Short answer, if that is all you need: Meta ads work in Nepal when four things are true — the Pixel and Conversions API are both installed and deduplicated, campaigns run against a real conversion objective rather than boosted posts, creative is tested systematically because creative is the main lever, and the destination page matches what the ad promised. Boosting posts fails on all four.

Why boosting does not work

Almost every Nepali business that tells me Facebook ads do not work has been boosting posts. It is worth being precise about why that fails, because it is not a matter of degree.

Boosting optimises for engagement on the post. It hands you a fraction of the targeting control, limited placement control, no proper conversion objective and effectively no measurement structure. It reliably produces likes, shares and comments from people who will never buy anything, and those vanity numbers make it feel like something is happening. A campaign built in Ads Manager against a lead or conversion objective is a fundamentally different mechanism.

Measurement: Pixel plus Conversions API

Browser-side tracking has degraded significantly. Ad blockers, browser tracking prevention and plain network failures mean a meaningful share of pixel events never arrive, and the loss is systematic rather than random — so Meta's delivery optimisation is working from a distorted picture of who converts.

The fix is server-side events through the Conversions API, sent alongside the browser pixel and deduplicated by event ID so nothing is double counted. Getting the deduplication wrong is worse than not sending server events at all, so this is implementation work rather than a checkbox.

For Nepali businesses there is an additional wrinkle: a large share of enquiries come through Messenger, WhatsApp or a phone call rather than a web form. If those conversions are invisible, the campaigns look far worse than they are. Making them countable is often the highest-value change in the account.

The full detail on how I structure this, including targeting and testing, is in the Meta advertising section of my Nepal growth system write-up.

Campaign structure and targeting

Meta's targeting has moved steadily toward broad audiences with the algorithm doing the finding, which works when it has good conversion signal and fails when it does not. So structure follows data volume: with strong signal, fewer and broader ad sets with consolidated budget; with thin signal, tighter targeting and manual control while the account builds history.

Retargeting is where most small accounts leave money behind. People who visited a pricing page, started a form, watched most of a video or engaged with the Page are dramatically cheaper to convert than cold traffic, and building those audiences early costs nothing.

Creative testing, done systematically

On Meta, creative is the lever. Bidding and structure matter, but a genuinely better angle beats an optimisation tweak by a wide margin. That means testing distinct concepts and messages rather than five colour variants of one idea, giving each enough budget and time to produce a decision rather than a hunch, and killing losers quickly instead of letting them bleed.

It also means creative that looks native to the feed. Ads that look like ads get scrolled past, and this is more true in Nepali feeds than most brand guidelines allow for.

The destination matters as much as the ad

A click is only the halfway point. If the ad promises something specific and the landing page is a generic homepage, you have paid for attention and discarded it. Where the destination is the bottleneck I will say so rather than keep tuning audiences against a page that cannot convert — the structural issues are set out in the five conversion leaks I fix on every project.

Ownership, and the trap to avoid

This is the most important paragraph on the page. Your Business Manager, Facebook Page, Instagram account, Pixel, ad account and custom audiences must all be owned by you, with any agency or freelancer added as a partner. A vendor holding these in their own Business Manager is common in this market and it means that leaving them can cost you your Page, your entire pixel history and every audience you built.

If you are currently in that position, sorting it out is worth doing before anything else, and I will help you do it even if you do not hire me for the ongoing work.

I charge a fixed management fee rather than a percentage of ad spend, and you pay Meta directly for media.

Common questions

Facebook or Instagram?

Usually both, with placements decided by where results actually come from rather than by assumption. The audiences skew differently by age and category in Nepal, and the data answers this faster than an opinion does.

Meta Ads or Google Ads?

Meta creates demand among people who were not looking; Google captures demand from people already searching. If nobody is searching for what you sell, Meta is the better starting point.

Can you guarantee leads?

No. I can commit to correct measurement, methodical testing and telling you early and honestly if the channel is not viable for your offer.

Can you audit an existing account?

Yes. Read-only access is enough to produce a written assessment of what is misconfigured and what it is costing you.

See Google Ads management, SEO consulting for the channel that keeps working after you stop paying, and AI automation for responding to leads within minutes. Full list of services.

Tell me what you are paying per lead

If you are already running Meta ads, that number plus read-only access is enough for me to tell you where the waste is. If you are starting out, tell me what you sell and who buys it.

Get in touch Read the blog