TikTok Advertising: What Works When the Algorithm Runs the Room

TikTok advertising works differently from every other paid social channel, and that difference is not cosmetic. The platform’s algorithm distributes content based on engagement signals rather than follower graphs, which means a well-made ad from a brand nobody has heard of can reach millions of people who would never have found it through search or retargeting. That is a genuine structural advantage for brands willing to make content that earns attention rather than buying it.

Done well, TikTok ads combine the reach mechanics of broadcast with the targeting precision of digital. Done badly, they are expensive interruptions in a feed designed to reward relevance. The difference between those two outcomes comes down to how well you understand the platform, not how much you spend.

Why TikTok Is a Different Kind of Advertising Problem

I spent a long time earlier in my career overvaluing lower-funnel performance channels. Cost per acquisition looked clean, attribution models felt reassuring, and the numbers made sense in a spreadsheet. What I missed, and what took me years to properly reckon with, is that a lot of what performance marketing gets credited for was going to happen anyway. The person who typed your brand name into Google after seeing your ad on TikTok does not show up in your TikTok attribution report. They show up as organic search revenue. The channel that created the intent gets none of the credit.

TikTok forces you to confront this problem directly because it is, structurally, a reach and awareness platform. You can run conversion campaigns on it, and you should test them, but the platform’s core value is its ability to put your brand in front of people who were not looking for you. That is demand creation, not demand capture. And demand creation is how you grow.

Think of it like a clothes shop. Someone browsing online who already knows your brand and searches for it is probably going to buy. But the person who walks past your window, stops, tries something on, and discovers you exist for the first time, that person is the one who expands your customer base. TikTok is the window. Most performance channels are the till.

If you want a broader view of how TikTok fits into a wider social strategy, the social media marketing hub at The Marketing Juice covers channel strategy, content planning, and paid social across platforms. This article focuses specifically on TikTok advertising: how the ad formats work, what creative actually performs, and how to build a testing approach that generates real learning rather than noise.

How TikTok’s Ad Formats Actually Work

TikTok Ads Manager offers several format types, and each one serves a different purpose in the funnel. Understanding the distinction matters because brands routinely use awareness formats to chase conversions and conversion formats to build awareness, then wonder why neither is working.

In-Feed Ads

In-Feed Ads appear in the For You page between organic content. They autoplay with sound, run up to 60 seconds, and support a call-to-action button. This is the workhorse format for most advertisers and the one where creative quality has the most leverage. Because the ad sits alongside organic content, anything that looks or feels like a traditional commercial is immediately identifiable as an ad, and users scroll past it without a second thought.

The best In-Feed Ads are indistinguishable from organic content in their first two seconds. They open with a hook that earns the next three seconds, which earns the next ten, and so on. The format rewards storytelling pacing, not production budget.

TopView

TopView is the first ad a user sees when they open the app. It plays for up to 60 seconds with sound on and no competing content. It is expensive relative to In-Feed placements and works best for brand moments: product launches, campaign activations, or awareness pushes where guaranteed reach justifies the premium. For most brands running ongoing paid activity, TopView is a supplement rather than a foundation.

Branded Hashtag Challenges

Branded Hashtag Challenges invite users to create content around a theme or action associated with your brand. When they work, they generate enormous volumes of user-generated content and organic reach that extends far beyond the paid placement. When they do not work, they are an expensive lesson in the difference between a brief that sounds good in a meeting and a brief that actually motivates people to participate.

The campaigns that succeed here tend to have a simple, repeatable action at their core: a dance, a transformation, a before-and-after. Complexity kills participation. The more you ask users to do, the fewer will do it.

Spark Ads

Spark Ads let you boost existing organic TikTok posts, either your own or a creator’s, as paid content. The post retains its original engagement metrics (likes, comments, shares) and links back to the original account. This is one of the most underused formats in TikTok advertising, and it is particularly effective when you have identified a piece of organic content that is already performing well. You are not guessing whether the creative works. The organic data tells you.

Spark Ads also work well in creator partnership campaigns. Rather than producing a separate ad, you amplify the creator’s own post, which preserves the authenticity that made it perform in the first place.

