---
title: "The Points Approval Period Is a Strategy Decision, Not a Fraud Setting"
slug: "points-approval-period"
url: "https://joy.so/blog/points-approval-period/"
kind: "post"
author: "Thomas Nguyen"
author_url: "https://joy.so/author/thomas/"
published_at: "2026-11-06T09:00:00.000+00:00"
updated_at: "2026-11-06T09:00:00.000+00:00"
description: "A 14-day hold makes BFCM points spendable in mid-December, after most calendars send the email telling customers to spend them."
category: "Program design"
tags: ["Program design", "BFCM & seasonal", "Loyalty Program", "How-to Guide"]
reading_time: 8
feature_image: "https://cdn-web.joy.so/cdn/image/2026/10/row15-approval-4e3db1.webp"
---

# The Points Approval Period Is a Strategy Decision, Not a Fraud Setting

Somewhere in your loyalty app there is a field with a number of days in it. It decides when points earned on an order stop being a display and become money a customer can spend. Most merchants have never opened it, and the ones who have opened it treated it as a fraud control.

It is a fraud control. It is also the setting that decides whether your December campaign is possible, because a 14-day hold on orders placed over the Black Friday weekend makes those points spendable around the middle of December — after the week in which most published BFCM calendars send the message telling customers to spend them.

**Key Takeaways**

- The approval period holds earned points unusable for a set number of days after an order is paid. It exists so a customer cannot spend points and then return the order.
- A 14-day hold on orders placed over the BFCM weekend makes those points spendable around the middle of December, which is after most vendors' "spend your points" email goes out.
- LoyaltyLion's 14-day default applies only to the Make a purchase rule created when the site was set up. On a rule you create yourself the field is empty, so a merchant who rebuilt the rule may have no hold at all.
- Rivo expresses the same control as a delay in seconds, up to a documented maximum of 60 days.
- Shortening the period for the BFCM cohort trades return exposure for a December redemption window. NRF and Happy Returns put the ecommerce return rate at 19.3% of online sales for 2025, which is the number that trade is priced against.

**Published 6 November 2026.** Vendor behaviour below was read from each help centre on 13 October 2026 and is linked. Where a vendor does not document the setting, this page says so.

## What the setting does

[LoyaltyLion's guide to the Make a purchase activity](https://help.loyaltylion.com/en/articles/1965585-activity-make-a-purchase) describes it as holding **"a customer's points in a pending state for a set number of days after their order is paid"**, and gives the reason without euphemism: **"This protects you from a customer spending their points straight after a purchase and then returning the order"**. The points are visible to the customer throughout. They simply cannot be redeemed yet.

The clock starts when the order is marked as paid, not when it ships, which matters in a season when fulfilment runs behind.

![Screenshot of LoyaltyLion's approval period documentation](https://cdn-web.joy.so/cdn/image/2026/10/shot-loyaltylion-f12838.webp)

_LoyaltyLion's own wording, read 13 October 2026. [LoyaltyLion Help Center](https://help.loyaltylion.com/en/articles/1965585-activity-make-a-purchase)_

## What each app does, and what it does not say

| App | What it is called | Default | Documented limits and advice |
| --- | --- | --- | --- |
| LoyaltyLion | A pending state for a set number of days after the order is paid | 14 days on the Make a purchase rule created at setup; the field is empty on a rule you create yourself | Recommends matching it to your returns policy, and keeping it to 30 days at most |
| Rivo | A delay between the order and the earning being approved, set in seconds | Not stated; 0 means no delay | Documented maximum of 5,184,000 seconds, which is 60 days |
| Joy | Pending Points — a delay in making points available after a purchase | None documented; the merchant sets the waiting period | Available on the Advanced and Ultimate plans. Pending points cannot be redeemed and do not contribute to tier status |
| Smile | Not covered in the order settings documentation | Not documented | Confirm in the app |

One row deserves particular attention, because it is the kind of detail that quietly undoes a policy. LoyaltyLion's 14-day default belongs to **the Make a purchase rule created when your site was first set up**. On a rule a merchant creates themselves, **the field is empty**, as [LoyaltyLion's own article on the purchase rule](https://help.loyaltylion.com/en/articles/1965585-activity-make-a-purchase) states. So a store that rebuilt its earning rules at some point — a rebrand, a migration, a new programme design — may have no hold at all and no indication that it lost one.

