Performance Marketing
2026-10-03 · 9 min read
Sapun Lamichhane
Founder & CEO of Arcetis
The single most common mistake is boosting a post for engagement or reach when the actual goal is inquiries or sales. Facebook's system optimizes for exactly what you tell it to optimize for, so a post boosted for engagement will reliably generate likes and comments, and nothing resembling a qualified lead.
If the goal is real inquiries, the campaign needs to be built as a lead generation or conversion campaign from the start, with the actual business outcome, a form fill, a message, a call, set as the objective Facebook is optimizing toward.
It is worth checking your current campaign's stated objective directly in Meta Ads Manager rather than assuming, since campaigns set up by someone unfamiliar with the platform are often left on a default objective that does not match the actual business goal.
“A Facebook ad optimized for engagement will get you engagement, if you want leads, you have to actually ask Facebook for leads.”
An ad that links straight to a Facebook page, with no clear next step, loses most of the people who were actually interested. A simple landing page or a clear in-platform action, like a lead form or a direct message prompt, gives interested people an obvious way to act.
In Nepal specifically, a large share of genuine interest converts through Messenger or WhatsApp rather than filling out a web form. Click-to-WhatsApp or click-to-Messenger ad formats often outperform standard lead forms here, simply because that is where people are already comfortable communicating.
Whatever the destination, it should load quickly and clearly state what happens next, a phone number, a simple form, or a chat prompt, within the first few seconds. A slow or confusing landing page undoes the work the ad already did to get someone interested.
Trust signals matter more than polish for most local audiences. Real Nepali testimonials, clear mention of payment options like eSewa, Khalti, or cash on delivery, and straightforward pricing information tend to outperform highly produced but vague creative.
Testing creative in both Nepali and English, and sometimes a mix, is worth doing rather than assuming one language performs better across the board. Different audiences and even different products within the same business can respond very differently.
Video content, even simple phone-shot footage showing a product or service in use, often outperforms static images for engagement and trust, particularly for service businesses where people want to see real work rather than stock-style photography.
A small business testing lead generation seriously should budget at least NPR 15,000 to 30,000 a month in ad spend to gather enough real data to judge performance, separate from any management fee if an agency is involved.
Cost per lead varies enormously by industry and offer, anywhere from roughly NPR 100 to over NPR 1,000 per lead depending on how competitive the niche is and how specific the targeting is. Treat any number quoted to you as an estimate to be tested, not a guarantee.
These ranges shift quickly with seasonality and competition too, running the same campaign during a festival shopping period versus a quiet month can produce meaningfully different costs, so judge performance over several weeks rather than a single week's data.
In most accounts we see, the biggest loss of value happens after the lead arrives, not in the targeting. A lead that gets a reply within minutes converts at a dramatically higher rate than one that waits hours or gets lost in a shared inbox.
Even a simple process, a dedicated person checking messages regularly, a basic spreadsheet tracking follow-up status, or a lightweight CRM, usually produces a bigger improvement in actual sales than tweaking targeting further.
Training whoever answers inquiries to ask a few qualifying questions early, rather than just confirming price, also helps separate genuinely interested buyers from casual browsers, which matters because chasing every single inquiry with the same intensity is not always the best use of limited time.
Running a single small test campaign is manageable in-house. Once a business needs reliable creative testing, proper tracking, and a follow-up system that does not depend on one person remembering to check their phone, specialized help starts to pay for itself.
Arcetis manages real Meta ad spend for Nepali businesses and spends as much time on tracking and follow-up systems as on the ads themselves, because that is usually where the actual leads are being lost.
A specialist is also useful simply for catching wasted spend early, a campaign that is technically running but poorly targeted can burn through a month's budget before an inexperienced manager even notices something is wrong.
Almost always because the campaign was optimized for engagement rather than leads or conversions. Facebook delivers exactly what objective you select, so an engagement-optimized ad will reliably produce engagement and little else.
Both can work, but in Nepal, click-to-WhatsApp or Messenger ads often convert better because that is where people are already comfortable talking to a business. Lead forms can work well too, especially when paired with fast follow-up, so testing both for your specific audience is worth the effort.
At least NPR 15,000 to 30,000 a month in ad spend for a meaningful test, run for a few weeks to a month before judging results, separate from any management fee.