Collection Ads and Shopping Ads

TikTok has built out its commerce capabilities significantly, and Collection Ads and Shopping Ads are designed for brands with product catalogues. Collection Ads open an instant gallery page within the app when tapped, reducing friction in the path to purchase. Shopping Ads integrate with TikTok Shop and allow users to buy without leaving the platform. For e-commerce brands, these formats are worth testing seriously, particularly in categories where impulse purchase is a real behaviour pattern.

Creative Is the Targeting

On most paid social platforms, targeting does a significant amount of the heavy lifting. You define your audience, set your parameters, and the platform delivers your ad to the people you specified. TikTok does not work this way, at least not primarily.

TikTok’s algorithm learns from engagement signals: watch time, replays, shares, comments, profile visits. It distributes content to users who have engaged with similar content before, regardless of whether they match your demographic targeting. In practice, this means the creative itself determines who sees your ad. Make content that resonates with 25-year-old women interested in skincare, and the algorithm will find them. Make content that resonates with nobody, and no amount of targeting precision will save you.

I have seen this play out repeatedly across categories. A client in a considered-purchase category was running TikTok ads with tightly defined audience parameters and wondering why reach was low and CPMs were high. We broadened the targeting significantly and focused the budget on improving the creative. Within three weeks, the algorithm had self-selected an audience that was converting at a better rate than the manually targeted campaigns. The creative was doing the targeting work.

This has a practical implication: your creative team is your most important TikTok investment. Not your media team, not your targeting strategy, not your bid management. Creative.

What High-Performing TikTok Creative Actually Looks Like

The single most common mistake brands make on TikTok is making content that looks like an ad. Polished production, branded endframes, voiceover narration, stock footage: all of these signal “advertisement” to a user who has been trained by thousands of hours of TikTok to recognise and dismiss them instantly.

High-performing TikTok creative tends to share several characteristics. It opens with a hook in the first one to two seconds that creates curiosity, poses a question, or shows something unexpected. It uses native formats: talking-head videos, duets, stitches, text overlays, trending audio. It feels like something a person made, not something a production company made. And it earns continued watch time by delivering value, entertainment, or genuine information, not by being loud.

The hook is the single most important element. TikTok’s own internal data consistently shows that the majority of viewers who will watch more than half of a video decide within the first two seconds whether to keep watching. If your first two seconds are a branded logo animation or a slow pan across a product, you have already lost most of your audience.

Effective hooks tend to fall into a few categories: a bold claim (“I spent £10,000 testing TikTok ads so you don’t have to”), a visual disruption (something unexpected or counterintuitive in the frame), a direct address (“If you’re running paid social and not doing this, stop”), or a problem statement that immediately resonates with the target audience. The hook does not need to be dramatic. It needs to be relevant and specific.

The Role of User-Generated Content and Creator Partnerships

User-generated content (UGC) and creator partnerships are not a nice-to-have on TikTok. For most brands, they are the most efficient path to creative that actually performs. Creators who have built audiences on TikTok understand the platform’s grammar in a way that most brand creative teams do not. They know what hooks work, what pacing keeps people watching, and what tone their specific audience responds to.

The most effective creator partnerships are not influencer campaigns in the traditional sense. You are not looking for someone with the largest following to hold your product and say something nice about it. You are looking for creators whose content style, audience, and authentic interests align with what your brand actually does. A creator with 80,000 followers and a highly engaged niche audience will often outperform a creator with 2 million followers and a diffuse one.

When briefing creators, the instinct to over-specify is almost always wrong. Give them the outcome you need (awareness of a specific product feature, traffic to a landing page, a particular offer to communicate) and let them determine how to deliver it in their own voice. The moment you hand them a script, you have made a brand ad with a creator’s face on it. That is not the same thing.

Campaign Structure and Bidding Strategy

TikTok Ads Manager follows a familiar three-tier structure: campaign, ad group, and ad. Campaign level sets the objective. Ad group level sets the audience, placement, budget, and bid strategy. Ad level is where the creative lives.

