--- title: "The synthesis nobody owns: getting from brand strategy and research to actual components" url: https://www.axelerant.com/blog/brand-research-synthesis-to-components published: 2026-08-26T07:26:03.215Z author: "Dheeraj Khindri, Director of Experience Design" lenses: experience source: Axelerant Thinking --- # The synthesis nobody owns: getting from brand strategy and research to actual components > Brand strategy and UX research do not tell you what a component should do. The synthesis step between them is unowned work, and skipping it has a cost. A redesign clears every review. The brand team confirms it expresses the new positioning. The research team confirms it addresses what the interviews surfaced. Then it goes into usability testing and people cannot find things, hesitate at the point of commitment, and describe the site as beautiful in the tone that means something else. Nobody was wrong. Nobody was even careless. What happened is that two documents arrived and neither one contained an instruction. Brand strategy produces a positioning: what the company stands for, what it must never sound like, where it sits against its competitors. Research produces findings: what people were trying to do, where they gave up, what language they used. Both are correct and neither one tells you what a component should do. Between them sits a translation step, and because it belongs to neither team it usually does not get produced as work. Insight goes in one end, an interface comes out the other, and the component library quietly encodes whichever opinion had the most seniority behind it in the room. That gap has a cost that only shows up later. A component library is not a set of styles. Every component is a frozen decision about what a person is permitted to do at that moment and how they will be told what happened. Ship the library and you have industrialized those decisions across every page and every future page. A design system scales whatever judgment went into it, including the judgment nobody made deliberately. ## Two layers, and only one of them is free The most useful thing we bring into a synthesis session is a distinction, and it resolves more arguments than any amount of workshop facilitation. Distinctiveness has two places it can live. The expressive layer is typography, color, spacing rhythm, photographic direction, motion character, editorial voice, density. The interaction layer is how navigation behaves, how filtering works, what a form control does, what happens on scroll, how state is communicated. The expressive layer will carry almost unlimited distinctiveness at no usability cost. A brand can look like nothing else in its category and still be effortless to use, and the strongest premium experiences generally are. The interaction layer is different, because distinctiveness there is not free. It is paid for out of conventions the user learned somewhere else. Jakob Nielsen's formulation of this is that people spend most of their time on other websites, so they expect yours to work the way those do. Every departure spends a little of that inherited fluency. Sometimes that purchase is worth making. It is almost never worth making by accident, and it is frequently made by accident, because a brand team asking for something that feels ownable and a design team obliging is one of the most reasonable-sounding conversations in the business. So the synthesis question, asked per component, is simply: which layer is carrying the difference here, and if it is the interaction layer, what convention are we spending and why. ## Why navigation is where this collides hardest Navigation is the component where the two inputs disagree most directly, and it is usually resolved by hierarchy rather than by reasoning. Brand wants navigation to express the story: the portfolio, the tiers, the architecture, the way the organization understands its own offer. Research almost always says users navigate by the job they arrived to do, which rarely matches the org chart or the product hierarchy. Both positions are legitimate. They are also incompatible at the level of a top-level menu, which has maybe six slots and no room for a compromise that satisfies both. Resolved by seniority, this produces a menu that mirrors the internal structure with a research-flavored label or two applied on top. Resolved by synthesis, somebody writes down which model the primary navigation serves, which model gets served elsewhere, and what the reason was. The written reason matters more than it sounds. It is what survives the next stakeholder, and it is the difference between a navigation decision and a navigation opinion. There is a related failure worth naming: navigation labels are also the language a retrieval engine reads. Reorganizing a menu around internal vocabulary is a discoverability decision as much as a usability one, which is a thread we pulled in more detail in [how universities stay visible in AI-first search](/blog/universities-stay-visible-in-ai-first-search). The same collision shows up across a multi-location estate, where templated pages carry brand authority and no local signal at all: [national rankings do not trickle down](/blog/national-rankings-local-visibility-multi-location). ## What the synthesis deliverable actually looks like It is smaller than people expect and less impressive than a strategy deck. For each significant component: what job it serves, drawn from research. What the brand needs it to express, drawn from positioning. Which layer carries the distinctiveness. What convention it honors or spends. What the alternative was and why it lost. A page or two, in a table, arguable by anyone who reads it. That last property is the point. A component library with no reasoning attached cannot be argued with, so it gets overridden instead. One with the reasoning attached can be challenged, defended, or deliberately changed, which is the only version that survives contact with a real organization over three years. ## The tradeoff This adds a stage. It produces a document that looks like overhead to anyone who did not sit in the session. And it forces brand and product to have their disagreement early, in a room, on the record, instead of late in a design review where it is easier to defer. Some organizations would genuinely rather have the late argument. It feels faster and it protects relationships. It is worth being honest that this is a real preference and not simply a mistake, because the alternative is pretending that a process change is free. ## The test for your own system Take the three most distinctive things about your current interface. For each one, decide which layer it lives in. If a distinctive thing turns out to be riding on an interaction pattern, name the convention it replaced and find out who decided that. If nobody can remember deciding, you have found something the system is enforcing that nobody chose. ## Frequently asked ### What is design synthesis, exactly? Synthesis is the translation step between brand strategy and user research on one side and component decisions on the other. It produces a per-component record of what job the component serves, what the brand needs it to express, which layer carries the distinctiveness, and what the rejected alternative was. It is a short document, not a deck. ### Why do design systems drift after launch? Usually because the reasoning was never written down. A component library with no attached rationale cannot be argued with, so teams override it instead of challenging it. One with the reasoning attached can be defended or deliberately changed, which is what lets it survive several years and several stakeholders. ### Can a brand be distinctive without hurting usability? Yes, if the distinctiveness sits in the expressive layer: typography, color, spacing, motion character, photography, editorial voice. Distinctiveness in the interaction layer is different, because it is paid for out of conventions users learned elsewhere. That purchase is sometimes worth making and should never be made accidentally. ### Should navigation follow the brand architecture or the user's mental model? They are usually incompatible in a menu with six slots, so the answer is to decide explicitly and record why, rather than let seniority settle it. Name which model the primary navigation serves, where the other model gets served instead, and what the reason was. The written reason is what survives the next stakeholder. ### Who should own the synthesis? It has no natural owner, which is why it goes missing. In practice it works when one person is made accountable for producing it and both the brand and research owners are required to sign it, so the disagreement happens once, early, on the record. ## Talk it through If you are heading into a redesign with a positioning document in one hand and research findings in the other, we are glad to talk through how the two get reconciled before anyone opens a design tool. We run this synthesis as a defined piece of work rather than an assumption. If that sounds useful for what you are planning, [get in touch](/contact). --- Read on the web: https://www.axelerant.com/blog/brand-research-synthesis-to-components