDING VIEWS 500 CLICKS ON THE LANDING 200 CLICKS ON OTHER PARTS 100 ADD TO THE CART 25 SALES Digital Marketing - Alba Ramírez Pagès optimization Digital Marketing - Alba Ramírez Pagès Digital Marketing - Alba Ramírez Pagès Digital Marketing - Alba Ramírez Pagès  Principles of persuation 1. Reciprocity 4. Shortage - Campaigns at "key moments": January, Black Friday, back to What can we give? Advice, immediacy, security, experience? school... - Incentive in exchange for discounts or promotions. - Commercial messages reinforcing "Today", "Now", - Advice at any time "Already"... 2. Consistency 5. Sympathy - Personalize retargeting messages - Opinator: should we add one? - Popup at process "abandonment" time - Chatbot: do we need it? - Messages during the process to encourage to move to the next - Chat or/and videoconference with a Specialist step - FAQS - Personalized messages for customers who have already taken out loans in the past. 6. Authority - Communicating years of experience? 3. Social Approval - Confidence of being a leader? - The loan most taken out by our clients - Recognizable people / government / entities / celebrities - So many people have taken out this product... - Testimonials - More than XXX loans approved among our clients - Play with approval rate What is left to finish?  1. Go to booking.com and identify as many persuasion elements as possible and categorize them 2. Create a brief to run a test on booking.com according to your criteria. You can create any scenario you consider necessary.