Personalizations
A personalization shows a different version of your site to a specific audience, such as returning customers, visitors from one country, or shoppers who arrived from a particular ad. Each variation has its own target, and visitors see the first variation whose target they match. A control group, 25% by default, keeps seeing the original so you can measure the lift the personalization brings.
Available on: all platforms. Theme, template, price and shipping personalizations are Shopify only. See Campaigns.
How it works
- A visitor reaches a page where the campaign applies. For content and URL split personalizations, the page must match the campaign's pages.
- Variations are checked in order. The first variation whose target the visitor matches is the one they get. Variations further down the list are ignored for that visitor.
- Traffic allocation is checked. If allocation is below 100%, only that share of visitors enters the campaign.
- The control group is decided once. The visitor is placed in control (sees the original) or not, based on the control group percentage. That decision is kept for the whole campaign, whatever variation they match later.
- A view is recorded, for the variation or for control. Control views remember which variation the visitor would have seen, so the report can compare personalized and control visitors per variation.
- On later pages, targeting mode decides what they see. With Live, targets are checked again on every page. With Permanent, a visitor keeps the variation they first got.
If a visitor matches no variation, nothing changes on the page and nothing is recorded.
Set up a personalization
- Go to Campaigns → New campaign, choose Personalization, then the type.
- Choose the target for the first variation, then define its changes.
- In the campaign editor, use Create variation and pick a target to add more variations. Each target can be used once per campaign.
- Drag variations into priority order. The most specific audience should usually come first.
- Set the Control group and Targeting mode.
- Add goals and choose the Primary metric.
- Preview each variation, then Start or Schedule the campaign.
Option | What it does | Values / default |
|---|---|---|
Variations | One version of the site per audience | One or more, each with its own target |
Variation order | Which variation wins when a visitor matches several targets | First match wins. Drag to reorder |
Control group | Share of matched visitors who see the original and are counted as control | 0–100%, default 25% |
Targeting mode | Live: targets are checked on every page. Permanent: the visitor sees the same variation once they've matched a target | Live. Price personalizations always use Permanent |
Traffic allocation | Share of matched visitors who enter the campaign at all | 0–100%, default 100% |
Goals and Primary metric | What's measured | Conversion rate, Revenue per visitor, Profit per visitor, Average order value |
Trigger (content personalizations) | Immediate runs as the page loads. Manual runs when a widget's Run campaign automation or ply.runCampaign(id) calls it | Immediate |
Behavior details and limitations
- Order is priority. A visitor who matches both "Returning customers" and "Visitors from Canada" sees whichever variation is higher in the list. Put narrow audiences above broad ones.
- The control group is per campaign, not per variation. A visitor in control stays in control for every variation of the campaign. The report still splits control visitors by the variation they would have seen.
- 0% control shows the personalization to everyone who matches. You won't be able to measure the lift, because there's no baseline.
- Live targeting can switch variations. With Live, a visitor who matches a different target on a later page (for example, after logging in) sees the variation for that target. A visitor who matches no target on a page sees the original there.
- Permanent targeting freezes the first match. With Permanent, the visitor keeps their first variation even if they stop matching its target or start matching a higher one. Use it when the audience is defined by something that changes during the visit, such as cart value or pages viewed, and you want a consistent experience.
- Personalizations can't be deployed. There's no winner to roll out: a personalization is meant to keep running. To stop measuring and show it to every matching visitor, set the control group to 0%.
- Reports compare personalized vs control. The report shows personalized and control totals, and per-variation results. When there's no control group, variations are compared to the worst-performing variation. See Reading campaign reports.
- Run from a widget. A content personalization with the Manual trigger can be started by a widget's Run campaign automation, for example after a visitor answers a quiz. See Widget automations.
Examples
- Geo-specific hero: variation 1 for visitors from the US with a "Free 2-day shipping" banner, variation 2 for visitors from the EU with "Free shipping over €50". 25% control.
- Ad landing personalization: a variation targeted at visitors who arrived with utm_campaign=summer-sale, repeating the ad's headline and offer on the homepage. The visitor's saved UTM values persist, so the visitor keeps matching on later pages, until a later link brings a different utm_campaign.
- VIP pricing: a product price personalization for customers tagged vip in Shopify, with a 10% control group to confirm the discount pays for itself.
Troubleshooting
- The wrong variation shows → the visitor matches a variation higher in the list. Reorder the variations or narrow the higher target.
- Nothing shows for a visitor who should match → they may be in the control group. Use a preview link to see a specific variation, or check the target in How targeting works.
- The variation flips between pages → the target matches on some pages and not others (for example, a page-based filter). Switch Targeting mode to Permanent.