Rivo takes the same idea and expresses it as a delay measured in seconds. [Its guide to points delays documents a ceiling of 5,184,000 seconds](https://help.rivo.io/en/articles/6952156-excluding-orders-by-tag-products-and-collections-and-adding-points-delays), which is 60 days. Rivo is also clear about what the delay does not do: it changes when earnings are approved, not how many points the order earns. The [cost of the programme is identical](https://joy.so/blog/loyalty-point-cost/) either way.

One more behaviour to know before you change anything, because it mirrors the discount-combination trap exactly: [LoyaltyLion's approval period documentation](https://help.loyaltylion.com/en/articles/1965585-activity-make-a-purchase) says **"Changing the approval period only applies to new purchases. If you change it from 14 days to 3 days, orders placed before the change keep their original 14 days"**. Shortening the hold on 1 December does nothing for the orders placed over the weekend, which are the only orders you wanted it to affect.

Joy calls the same control **Pending Points**, and it does one thing the others do not document: **"Only Available Points can be redeemed for rewards or contribute to the customer's tier status"**, as [Joy's guide to Pending Points](https://help.joy.so/reward-programs/advanced-settings/pending-points/) puts it. So the hold covers tier progress as well as redemption, which matters if your December plan leans on customers crossing a threshold. It is [available on the Advanced and Ultimate plans](https://joy.so/pricing/), and since April 2026 merchants on those plans can switch it on themselves rather than raising a ticket. No default waiting period is documented — the figure in Joy's example is an example, not a setting you inherit.

![Screenshot of Joy's Pending Points documentation showing plan availability](https://cdn-web.joy.so/cdn/image/2026/10/shot-joy-pending-f1dc53.webp)

_Joy documents the same control as Pending Points. Read 13 October 2026. [Joy Loyalty Help Center](https://help.joy.so/reward-programs/advanced-settings/pending-points/)_

## The December arithmetic

Put the 2026 calendar against a 14-day hold. [Black Friday falls on 27 November](https://joy.so/blog/shopify-bfcm-guide/) and Cyber Monday on 30 November, so orders from that weekend clear their approval period between roughly **11 and 14 December**. Before that date, a campaign asking customers to spend their new points is asking them to do something the system will not let them do.

This is where the setting stops being an operations detail. The post-weekend email sequence that every vendor recommends — thank you, then a reason to come back — collides with a hold that nobody in the marketing conversation knows about. The customer clicks, finds a pending balance, and the moment is spent.

There are only three honest ways out, and picking one in October is much cheaper than discovering the problem on 8 December. Shorten the period for the BFCM cohort and accept more return exposure. Keep the period and move the campaign to the window after it clears. Or keep both and change the message, so the December email tells customers what they have earned and when it unlocks, which is a reason to come back with a date on it rather than a broken call to action.

## What you are trading, in numbers

Shortening the hold buys a redemption window and sells return protection, so it is worth knowing the scale of the thing you are selling. The most recent figures with a published method come from NRF and Happy Returns, and each one carries a different population, which is exactly how this set gets misquoted.

| Figure | Population | Note |
| --- | --- | --- |
| 15.8% / $849.9 billion | Of the annual sales of the retailers surveyed | Not an economy-wide figure, and often misquoted as one |
| 19.3% | Of online sales, forecast for 2025 | The number that matters to an ecommerce loyalty program |
| 17% | Of holiday sales, as retailers expect | The seasonal version, and the one to set the approval period against |
| 9% | Of all returns, fraudulent | The reason the setting exists in the first place |
| 7.7 returns | Average for surveyed consumers aged 18 to 30 — and the survey only included people who had already returned at least one online order | More than any other generation surveyed |

[NRF's press release on returns](https://nrf.com/media-center/press-releases/consumers-expected-to-return-nearly-850-billion-in-merchandise-in-2025) gives the method behind them: 2,006 consumers who had returned at least one online purchase in the previous twelve months, and 358 professionals working in ecommerce at US merchants above $500 million in revenue, both surveyed in the summer of 2025. Large US merchants are not a proxy for your store, and the right use of these numbers is as an order of magnitude against which to read your own.

