Meta Ads in Turkey: Facebook and Instagram Advertising for Foreign Brands

Instagram and Facebook are not interchangeable placements in Turkey, and the gender split is the clearest evidence. According to DataReportal’s Digital 2026 Turkey report, published on 8 November 2025 using Meta’s own advertising tools, Instagram’s adult advertising audience in Turkey was 46.5% female and 53.5% male as of October 2025. Facebook’s was 34.2% female and 64.4% male. Instagram’s reported potential ad reach was 62.3 million against Facebook’s 34.7 million, and Instagram covered 92.1% of adults aged 18 and over against Facebook’s 53.2%.

A single campaign running across both placements with automatic distribution will therefore deliver two quite different audiences and report one blended number. That is the most common and most expensive mistake foreign brands make on Meta in Turkey. This article covers how to read the reach data, how to structure campaigns, what targeting you can lawfully build under Turkish data protection rules, how to measure when consent limits your signal, and what to review weekly.

Table of Contents

• The reach picture, and how to read it
• Instagram and Facebook are not one audience
• What to settle before you open the account
• Campaign structure: automated or manual
• Targeting under Turkish data protection rules
• Pixel, Conversions API and the consent problem
• Creative that works in Turkish feeds
• Budget and bidding when the currency moves
• Branded content and disclosure
• What to review weekly
• Frequently asked questions

The reach picture, and how to read it

The Digital 2026 Turkey report gives the following for October 2025, all from Meta’s advertising tools.

Instagram. Potential ad reach 62.3 million, equal to 70.9% of the total population, 92.1% of adults aged 18 and over, and 80.4% of internet users. Reported reach was up 5.80 million year on year, a change of 10.3%, and up 1.00 million over the preceding quarter.

Facebook. Potential ad reach 34.7 million, 39.5% of the total population, 53.2% of adults 18 and over, and 44.8% of internet users. Reported reach was up 350,000 year on year, a change of 1.0%, and down 450,000 over the preceding quarter.

Messenger. Potential ad reach 12.8 million, 14.5% of the total population and 19.4% of adults 18 and over, with reported reach down 500,000 year on year.

Read all of these carefully. They are Meta’s estimates of the audience an advertiser could potentially address, not monthly active users, and DataReportal is explicit that they should not be read as active user numbers. Meta has also revised its figures, so they are not directly comparable with earlier reports, which means a reported year-on-year change can reflect a methodology revision as much as a movement in real usage. Facebook’s quarterly decline and Messenger’s annual decline are both the kind of figure that may partly reflect platform corrections rather than people leaving.

What the data supports is a statement about relative scale and composition, not a growth narrative. Instagram addresses a substantially larger and more gender-balanced audience in Turkey than Facebook does. That is the planning input.

Instagram and Facebook are not one audience

	Instagram, Turkey	Facebook, Turkey
Potential ad reach	62.3 million	34.7 million
Share of adults 18+	92.1%	53.2%
Share of internet users	80.4%	44.8%
Adult ad audience, female	46.5%	34.2%
Adult ad audience, male	53.5%	64.4%
Reported year-on-year change	+10.3%	+1.0%

Gender percentages may not sum to 100% in Meta’s reporting, and these remain ad-audience estimates rather than verified user counts. Even allowing for that, a roughly thirty-point gap in female composition between the two placements is not noise.

Three practical consequences.

If your product skews female, automatic placement distribution works against you. Letting Meta allocate freely across Instagram and Facebook in Turkey can send budget to the placement with a markedly lower share of your actual buyer. Test placements separately before you consolidate.

Facebook is not dead in Turkey, and writing it off is a different error. It still reaches over half of Turkish adults. In categories with older or more male-weighted buyers, and for lead generation in particular, it often delivers lower costs than Instagram precisely because competition for that inventory is thinner.

Report placement-level results, always. A blended cost per result across two audiences this different tells you almost nothing actionable. If your reporting template does not break out placement, fix the template before you increase the budget.

