AR Marketing App Development: Choosing a Format That Pays Off

This guide walks through each.

What an AR marketing app is

An AR marketing app uses the phone camera to place digital content in the real world so a customer can see, try or explore a product before buying. The content can be a sofa in a living room, a pair of glasses on a face, or an animation that appears when someone scans your packaging.

On iPhone this is powered by Apple's ARKit, and on Android by Google's ARCore. Both detect floors, walls and faces and keep a 3D object steady as the phone moves. Browsers can now do a simpler version of the same thing, which matters for marketing because it removes the app download. For the engineering side, see our guides on AR app development and how to build an AR app.

WebAR, native app or social filter

Pick the format by asking how much friction your audience will accept. A one-off campaign cannot ask people to install an app. A tool your customers use again and again can.

Many brands combine two: a product page viewer for every shopper, and a richer AR feature inside their app for loyal customers. If you already have an app, adding AR to it is usually better than launching a separate one.

Use cases that earn their cost

AR earns its cost when it answers a question that stops a purchase: will it fit, will it suit me, how does it work. It rarely earns its cost as a novelty.

• Place in your room: furniture, appliances, lighting, art, flooring and outdoor equipment at true scale.
• Virtual try-on: glasses, makeup, jewellery, watches and footwear.
• Packaging and print activation: scan the label to see a recipe, a how-to or an offer.
• Trade shows and showrooms: show a machine or vehicle that is too large or costly to bring.
• Property and interiors: preview finishes and layouts.
• Guided setup: show the customer how to install or use what they bought, which also reduces returns and support calls.

A simple test before you commit: if a good photo or a 20-second video would do the same job, skip AR. Validate demand the same way you would for any product. Our app idea validation checklist can help.

3D assets are half the project

The quality of an AR experience depends more on the 3D models than on the code. A model that looks wrong in size, colour or material does more harm than no AR at all.

• Source: models can be converted from existing CAD files, built by a 3D artist from photos and dimensions, or captured by 3D scanning. CAD files are usually far too detailed for a phone and must be simplified.
• Formats: USDZ is used for Apple's AR viewer, and glTF or GLB for Android and the web. Plan to export both from one master file.
• File size: models must load fast on mobile data. That means careful polygon counts and compressed textures.
• Materials and lighting: fabric, metal and glass need realistic materials, and the app should match the room's lighting.
• Scale: true-to-size placement is the whole point for furniture and equipment. Test with a tape measure.
• Catalogue size: ten products is a project. A thousand products is a pipeline with its own process, naming rules and quality checks.

Ask for a content management approach that lets your team add or swap models without a new app release.

How to measure results

Decide what you will measure before you build, and wire the events in from the start. Useful measures include:

• How many people open the AR view, out of those who see the button.
• How many successfully place the object. Low numbers point to poor instructions or poor lighting guidance.
• Time spent and products viewed in AR.
• Add to cart and purchase rate for shoppers who used AR compared with those who did not.
• Return rate on products bought after an AR view.
• Scans per QR code location for packaging and print campaigns.

Be careful with the comparison. Shoppers who open AR are often more interested to begin with, so an A/B test gives a fairer answer than a simple before and after. We will not promise a lift, and you should be wary of anyone who does. See the mobile app analytics guide for event planning.

Privacy matters too. The camera permission prompt needs a clear reason, and face data used for try-on should be processed on the device and not stored unless you have a lawful reason and consent.

What drives the cost

Cost varies widely, so we quote after we know the format and the catalogue. The biggest drivers are:

1. Format: a product page viewer is the smallest, WebAR campaigns sit in the middle, and native AR inside an app is the largest.
2. Number and complexity of 3D models, and whether usable source files exist.
3. Tracking type: floor placement is simplest. Face, body, image and object tracking each add work.
4. Interactivity: colour and option pickers, animation, measuring, saving and sharing.
5. Commerce integration: linking to your catalogue, cart and stock.
6. Content management for models and campaigns.
7. Licences: some WebAR platforms charge ongoing fees, which you should budget separately from the build.
8. Device testing across older phones, since AR is demanding.

For a rough range on the app portion, try the app cost calculator.

Steps to a first AR campaign

Start small with a handful of products and one format, measure, then expand.

1. Pick the customer question AR will answer and the metric that proves it.
2. Choose three to ten products where fit or look is the main barrier.
3. Choose the format using the table above.
4. Produce and approve the 3D models first. Check size, colour and materials against the real product.
5. Build a prototype and test it with real customers in ordinary rooms and lighting.
6. Add analytics, clear on-screen instructions and a fallback for phones that do not support AR.
7. Launch to part of your audience, compare results, then scale the catalogue.

If AR will live inside a shopping app, our guide to app development for retail covers the rest of the product. When you are ready, tell us what you sell and who you want to reach. We will suggest a format, outline the 3D work and send a fixed-scope quote.