Product and brand as one system: what eight years at Plume taught me
Customers feel the seam between brand and product long before they can name it. The ad promises calm and the setup flow feels like a router admin page. The box looks like jewelry and the support email looks like an invoice. Nobody files a ticket about it. They just trust the company a little less.
I spent eight years at Plume (2016–2024), leading brand and then design, and Head of Design from 2021 to 2024. Across that stretch the company went from Series B to a unicorn valuation, and from a consumer mesh WiFi product to a platform used by service providers around the world. The lesson I keep coming back to is simple to say and hard to run: product and brand only stay coherent when one system, and ideally one owner, holds both.
The seam: where product and brand drift apart
Drift rarely starts as a decision. It starts as a reasonable split. Marketing owns the launch, product owns the app, someone in operations owns the emails, and a partner team owns whatever the service provider sees. Each group does good work. The seams show up where their work meets:
- Onboarding versus advertising. The campaign sells a feeling, and the first five minutes in the app have to deliver it or the feeling turns into a refund.
- App versus packaging. A premium unboxing sets an expectation the app has to meet in the very next minute. If the hardware feels crafted and the app feels like a settings panel, the customer notices the drop even if they can’t say why.
- Consumer app versus operator portal. From 2018 the customer list included Communications Service Providers (CSPs and ISPs), not only homeowners. A portal that felt foreign to the consumer app would have told partners we were two companies stitched together.
The fix isn’t a brand police function. It’s making the same small set of decisions flow into every surface.
One house, many rooms
The hardest brand problem at Plume was audience, not aesthetics. A homeowner wants to hear that the WiFi just works. A service provider wants to hear about fewer support calls, network visibility and new revenue. Write two brands for that and you get two companies.
We used a "one house, many rooms" model. The house is the shared promise: effortless, adaptive, quietly intelligent. Each room is an audience with its own vocabulary and density. The consumer room is warm and spare. The partner room is precise and proof-heavy. The rooms can look different from the inside, but you never walk through a door and wonder whose house you’re in.
In practice that meant one set of messaging pillars, with tone adjusted by context rather than rewritten per team. When we moved Plume from "WiFi" toward a service people subscribe to, which was a long argument that ended in our biggest rebrand in 2020, the same pillars carried into the app, the partner kits and the membership story at once.
Shared parts, not shared PDFs
A brand guide is a document. A system is a set of parts people actually build with. The distinction matters because documents get skimmed and parts get reused.
We ran product and brand as one system through atomic design: the same building blocks of color, type, spacing and modules carried across the app, the web, email and brand work. We built those systems in Sketch and Zeplin before Figma took over collaboration, and moved them into Figma later. Today I’d go one step further and share semantic tokens: brand color and type defined once, consumed by the app, the marketing site and the operator dashboards, so a change ships everywhere as one reviewable diff instead of a dozen redlines.
The principle didn’t depend on the tool. If the brand team and the product team each keep their own source of truth for color, you have two brands within a year.
The small surfaces where the system proves itself
Big launches get attention. The system is tested on the surfaces nobody budgets for.
Email. Requests came at my team from product, marketing, legal and billing. Instead of four email styles, we built one atomic email system: modular, swappable components where a single message could change layout and even data depending on day or night, mobile or desktop. Email is a more direct channel of identity than the website for most customers, and treating it as product rather than admin was one of the better calls we made.
Plume IQ. For four to five years I designed a data insight graphic every month from our network of smart homes, working with a lead data scientist and our CSP marketing lead. It shipped on time every month. That cadence did two things: it trained the brand to speak in evidence, and it became the springboard for the much larger set of operator dashboards that followed.
Support and billing. Moving Plume from "WiFi" to a membership meant explaining, again and again, that "it’s not WiFi", it’s a smart home service that keeps getting more capable. That argument had to hold in the app, in billing messages and in the scripts support teams used, or the subscription felt like a fee for something people thought they already owned. Consistency there did more for the brand than any campaign.
Moving from leading brand to Head of Design
I led brand first and became Head of Design in 2021, with a team that grew to about nine designers, supported by freelancers and agencies, covering consumer product, operator tools, email and the brand itself. People ask what changed. Less than you’d think, and more than the title suggests.
What changed:
- The ratio of making to deciding. In the early years, with two or three internal designers and agencies as needed, I designed a large share of the work myself. As Head of Design, the job tilted toward setting defaults, reviewing, and making the calls that unblock a team.
- The audience for my arguments. Brand arguments land with marketing and leadership. Product arguments land with engineering and product managers, who care about scope, state and cost. I had to get fluent in both.
- Where quality gets defended. In brand, quality lives in the craft of each asset. In product, it lives in the system and in what the code base makes easy or hard.
What didn’t change: the belief that a customer meets the logo, the screen and the sentence as a single thing. The title moved. The scope of what I considered "my problem" was already the whole thing.
A checklist for teams with split brand and product orgs
You don’t need to reorganize to get most of the benefit. Start here:
- Name one owner for the seam. Not for brand or product, for the handoff between them. If nobody owns it, it drifts.
- Write pillars once. One promise, a tone-by-context table, and a short list of words you never use. Both teams inherit it.
- Share the parts. One atomic system, and today one token set, for color, type and spacing that the app, the site and email all consume.
- Audit the small surfaces. Pull ten real emails, receipts, error states and partner slides. Score them against the pillars. The gaps you find are the brand.
- Review across the line. Product reviews include someone from brand, and campaign reviews include someone who knows the first-run flow.
- Treat partner and operator surfaces as brand. If you sell to businesses as well as consumers, the dashboard is the brand for that buyer.
When I look at what held up at Plume through a rebrand, a platform pivot and years of venture pace, it wasn’t any single asset. It was the decision to run brand and product as one system. More detail on the work itself is in the Plume brand, Plume app and Plume web dashboards case studies, and the growth side of the story is in What I’d Build Earlier.
Related Designing data-dense dashboards: role-based views that surface the next actionConnected hardware: designing the box, the app and the brand as one launchDeep-tech brand and product: making an invisible product visible