Most Facebook and Instagram ads for small business NZ owners run end the same way: a few hundred dollars spent, a handful of clicks, and no clear sense of what actually worked. That is rarely a budget problem. It is a structure problem: the campaign was never set up to produce a clean answer in the first place. Get the structure, targeting and creative right before you spend a dollar, and even a modest budget will tell you something useful.

How a Meta ad campaign is actually structured

Facebook and Instagram ads run through Meta Ads Manager, which organises every campaign into three layers. Understanding this structure matters more than any single setting, because it decides what data you get back at the end.

• Campaign level: You choose one objective here (Awareness, Traffic, Leads, or Sales). One campaign should chase one goal. Mixing objectives inside a single campaign muddies the results and gives Meta’s algorithm conflicting signals to optimise against.
• Ad set level: This is where targeting, placements, budget and schedule live. Each ad set defines who sees the ads and how much you are willing to spend reaching them.
• Ad level: The actual creative: image or video, copy, and call to action. Several ads can sit inside one ad set so you can test which version performs best.

For a first campaign, keep it simple: one campaign, two or three ad sets testing different audiences, and two or three ads inside each set testing different creative. Anything more complex than that splits your budget too thinly to produce a reliable result.

What to spend before you touch targeting

Budget decides how fast Meta’s algorithm can learn, not just how far your reach goes. As a rough guide, a campaign needs somewhere between five and ten conversion events a week before the algorithm has enough data to optimise properly. Below that, it is still guessing.

We have covered the pricing side of this in detail elsewhere, including what social media marketing actually costs in Auckland and how much a NZ small business should budget for social media. The short version for paid ads specifically: start at ten to thirty dollars a day per ad set, run it for at least four to five days before judging results, and only increase spend once you have a genuine baseline to compare against.

Targeting that works for a small NZ audience

New Zealand’s population is small and regionally concentrated, which changes how targeting should be approached compared with larger markets. Stacking five or six interests on top of narrow age brackets and a small radius often shrinks the audience so far that Meta cannot deliver ads efficiently.

A more reliable approach for local businesses:

1. Start broad. Let Meta’s delivery system find the right people within a wide age range and a sensible location radius, rather than pre-guessing who will convert.
2. Layer in first-party data once you have it. Website visitors, email lists and past customers make far stronger audiences than interest categories, because they are based on real behaviour.
3. Build lookalike audiences from actual conversions. Once you have thirty to fifty genuine leads or sales, a lookalike audience built from that data will usually outperform manual interest targeting.
4. Test placements, not just audiences. Feed, Stories and Reels behave differently. Let the platform’s Advantage+ placement options run rather than restricting to one format from the start.

Platform choice still matters before you get to targeting. If you have not settled that yet, our guide to choosing the best social media platform for a NZ business is a useful starting point.

Creative that actually gets clicked

Targeting decides who sees your ad. Creative decides whether they act on it, and in Meta’s current system, creative quality carries more weight than granular targeting ever did. A broad audience paired with strong creative consistently outperforms a narrow audience paired with generic creative.

For NZ service and retail businesses, a simple formula holds up well:

• Problem: Open with the exact pain point your customer already has, in their own words.
• Solution: Show plainly how your product or service solves it.
• Proof: A review, a customer count, or a before-and-after result.
• Specific call to action: “Request a Free Quote” or “Book a Free Assessment” outperforms a vague “Learn More”.

Vertical video (9:16) built for Reels and Stories now outperforms static images for most objectives, and it does not need to be polished. Footage shot on a phone, provided it is well lit and clearly framed, regularly beats studio production because it looks native to the feed rather than like an ad.

The compliance rules most small businesses miss

This is the part most generic advertising guides skip, and it is where NZ businesses carry real risk. Every ad served to a New Zealand audience, including on Facebook and Instagram, sits under the Fair Trading Act. Claims must be accurate and capable of being backed up, not just persuasive.

Before you publish an ad, check it against these points:

• No unsubstantiated claims. “Voted Auckland’s best” or “guaranteed results” needs evidence behind it, not just confidence.
• Testimonials must be genuine and current. A review used in an ad should be real, attributable, and not materially out of date.
• Before-and-after imagery needs context. Typical timeframes and conditions should be disclosed, particularly for services where results vary person to person.
• Pricing claims must match what a customer actually pays. “From $X” pricing that almost nobody qualifies for is a common source of complaints.
• Sponsored or partnered content needs disclosure if you are running ads on behalf of, or featuring, another party.

None of this should slow a campaign down. It takes minutes to check and it is far cheaper than a Commerce Commission complaint or a platform takedown mid-flight.

Testing and scaling without resetting the algorithm

Once a campaign is producing results, the instinct is to scale fast. Resist it. Jumping a budget from fifty to two hundred dollars a day overnight typically resets the ad set’s learning phase, and performance drops before it recovers.

1. Increase budget in steps of twenty to thirty percent, spaced a few days apart, rather than in one large jump.
2. Test one variable at a time. Swap creative, then wait for results, before touching audience or copy.
3. Keep a winning ad running while new variants are tested alongside it, rather than pausing it to try something new.
4. Retire creative before it fatigues. Falling click-through with rising frequency is the signal to refresh, not to keep pushing spend into it.

Your first-campaign checklist

• One campaign, one objective, two to three ad sets
• Ten to thirty dollars a day per ad set to start
• Broad targeting first, first-party data layered in once available
• Two to three ad variations using the problem-solution-proof-CTA formula
• At least one vertical video creative in the mix
• Every claim checked against the Fair Trading Act before publishing
• No budget or audience changes inside the first four to five days
• Scale spend in small steps, not sudden jumps

FAQ

How much should a NZ small business spend on Facebook and Instagram ads to start?

Ten to thirty dollars a day per ad set is a workable starting point for most small businesses, run for at least a week before judging results. For a full breakdown of realistic monthly figures, see our guide to social media marketing budgets for NZ small business.

Is Facebook or Instagram better for paid ads in NZ?

Both run through the same Meta Ads Manager and can be included in the same ad set, so it is rarely an either-or choice. Instagram tends to suit visual and lifestyle-driven brands, while Facebook still reaches a broader, slightly older local audience.

How long before a Facebook or Instagram ad campaign shows results?

Allow four to five days minimum for the ad set to exit its learning phase before drawing conclusions. Meaningful trend data usually takes two to three weeks of consistent, unchanged delivery.

Do testimonials in ads need to be real customers?

Yes. Under the Fair Trading Act, testimonials used in advertising must be genuine, attributable and not materially misleading about typical results.