Four Surfaces Is Not Four Products
RevSprint runs on four distinct surfaces. A browser application for deep work. A desktop overlay called RIBA that sits on top of whatever you're already using. A mobile app for decisions on the move. And Orbit, a customer-facing intelligence surface deployed on your sites.
The obvious approach is to build four products, with four backends, four data layers, and four parallel teams maintaining four sets of business logic that drift out of sync the moment the roadmap diverges. Plenty of companies have shipped this shape and the result is always visible at the seams: the mobile app trailing the web app by a couple of quarters, the desktop client carrying features the browser does not, the customer-facing widget feeling like a different product by a different vendor that happens to share the logo.
We rejected that approach on day one. RevSprint has one intelligence layer, and every surface is a rendering target for the same underlying brain. A signal that surfaces as a desktop notification through RIBA appears at the same moment in the browser dashboard, on the user's phone, and in the way Orbit speaks to the affected customer on your website.
“The moment you let surfaces diverge in their intelligence, you've built four products wearing a trench coat pretending to be one platform. Users notice. They just can't always articulate why the experience feels inconsistent.”
The Render Adaptation Problem
The hard part is not sharing the intelligence between surfaces; it is adapting the output to surfaces with very different constraints on attention, real estate, and disclosure. A desktop overlay has to live above the user's actual work without obstructing it, a browser application can spread out across a full viewport, a mobile screen is small but has the user's full attention for the eight seconds they hold it, and Orbit faces external users who must never see anything the data boundary should not allow them to see.
- Desktop overlay: intelligence must be concise, actionable, and dismissible. You can't show a full dashboard on top of someone's email client.
- Browser app: intelligence can be expansive. Full context, deep drill-downs, historical patterns. This is where analysis lives.
- Mobile: intelligence must be decisive. Push notifications that surface the one thing you need to know. Quick actions you can take with a thumb.
- Orbit (customer-facing): intelligence must respect data boundaries absolutely. Internal deal health, support sentiment, team conversations, none of that can leak to external users.
The engineering challenge is building a render adaptation layer that takes the same intelligence payload and transforms it appropriately for each surface. Not dumbing it down for mobile or stripping it for the overlay, but genuinely adapting the presentation to match what each surface does well.
This means the intelligence layer cannot produce fixed-format output. It produces structured intelligence with priority ordering, action suggestions, and a context depth that each surface can sample at its own granularity. The overlay surfaces the headline and the single top action a user can take in two seconds. The browser pulls the full depth for the work that needs spreading out. Mobile takes the notification summary and a deep link back to whichever surface can handle the next step. Orbit takes only the slice the data boundary allows it to receive in the first place.
Surface Tagging and Trust Progression
Every interaction with RIBA is tagged with the surface it originated from. This isn't just analytics. It drives trust progression. If a user primarily interacts through the desktop overlay, RIBA adapts its communication style for that surface. If they shift to mobile during evenings, the intelligence adapts again.
Surface tagging also feeds the progressive autonomy model. RIBA can be granted different authority levels on different surfaces. Perhaps you trust it to auto-schedule follow-ups from the browser but want it to ask for confirmation on mobile. The intelligence is the same. The autonomy envelope is surface-specific.
This is what it means to build a Symbiotic Intelligent Operating System that meets you where you are. Not four apps that happen to share a brand. One intelligence layer that adapts to whatever surface you're using, with the full organisational context behind it every time.
We frame the same idea from the product side in One Product, Four Surfaces, and Nielsen Norman Group's research on cross-surface and context-adaptive interfaces consistently finds that surface-adapted intelligence outperforms one-size-fits-all interfaces. To see the same brain on every surface you actually work from, get early access.