Messenger deserves a deliberate decision rather than inclusion by default. Its reported potential ad reach in Turkey is 12.8 million — 19.4% of adults 18 and over — with an adult ad audience of 29.9% female and 69.3% male, the most skewed of the three, and reported reach down 500,000 year on year. For most consumer brands that makes it a small, narrow placement rather than a meaningful share of budget. Audience Network placements warrant the same scrutiny: they expand delivery into third-party apps where the brand-safety and quality position is harder to see, so include them knowingly or exclude them, but do not leave the decision to a default.

What to settle before you open the account

Three things belong in place before the first campaign, and none of them is a campaign setting.

The commercial and tax position. Advertising payments in Turkey carry specific tax treatment — withholding under Presidential Decision No. 476, VAT accounted for by reverse charge where the supplier has no Turkish presence, and a digital services tax levied on the platform rather than on you but capable of reaching you through platform pricing. These are not Meta-specific settings but they change your effective cost per result, and they are set out in our guide to advertising taxes in Turkey. Resolve this with tax advice before you budget, not after the first invoice.

Business asset ownership. Decide who holds the Meta Business Manager, the ad account, the Pixel or dataset, and the page: the brand, with agencies and partners granted access rather than ownership. An ad account and a conversion dataset built inside an agency’s or a distributor’s business portfolio carry historical learning you may not be able to take with you. This is the single most recoverable mistake if you fix it on day one and the least recoverable if you do not.

Where the traffic lands. Meta will deliver clicks competently to a page that cannot convert Turkish visitors, and the resulting cost per acquisition looks like a media problem. Confirm the destination works — Turkish language throughout, local payment presentation, mobile-first — before spending against it. Our web design and software service covers that build side.

Campaign structure: automated or manual

The honest answer in a market like Turkey is that automated campaign types work well once you have conversion volume and poorly before you do. Meta’s automated products optimise by learning from conversion signal. A new market entrant typically has no Turkish conversion history, limited volume, and a signal weakened by consent choices — which is precisely the condition in which automation has least to work with.

A workable sequence:

• Phase one, learn deliberately. Manual campaigns with separated placements, a small number of clearly distinct audiences, and a conversion event that actually fires often enough to optimise against. Accept higher management overhead in exchange for knowing which variable moved.
• Phase two, consolidate what you proved. Fold the audiences and creative that worked into broader structures, keeping placement separation where the Instagram and Facebook gap matters for your category.
• Phase three, automate with guardrails. Move to automated campaign types once conversion volume supports learning, retaining exclusions, brand-safety settings and a reporting view that still shows placement and creative level detail.

Two structural cautions specific to entering a new market. Optimise toward an event with enough volume to leave the learning phase — optimising a low-volume purchase event in a market with no history produces unstable delivery and misleading early results. And resist running too many small test cells at once: a budget split across fifteen ad sets in a market you do not yet understand generates noise rather than learning.

Audience overlap is a sharper problem in Turkey than in larger markets for a simple structural reason: a 62.3 million addressable pool sounds generous until you have narrowed it by city, age, language and interest, at which point several ad sets are frequently bidding for the same people. That raises your own costs and destabilises delivery. Check overlap between ad sets before launch rather than diagnosing it afterwards from erratic results, and consolidate rather than adding another cell when two audiences substantially intersect.

Targeting under Turkish data protection rules

Audience building on Meta in Turkey is constrained by Law No. 6698 on the Protection of Personal Data (KVKK), and the constraints bite before the campaign rather than during it.

Customer list audiences. Uploading a customer list to build a custom audience is a processing and transfer of personal data. You need a lawful basis for that specific use, and your privacy notice must actually cover it. A list collected for order fulfilment does not automatically carry permission to be used for advertising audience building.

Website and app audiences. Retargeting pools built from Pixel or SDK events depend on the consent position of the people in them, which is covered below and in our guide to cookie consent and tracking compliance in Turkey.

Children. Targeted advertising aimed at children that relies on profiling with personal data is prohibited following the amendment to the Commercial Advertising and Unfair Commercial Practices Regulation published in the Official Gazette on 1 July 2026. If a campaign’s audience cannot reliably exclude under-18s, the safe design is not to build it on profiling. The broader implications are in our guide to marketing to Gen Z and young consumers in Turkey.

