Client name withheld under NDA.
The store sold pram accessories, feeding sets, toys and clothing for babies and toddlers. Most orders came from Google Shopping and paid social, and the owner worried about how much each new customer cost. She had put an affiliate link programme on the website a year earlier, after two parenting bloggers asked for one. Six people used it. Nobody was recruiting, the commission was a flat 15% on everything, and the links often failed to credit the sale when a visitor came back a day later on a different device.
The owner wanted a channel where she paid only when a sale happened, and she wanted it to be honest. Parents trust other parents, so a recommendation had to look like one. We started by testing the links that already existed and found that about one in five sales was not being credited. We recorded the programme's order counts, revenue and cost per order before changing anything, so the later numbers would mean something.
We tested every existing link on phones and computers, in different browsers, and with a delay of one to seven days between click and purchase. We fixed the tracking, set a 30-day cookie window and added a server-side order check so a sale made after a return visit still counted. We also read the programme terms, which were one paragraph long.
If a sale is not credited, the affiliate stops promoting and the store loses its best partners. Tracking has to be right before anyone is recruited. A clear set of terms also protects the store when a partner breaks the rules.
We replaced the flat 15% with tiers by category and by volume. Feeding and clothing paid 12%, toys paid 10% and bundles paid 8%. Affiliates who sent more than 20 orders in a month moved up two points. We wrote plain-English terms: no bidding on the store's brand name, no coupon pop-ups that overwrote another partner's link, and a clear note on every post that links were affiliate links.
Commission is the whole business model. The store needed margin on each sale, and the affiliate needed enough to keep posting. Tiers protected low-margin items and rewarded the partners who sent real volume. Disclosure rules keep the recommendations honest and keep the store in line with Singapore advertising guidelines.
We built a list of about 150 possible partners: parenting bloggers, mum communities, product review sites, cashback sites and a few newsletters. We sent personal messages with a sample box offer for fit partners, not a generic invitation. Each new affiliate received a short welcome pack with product photos, a reviewer's checklist and a link to a private dashboard.
Quality beat quantity. A mum blogger with 3,000 loyal readers outsold a large coupon site that sent traffic with no intention to buy. We spent most of our time on the first group and kept coupon partners on a short leash.
We gave partners product facts, accurate sizing guides, a seasonal calendar and approved images. We also built two comparison landing pages on the store's site, one for newborn feeding sets and one for toddler gifts, which partners could link to. The store's own blog posts were updated so affiliates could quote them.
Affiliates sell better when the page they send readers to answers the questions readers have. A tidy landing page raised the average order value because shoppers bought sets instead of single items.
We checked orders every week for self-referrals, unusual coupon use and refunds. A handful of orders were reversed, and one partner was removed for bidding on the store's name. A monthly report showed orders, revenue, cost per order and the top ten partners, and each partner received a short note on their own results.
Fraud control keeps the programme profitable and the good partners confident. A report that compares cost per order against other channels also gives the owner a reason to keep investing.
Six months in, the affiliate programme had grown from a side project into a real sales channel. Active affiliates went from 6 to 41, monthly orders from 22 to 238 and monthly revenue from S$1,400 to S$17,600. The store paid S$10.20 per affiliate order, below what it was paying per order on paid social, and the share of online revenue from affiliates reached 11%.
Tracking repaired, commission tiers and terms written, existing affiliates re-briefed.
About 150 partners approached, welcome pack and dashboard live, first new affiliates selling.
Landing pages published, fraud checks running weekly, monthly reporting and tier reviews in place.
About one in five sales had not been credited. Once credit reached 98% of sales, partners trusted the programme and posted more. Recruiting into a broken system only wastes the partners you worked to find.
Forty-one active affiliates produced S$17,600 a month, and the top ten brought in most of it. Parenting bloggers and community owners outsold coupon sites by a wide margin. A small group of partners whose readers trust them is worth more than a long list.
Tiered commission held the cost per order at S$10.20 while the average order value rose from S$64 to S$74. Paying more for volume and less for low-margin items is how an affiliate programme stays profitable as it grows.
How we set up and run affiliate programmes for Singapore brands.
Working with creators whose followers trust their opinion.
Online stores built for sales and easy tracking.
Another online store, grown through Google Shopping.
Tell us about your store and margins. We will tell you if affiliate marketing fits and what a realistic six-month plan looks like.