![Bar chart of return rates by population from NRF and Happy Returns](https://cdn-web.joy.so/cdn/image/2026/10/fig-row15-returns-fd7e5b.webp)

_Each figure with its population attached._

One more thing the source says that the category routinely gets backwards: the rate is **down**, not up. [NRF puts 15.8% against 16.9% the previous year](https://nrf.com/media-center/press-releases/consumers-expected-to-return-nearly-850-billion-in-merchandise-in-2025), when returns totalled $890 billion, and describes this year's figure as in line with it. If you have read that returns are climbing, that framing is not coming from this dataset.

The figure that matters most here is **19.3% of online sales returned**. If roughly one order in five comes back, a hold that expires before your return window closes means a measurable number of customers will [spend points on orders they then return](https://joy.so/blog/loyalty-fraud-prevention/) — and those points are removed afterwards, which is a worse conversation than the pending balance would have been.

## And when the order does come back

LoyaltyLion removes the points automatically through webhooks, proportionally on a partial return, and [its guide to refunds and returns](https://help.loyaltylion.com/en/articles/1965601-refunds-and-returns-on-purchased-products) documents the uncomfortable part: it happens **"regardless of whether the customer has points in their account"**. A customer who already spent them goes negative.

That is the scenario the approval period exists to prevent, and it is also the scenario a shortened period makes more likely. If you shorten it for the BFCM cohort, decide in advance what your support team says to the first customer who finds a negative balance, because in December they will not have time to invent an answer.

## How to decide, in one sitting

Start from your returns policy, which is what LoyaltyLion recommends and what most merchants have never aligned the setting to. Then look at the December campaign you intend to run and work backwards from the date it needs points to be spendable. If the two dates conflict, the campaign is usually the cheaper thing to move.

Then check the setting actually holds what you think it holds, especially if your earning rules have been rebuilt at any point. An empty field looks exactly like a field you have not configured, and it behaves like no protection at all.

The mechanics that work on a loyalty base during the weekend itself are in our [BFCM loyalty strategies guide](https://joy.so/blog/loyalty-strategies-for-bfcm/), and the order of work for the weeks before it is in our [Black Friday checklist](https://joy.so/blog/black-friday-checklist/).

## A field with a date in it

Loyalty programs are full of settings that look operational and behave strategically. This is the clearest example: a number of days, entered once, that decides whether the most valuable cohort you acquire all year can act on what you gave them during the window you planned to reach them in. Open it before November, not after.

## FAQs

### What is a points approval period?

A hold. Points earned on an order sit in a pending state for a set number of days after the order is paid, visible to the customer but not spendable. LoyaltyLion states the reason plainly: it protects you from a customer spending their points straight after a purchase and then returning the order.

### What should I set it to?

Your returns window, as a starting point, which is LoyaltyLion's own advice. Then ask whether the BFCM cohort deserves a different answer, because a hold sized for an ordinary month pushes BFCM points past the window in which you want them spent.

### When do BFCM points become spendable on a 14-day hold?

Orders placed between 27 and 30 November 2026 clear around 11 to 14 December. If your December campaign is built on customers spending points, that is the earliest it can work, and most published BFCM email calendars send the "spend your points" message before it.

### Is it safe to shorten the period for Black Friday?

It is a trade, not a free move. You are exchanging return exposure for a redemption window. Price it against your own return rate rather than an industry one: the figures below are survey averages across large US merchants, and your category may be nothing like them.

### What happens to points if the order is returned?

They are removed automatically, proportionally on a partial return, and LoyaltyLion documents that this happens regardless of whether the customer still has the points in their account. A customer who spent them goes negative, which is a support conversation you would rather not have in December.

### Does the delay change how many points are earned?

No. Rivo is explicit that an earnings delay changes when earnings are approved, not how many points or how much store credit the order earns. The cost of the programme is unchanged; only the timing moves.