Targeting transparency. The same July 2026 amendment requires advertisers running targeted advertising to tell consumers clearly and accessibly which criteria were used and how those criteria can be changed. That is a disclosure obligation on you, not only a platform setting.

The practical effect is that broad targeting with strong creative is often the better route in Turkey than narrow interest stacking — not because narrow targeting performs worse, but because the compliance and signal costs of maintaining it frequently exceed its benefit.

Pixel, Conversions API and the consent problem

Your conversion signal in Turkey will be incomplete, and the measurement plan has to assume that from the start. Browser-side tracking depends on consent, consent rates are never complete, and a platform optimising on partial signal will under-report and under-optimise.

Three things worth doing properly.

Implement server-side conversion sending alongside the browser pixel. Meta’s Conversions API improves signal durability, but it does not override consent: a server-side event about someone who refused tracking is still processing their data. Build the consent state into what you send rather than treating the API as a workaround. That distinction is the one that matters legally and the one most commonly got wrong.

Deduplicate and verify. Running both browser and server events without proper event identity matching produces inflated conversion counts, which flatters the channel and corrupts the optimisation.

Hold a source of truth outside the platform. Meta-reported conversions and your own order data will not agree. Decide in advance which is the commercial number, and report the platform figure as an optimisation input rather than a result. For brand effect, platform metrics cannot answer the question at all — that needs survey measurement, with the local considerations set out in our guide to brand tracking and consumer research in Turkey.

Creative that works in Turkish feeds

Creative does more of the work on Meta than targeting does, and localisation failures are visible in the first two seconds.

Write Turkish copy from a brief, not from a finished English ad. Turkish places the verb late and runs longer than English for the same content, so a translated headline arrives at its point in the wrong position and overruns the space it was designed for. Primary text, headline and description all need writing natively.

Design for Turkish text lengths. Static and video assets built around English word counts break when the copy is swapped. On-image text should be set in Turkish from the start rather than retrofitted.

Vertical first, sound-aware. Reels and Stories placements require vertical framing. Recovering a vertical crop from footage framed only for horizontal is a compromise viewers notice, so decide the output set before you shoot.

Localise the offer, not just the words. Price in lira, show payment presentation Turkish buyers expect, and make the call to action point at something they can actually complete. A beautifully localised ad leading to a global checkout is a wasted impression.

Test fewer, bigger creative differences. In a new market, variation between genuinely different concepts teaches you more than variation between two button colours, and it teaches it faster.

Budget and bidding when the currency moves

Turkey’s inflation and currency movement affect paid social planning in ways that catch out centrally managed budgets.

Set the budget in the currency you are accountable in, and watch the other one. A budget fixed in euros or dollars buys a changing amount of Turkish inventory over a year; a budget fixed in lira represents a changing cost to the parent company. Neither is wrong, but you need to know which one you are managing and report both.

Expect auction costs to move in nominal terms. Rising nominal media costs in a high-inflation economy are not automatically a performance deterioration, and a year-on-year cost per result comparison in nominal lira will mislead. Compare in real terms or against a competitive benchmark rather than against last year’s number.

Watch the value side of the ratio, not just the cost. If your product price rises with inflation while your cost per acquisition rises at a similar rate, the return is stable even though every input looks worse. The pricing decisions behind that sit in our guide to pricing strategy in Turkey.

Review budget cadence more often than you would elsewhere. Quarterly budget setting with no mid-period mechanism is a poor fit for this market. Agree a review trigger in advance.

Branded content and disclosure

Where a creator’s post is amplified as an ad, or where a creator receives any benefit for a post, Turkish disclosure rules apply to the content.

Following the 1 July 2026 amendment to the Commercial Advertising and Unfair Commercial Practices Regulation, the obligation is triggered by any benefit — payment, a discounted product or service, event participation — and the content must carry a clear label such as “reklam” or “tanıtım”. Sending a creator a free product in the hope of coverage is within scope.

