# Should I use in-flow save offers or post-cancellation win-back?

Source: Lazyweb Research
Published: 2026-07-07
Sample size: n=48
Tags: cancellation, retention, monetization, upsell, experiments
HTML: https://www.lazyweb.com/research/win-back-vs-in-flow-save-offers-cancellation
Markdown: https://www.lazyweb.com/research/win-back-vs-in-flow-save-offers-cancellation.md

**Answer.** Both appear in the field, and they operate at different moments. In-flow, discount offers dominate (30 of 48 apps, 63%) while pause and downgrade are rarely observed (0 of 48 each) [1][2]. Post-cancellation, named win-back plays like Paramount+'s '1 month free' restart modal and DoorDash's post-cancel recovery banner target users at the moment of loss [3][4].

> In-flow discount in 63% of 48 cancel apps; named win-back modals run after cancel, July 2026.

## In-flow: discount is the workhorse

On the cancel screen itself, **discount/offer language appears in 30 of 48 apps (63%)**, while pause and downgrade are each observed in 0 of 48 [1][2]. Headway's in-flow style pairs a **33% discount** with loss-aversion on a 'trial expired' modal, listing the benefits the user is about to lose [5]. In-flow, discount plus loss-aversion is the dominant observed combination.

## Post-cancellation: win-back at the moment of loss

Win-back fires **after** the user leaves. Paramount+ shows a restart modal offering '1 month free trial' with a benefit list and a 'Restart Subscription' CTA — a resubscribe play, not an in-flow save [3]. DoorDash A/B-tested replacing a static saved-store card with an **order-cancellation recovery banner** ('Your order was canceled, we'll help you find a replacement') surfaced immediately at the moment of loss [4].

## How to apply it

If you can only build one path first, in-flow **discount** is the well-trodden choice with the most observed prevalence [1]. Layer win-back (free-month restart, recovery banner) as a second surface for users who complete cancellation, timing it to the moment of loss where the field places it [3][4]. Pause and downgrade remain whitespace in this corpus if you want a differentiated in-flow offer [2].

## The numbers

| Stat | Computed from |
| --- | --- |
| 30 of 48 apps (63%) show in-flow discount/offer language | save_offer_offer_language |
| Pause 0/48 and downgrade 0/48 apps in-flow | save_offer_pause |
| Paramount+ win-back modal offers '1 month free trial', 'Restart Subscription' CTA | qualitative / paramount win-back-modal |
| DoorDash A/B added post-cancellation order-recovery banner | qualitative / doordash experiment |
| Headway 'trial expired' modal offers 33% discount with loss-aversion | qualitative / headway discount-to-retain |

## Methodology

Universe: 48 apps with genuine cancel-intent screens (401 screenshots) plus named win-back examples, July 2026. Method: company-level keyword prevalence and named-example vision descriptions. Caveat: win-back examples are illustrative named cases, not a measured rate.

## Sources & citations

- [1] Lazyweb Research analysis of 48 apps (genuine cancel-intent screens), July 2026. In-flow discount and pause/downgrade prevalence.
- [2] Lazyweb Research analysis of 401 cancel-intent screenshots (48 apps), July 2026. Headway in-flow discount modal vision description.
- [3] Lazyweb Research analysis of Paramount+ and DoorDash win-back examples, July 2026. Post-cancellation restart modal and order-recovery banner.

## Related questions

- [What percent of cancel flows show a save offer, and which type?](https://www.lazyweb.com/research/what-percent-of-cancel-flows-show-a-save-offer)
- [Are pause and downgrade save offers worth building for cancellation?](https://www.lazyweb.com/research/are-pause-and-downgrade-save-offers-worth-building)