Meta Ads

Facebook Ads for eCommerce

Facebook ads for eCommerce are campaigns optimized for purchases, not traffic: the Pixel and Conversions API send every view, add-to-cart and purchase with its value, and the structure moves buyers from first sight to checkout to a second order. This is the store-focused service within my Meta Ads services.

  • Purchase events with values before any spend
  • Prospecting, catalog retargeting and cart recovery
  • Nepal stores and international brands

Who this service is for

Online stores that already take orders on a website: Shopify, WooCommerce, Magento or a custom cart, selling in Nepal or to Australia, the UK, the USA, Canada or the UAE. It also fits Nepali brands that sell through Facebook and Instagram messages but want a store to carry the checkout, because a message-to-order flow cannot be measured as return on ad spend. If you sell services or collect enquiries rather than orders, Facebook lead ads is the right page; if you have a catalog but no ads history, start with catalog and dynamic product ads.

What running eCommerce ads on Meta includes

  • Purchase tracking that Meta can learn from. ViewContent, AddToCart, InitiateCheckout and Purchase events with product IDs, currency and value, sent from the browser and the server and deduplicated. This is the Pixel and Conversions API setup step and it happens before any budget is spent.
  • A product catalog in Commerce Manager synced from your store so ads can show the exact items a person viewed.
  • Campaign structure by funnel stage: prospecting to cold audiences, retargeting to viewers and cart abandoners, and a retention layer for past buyers.
  • Creative for products: product video, user-generated content, carousel and collection formats, with copy in Nepali or English for the Nepal market and native English for overseas stores.
  • Weekly optimization on return on ad spend and cost per purchase, and a monthly report that reconciles Meta’s numbers with store orders.

How the funnel is built

  1. Audit the store’s checkout and events. I place test orders and confirm every event fires once with the correct value. A store that reports two purchases for one order will mislead every decision that follows.
  2. Prospecting. Broad or interest-based audiences see best-seller and collection creative. On accounts with clean purchase data I test Advantage+ Sales campaigns (Advantage+ Shopping until its 2025 renaming) against a manually structured campaign and keep whichever produces a lower cost per purchase; Meta merged the two setups into one campaign flow in 2026, so the comparison is now a settings choice rather than two campaign types. Details are on the Advantage+ campaigns page.
  3. Catalog retargeting. People who viewed or carted a product see that product, with a reason to return: free delivery in Kathmandu Valley, a stock warning, or a time-limited offer.
  4. Cart recovery. A short window, usually one to three days, with the strongest offer the margin allows, then a longer window with product reminders.
  5. Retention. Past buyers see new arrivals and replenishment reminders, excluded from prospecting so the budget does not pay to re-acquire them.
  6. Scale. Budgets rise in steps on ad sets that hold their return on ad spend; creative is refreshed every cycle because product ads fatigue faster than lead ads.

What is different for stores in Nepal

Cash on delivery is still the majority of orders, so the Purchase event has to fire on order confirmation, not on payment, and the report has to allow for returns and refused deliveries. Delivery radius matters: a Kathmandu store that ships nationally by courier still converts best inside the Valley, so campaigns are often split by delivery zone. Paying Meta is a constraint of its own: Nepali individuals can hold prepaid dollar cards limited to USD 500 a year, an April 2026 Nepal Rastra Bank amendment lets banks set their own prepaid-card limits, and a registered IT business can obtain a card with a USD 3,000 to 5,000 annual limit. Stores that cannot pay Meta directly run through my Facebook boost service and are invoiced in rupees. Instagram Shopping and checkout features depend on Meta’s commerce eligibility for the country.

What is different for stores abroad

Overseas stores usually have cleaner payment data and a higher share of iOS traffic, which makes the Conversions API and event match quality the deciding factor in how well Meta attributes purchases. Competition per impression is higher, so creative volume, offer strength and landing page speed carry more weight than audience choice. I report in the store’s currency and reconcile against Shopify or WooCommerce orders rather than trusting Ads Manager alone.

Mistakes I see in eCommerce accounts

  • Campaigns optimized for link clicks or add-to-cart because the Purchase event was never installed.
  • No values on events, so return on ad spend cannot be calculated and Meta cannot bid toward higher-value orders.
  • Retargeting audiences that include recent buyers, who then see ads for what they just bought.
  • One catalog with no product sets, so a sale on shoes shows ads for bags.
  • A single static image running for months while cost per purchase climbs.

What you receive

A verified event setup, a catalog connected to your store, the three-stage campaign structure, creative briefs and copy for each stage, weekly optimization notes, and a monthly report that shows spend, purchases, revenue and return on ad spend against store orders. Ownership of the Business Manager, Pixel, catalog and ad account stays with you.

Budget, pricing and next step

For Nepal stores a starting ad budget of NPR 15,000 to 30,000 a month is enough for the funnel to gather purchase data; international stores usually start at USD or AUD 500 to 1,000 a month and scale on results. The management fee sits within the ranges on the Meta Ads packages and pricing page. If you already run ads, a free Meta Ads audit of your event setup and funnel is the fastest way to find out what is being wasted.

Frequently asked questions

What budget should I start with for eCommerce ads?

For Nepal campaigns, NPR 15,000 to 30,000 a month in ad spend is a workable start; for international campaigns, USD or AUD 500 to 1,000 a month. The right figure depends on your average order value, margin and audience size, and I confirm it after looking at your store and its events.

Do I need a website, or can I sell through Facebook messages?

You can run message campaigns without a store, and many Nepali brands do, but you cannot then measure return on ad spend or use catalog retargeting. A simple store, even one that hands off to cash on delivery, lets the Purchase event drive optimization. I can help specify the store; the build sits under my web development services.

Do you handle Shopify and WooCommerce tracking?

Yes. Both have native Meta integrations that send Conversions API events, which I configure and then verify with test orders in Events Manager. Custom carts get Pixel events through Google Tag Manager plus a server-side connection.

Should I use Advantage+ Sales campaigns?

Only once the Pixel has enough clean purchase data. On a new account I start with a structured campaign, then test Advantage+ Sales against it. On stores with an established purchase history it often wins on cost per purchase, but I keep a manual campaign for launches and promotions where I need control.

Can you run ads for my store if I cannot pay Meta from Nepal?

Yes. Through the boost service your campaigns run from an agency ad account and you pay the ad budget and fee in rupees. Tracking, catalog and campaign structure are identical.

How do you count cash-on-delivery orders?

The Purchase event fires on order confirmation so Meta can optimize, and the monthly report shows confirmed orders separately from delivered and paid orders, using your store or courier data. That way the return on ad spend figure reflects money actually received.

Ready to talk about your project?

A free 30-minute call, a straight answer about what would move the numbers, and a written proposal within 48 hours if we are a fit.