Alex Harris

What I used it for Web and feature experimentation, personalization, customer data through ODP, CMS experiences, and AI-assisted workflows with Optimizely’s Agent Platform.

The problem Organizations often had the platform but needed a better system around it: stronger prioritization, better measurement, faster execution, more meaningful personalization, and a way to carry learning from one experiment into the next.

My role I’ve helped shape and sell programs, define experimentation and personalization strategy, build roadmaps and test plans, work across design, development and analytics, review measurement and results, and establish the operating model that keeps the program moving.

What it taught me The technology can make experimentation possible, but the value comes from the questions, the evidence, the measurement and what the team does with the answer. AI is making execution faster, which makes judgment and guardrails even more important.

Related experience Enterprise ecommerce, healthcare, financial services, and membership organizations.

Credential Certified in Optimizely

What I used it for A/B and multivariate testing, experience targeting, and personalization within Adobe-based marketing and experience stacks, with measurement connected to Adobe Analytics where appropriate.

The problem Mature organizations often had strong Adobe technology but experimentation, analytics and business measurement were not always operating as one system. That made results harder to trust, prioritize and turn into the next decision.

My role I shaped experimentation strategy and roadmaps, defined hypotheses, audiences and success metrics, worked with developers and analysts on implementation and measurement, and turned results into decisions about what to change next.

What it taught me A test is only as trustworthy as the measurement behind it. Define what success means before the experiment launches, not after the results come in.

Related experience Enterprise healthcare, financial services, and ecommerce organizations using Adobe-based experience stacks.

What I used it for Onsite personalization, product recommendations, audience targeting, and testing personalized experiences in ecommerce.

The problem Personalization programs could launch a lot of experiences without proving which ones actually performed better than a strong default. That spread effort thin and made it harder to know what should scale.

My role I shaped personalization strategy and use cases, designed audiences and experience variations, defined how lift would be measured against a control, and worked with merchandising, design and development teams to decide what to scale, change or retire.

What it taught me If personalization can’t beat a strong default experience, it’s adding complexity, not value.

Related experience Ecommerce, retail, and direct-to-consumer brands.

What I used it for Session replay, behavioral friction analysis, funnel/context review, and understanding why customers were struggling in key digital journeys.

The problem Analytics could show where conversion dropped, but not always why. I used behavioral evidence to understand the friction behind the numbers and turn it into better experiment, personalization and journey ideas.

My role I used Quantum Metric as part of the research and optimization process: identifying problem areas, shaping hypotheses, connecting findings to the roadmap, and working with analysts and delivery teams to decide what should change next.

What it taught me Analytics tells you where the problem is. Watching the experience shows you what the customer is actually dealing with. The best decisions come from using both.

Related experience Streaming and subscription, ecommerce, healthcare, financial services, and mobile experiences.

What I used it for Measurement planning, event and conversion setup reviews, funnel and path analysis, audience insights, and reading experiment and campaign results.

The problem Teams often had plenty of data in GA4, but the reports still did not answer the questions the business actually needed to answer. Decisions then fell back to opinion or to whichever number was easiest to pull.

My role I started with the business questions, defined the measurement plan and key conversions around them, reviewed tracking with analysts and developers, and turned the analysis into test ideas, priorities and readouts leaders could act on.

What it taught me Decide what you need to know before you decide what to track. Most analytics problems start as question problems.

Related experience Ecommerce, lead generation, membership, content, and enterprise digital programs.

What I used it for Enterprise reporting and segmentation, journey and conversion analysis, experiment readouts alongside Adobe Target, and finding opportunities for testing and personalization.

The problem Large organizations often had multiple teams using the same Adobe data but different definitions of success. Meetings could turn into debates about whose number was right instead of what the business should do next.

My role I worked with analytics, marketing and product teams to agree on the metrics and segments that mattered, used the analysis to find and prioritize opportunities, and made sure results were reported in a way leaders could trust and act on.

What it taught me The hardest part of enterprise analytics is often not getting the data. It’s getting everyone to agree on what the number means.

Related experience Enterprise healthcare, financial services, and ecommerce organizations.

What I used it for Customer-data collection and activation, event and attribute planning, audience signals, and connecting behavioral data to analytics, experimentation and personalization platforms.

The problem Customer behavior was being captured differently across platforms, which made audiences, analytics and personalization harder to trust. The challenge was creating a dependable data foundation that multiple systems could use consistently.

My role I helped define the events and customer attributes that mattered, how they should be structured and named, which systems needed them, and how that data would support analytics, experimentation, personalization and lifecycle use cases.

What it taught me The value of a data layer isn’t collecting more events. It’s giving every downstream system the same reliable understanding of the customer.

Related experience Ecommerce, B2B/SaaS, membership and subscription, and enterprise digital programs.

What I used it for Audience and identity strategy, segmentation, activation planning, and connecting customer data to personalization, campaigns and next-step experiences.

The problem Organizations often had customer data spread across systems, or even a CDP in place, but that data was not changing the experience when it mattered. The challenge was turning unified data into something marketing and experience teams could actually act on.

My role I shaped the use cases, defined which customer and account signals mattered, designed audience and activation strategies, and connected the data layer to personalization, lifecycle marketing, ABM and digital experiences.

What it taught me Unified data has no value by itself. It becomes valuable when it changes a decision, an experience, or what happens next for the customer.

Related experience Enterprise B2B, financial services, ecommerce, and membership/lifecycle organizations.

What I used it for Designing, building and optimizing more than 50 Shopify sites, including themes, product and collection pages, cart and checkout flows, analytics, and experimentation around the parts of the experience that mattered most to conversion.

The problem Many stores looked good but did not make buying easy enough. Customers could reach a product page and still be unclear about the value, the right option, what would happen next, or whether they could trust the purchase.

My role I was hands-on across research, design, build and optimization, then measured what changed and kept improving the pages that mattered most to revenue. On larger projects, I also worked with and directed designers and developers.

What it taught me In ecommerce, clarity usually beats cleverness. If the product, value, next step or reassurance is unclear, conversion suffers no matter how polished the store looks.

Related experience Direct-to-consumer brands, specialty retail, and small and mid-size ecommerce businesses.

What I used it for AI prototyping, research synthesis, content operations, experimentation workflows, agent orchestration, and connecting analytics, search, browser testing, CMS content and existing knowledge through MCP.

The problem The information needed to make a good decision usually lives in different places: analytics, research, past experiments, content, documentation and individual people’s knowledge. I wanted AI to help connect that context around the problem being solved instead of treating every task as a new conversation.

My role I design and build these systems hands-on: connecting sources, creating workflows and agents, defining what AI is allowed to trust or change, establishing human approval points, testing the outputs, and building the guardrails around them.

What it taught me AI will accelerate a broken process just as readily as a strong one. The useful part is building a system that learns from failure: when something goes wrong, turn it into a rule, check or guardrail so the same mistake is harder to make twice.

Related experience Marketing operations, experimentation, personalization, content operations, analytics and research, SEO and AI search visibility, prototyping, and workflow automation.

Example One content workflow connects research and source material to AI-assisted drafting, then checks factual claims, approved proof, voice and publishing rules before anything can reach the site. Publishing remains drafts-first with human approval.