Two operational points for paid amplification specifically. Confirm the disclosure is present in the asset itself before you put budget behind it, because amplifying undisclosed branded content extends the exposure rather than limiting it. And put the required wording in the brief and the contract rather than relying on the creator’s habit or on a platform toggle alone.

What to review weekly

Meta reports a great deal that does not answer whether the account is working. Five things are worth a standing weekly look in a Turkish account.

Placement-level cost per result. Given the Instagram and Facebook composition gap, a blended figure hides the decision you need to make.

Creative fatigue indicators. Frequency rising while click-through declines on the same asset is the signal to refresh, and it arrives faster on a small, concentrated audience than planners from larger markets expect.

Learning phase status and delivery stability. Ad sets stuck outside stable delivery are spending without optimising, and in a low-volume new market this is common.

Platform-reported conversions against your own order data. Track the gap as a ratio over time. A sudden change in the ratio usually means a tracking problem rather than a performance one.

Comment volume and sentiment under paid posts. Paid reach recruits people with no relationship to the brand, so comment volume and tone differ from organic. Unmoderated comment threads under an active campaign are a reputational exposure the media report will not show you.

Medyae’s digital marketing service covers paid media planning, execution and reporting of this kind, delivered as an ongoing programme rather than a one-off campaign build.

Frequently asked questions

Is Instagram or Facebook better for advertising in Turkey?

They reach different audiences, so the answer depends on your buyer. DataReportal’s Digital 2026 Turkey report puts Instagram’s potential ad reach at 62.3 million, covering 92.1% of adults 18 and over with an adult ad audience of 46.5% female and 53.5% male. Facebook’s reach is 34.7 million, covering 53.2% of adults, with an adult audience of 34.2% female and 64.4% male. Instagram is larger and more gender-balanced; Facebook still reaches over half of Turkish adults and is often cheaper for older or male-weighted audiences. Test placements separately rather than letting distribution run automatically.

Can we upload our customer list to build a Meta audience in Turkey?

Only with a lawful basis for that specific use under Law No. 6698 (KVKK), and with a privacy notice that actually covers advertising audience building. A list collected for order fulfilment does not automatically carry permission to be used this way. Take Turkish advice on your particular data before uploading.

Does the Conversions API get around consent requirements in Turkey?

No. Server-side conversion sending improves signal durability, but sending an event about someone who refused tracking is still processing their personal data. Build the consent state into what you send, and deduplicate against browser events to avoid inflated conversion counts.

Should we use Meta’s automated campaign types when entering Turkey?

Not at first. Automated products optimise by learning from conversion signal, and a new entrant typically has no Turkish conversion history, low volume and a signal weakened by consent choices. Start manual with separated placements and a conversion event that fires often enough to optimise against, then consolidate and automate once volume supports it.

Why do our Meta costs keep rising in Turkey?

Partly because nominal media costs rise in a high-inflation economy, which is not the same as performance deteriorating. Compare cost per result in real terms or against a competitive benchmark rather than against last year’s nominal lira figure, and check whether your own prices and order values rose at a similar rate — if they did, the return may be stable even though every input looks worse.

Conclusion

Meta is the largest paid social opportunity in Turkey, and the figures that matter are composition rather than headline reach. Instagram addresses 62.3 million with a near-balanced gender split; Facebook addresses 34.7 million with a markedly male-skewed one. Treating them as one placement and reading one blended result is what turns a workable account into an unexplainable one.

Four things to get right. Separate placements and report at placement level, at least until you have proved what your category needs. Settle the tax position and business asset ownership before the first campaign rather than after the first invoice. Build the consent state into your measurement rather than treating server-side sending as a way around it. And write Turkish creative from a brief, because on Meta the creative does more work than the targeting.

For how paid social fits the wider channel picture, our pillar guide to marketing in Turkey for foreign brands is the place to start.

This article is general information about paid social advertising in Turkey, and the figures cited carry the dates of the sources given. It is not legal or tax advice. Data protection obligations, advertising disclosure requirements and the tax treatment of advertising payments are fact-specific and should be confirmed with qualified Turkish advisers for your situation.