I've had this conversation a hundred times, and the answer is never the same. Build a web app or build a native app - reach everyone instantly through a browser, or build for a platform and get something better at a higher cost and a slower timeline. It's an old argument. It never really gets settled, it just shows up again every time a new medium forces someone to make the call.
Right now that medium is AR. If your brief includes a filter, a try-on, or anything where someone points their phone and something happens, you're about to make this same call. Even if nobody frames it that way.
Web usually wins here. Build it once, it runs on basically any phone. App-based, native tools, or built inside Snapchat or TikTok, usually costs more, because you're either building for one ecosystem specifically or adapting for several.
Don't take "web" as shorthand for "cheap," though. In both cases, the thing that actually eats your budget usually isn't the platform, it's the content. Getting a 3D product or character to look right takes more time and money than the build itself, most of the time. If a quote only covers technical build and skips content production, ask what's missing.
Web ships the moment it's done. No app store review, no waiting for someone else to approve your release. That matters if you're tied to an event date or a media buy that can't move. App-based work adds a review step you don't control. Usually not a long one, but it's not yours, and if your date is fixed, that's a risk.
This is the one that bites people later, usually in a room with a client. Native AR generally looks and feels better. It's steadier, more responsive, with better access to a phone's sensors. Web has closed a lot of that gap and is genuinely fine for most of what brands actually need, but for the demanding stuff, like precise object placement, longer immersive sessions, native still has the edge.
The real question isn't which one's better. It's whether the gap matters for what you're building. A quick shareable filter doesn't need the polish a virtual showroom does. Match the format to the job, not to whichever one sounds more impressive in the deck.
Here's what actually surprises people, and it's not the technical trade-off. You're betting the platform still exists when you need to update the campaign next year.
Two examples, both recent. Meta Spark, the tool behind Instagram and Facebook filters, shut down in January 2025. Anyone who'd built their AR strategy around Instagram lost that channel overnight, no warning baked into their plans. More recently, 8th Wall, one of the most established browser-based AR platforms, announced it's winding down too. No new projects since February 2026. Everything still live goes offline for good in February 2027.
Nobody using those platforms did anything wrong. The platforms just made a call that had nothing to do with any single brand's campaign.
I'm not going to tell you to avoid web AR or avoid app-based AR. Both are still solid choices, and everything above still applies. What I'd add is: ask about platform risk with the same seriousness you'd ask about cost and timeline. Before you sign off on anything, ask whoever's building it, us included, what happens to the experience if the platform underneath it shuts down, and how much of the work you keep if you need to move it somewhere else.
That question costs you nothing to ask now. But not asking it can cost you a campaign a year from now. It's also a question I think any AR partner worth working with should be able to answer straight.
Not sure whether your next campaign should be web or app-based? Let's talk it through.