Campaign objectives map roughly to funnel stages: reach and video views for awareness, traffic and app installs for consideration, conversions and catalogue sales for lower-funnel activity. Choosing the right objective matters because it determines how the algorithm optimises delivery. If you select a conversion objective but your pixel is not firing reliably or your conversion volume is too low for the algorithm to learn from, you will get poor delivery and inflated CPAs. In those cases, a traffic objective with strong landing page quality is often more effective until you have enough conversion data for the algorithm to work with.

Bidding: Lowest Cost vs. Cost Cap vs. Bid Cap

Lowest Cost (formerly “automatic bidding”) tells the algorithm to spend your budget and get as many results as possible at the lowest achievable cost. It is the right starting point for most campaigns because it gives the algorithm maximum flexibility to learn. Cost Cap sets a target cost per result and the algorithm tries to stay near that target while spending your budget. Bid Cap sets a hard ceiling on what you will pay per click or impression, which limits reach but controls cost more tightly.

For most advertisers starting on TikTok, Lowest Cost with a daily budget gives the algorithm the data it needs to optimise. Move to Cost Cap once you have established a baseline CPA and want to manage efficiency at scale. Bid Cap is rarely the right choice for new campaigns because it tends to restrict delivery before you have enough data to set a meaningful ceiling.

Audience Targeting: How Much to Specify

As noted above, TikTok’s algorithm is unusually capable of finding the right audience through content signals rather than demographic parameters. That said, some targeting inputs are worth using: interests and behaviours can help seed the algorithm in the right direction, particularly for new campaigns with no prior data. Custom audiences (retargeting website visitors, customer lists, video viewers) are effective for mid-funnel and lower-funnel campaigns. Lookalike audiences built from high-value customer lists can extend reach efficiently.

What you should avoid is over-targeting. Narrow audience definitions on TikTok tend to restrict delivery, inflate CPMs, and prevent the algorithm from finding the pockets of audience it would have found on its own. Start broader than feels comfortable and let the creative do the qualification work.

Measurement and Attribution: What the Numbers Are Actually Telling You

Attribution on TikTok is genuinely complicated, and I say that as someone who has spent a career looking at attribution models with a degree of professional scepticism. The platform’s default attribution window (7-day click, 1-day view) will capture some of the value TikTok generates, but it will miss a significant portion. Users who see your ad, search your brand name two days later, and convert through Google are counted as organic search conversions. The TikTok ad that created the intent gets nothing.

This is not a TikTok-specific problem. It is the fundamental limitation of last-click attribution in a multi-touchpoint world. But it is more acute on TikTok because the platform sits further up the funnel for most categories, which means the gap between exposure and conversion is longer and more likely to involve other touchpoints.

A few approaches help. Brand search volume is one of the cleanest signals of TikTok’s impact: if your campaigns are generating awareness, you should see branded search queries increase. Incrementality testing, where you hold out a portion of your target audience from TikTok exposure and compare conversion rates, is the most rigorous approach but requires enough volume to be statistically meaningful. Marketing mix modelling, when done properly, can attribute TikTok’s contribution to overall revenue growth even when last-click attribution cannot.

The honest position is that TikTok’s in-platform reporting will understate the channel’s true contribution for most brands. That does not mean you should ignore it. It means you should treat it as a floor, not a ceiling, and look for corroborating evidence in other data sources before making budget decisions based on ROAS alone.

When I was at iProspect, we managed hundreds of millions in ad spend across dozens of categories. The clients who made the best long-term decisions were the ones who understood that their attribution tools were giving them a perspective on reality, not reality itself. The ones who made the worst decisions were the ones who optimised purely for what their dashboards could measure, and gradually starved the channels that were building their brand while feeding the channels that were harvesting the demand those brand-building channels had created.

Building a TikTok Testing Framework That Generates Real Learning

The brands that win on TikTok are not the ones that found the one viral creative concept and scaled it. They are the ones that built systematic testing processes that generate consistent learning and allow them to improve creative quality over time.

A useful testing framework has three components: a hypothesis, a variable, and a sample size large enough to draw a meaningful conclusion. “Let’s try a different hook” is not a hypothesis. “Opening with a problem statement will generate higher watch time than opening with a product shot because TikTok users respond to content that addresses their context before introducing a brand” is a hypothesis. The difference matters because only the second version tells you something you can act on.

What to Test and in What Order

Hook is the highest-leverage variable to test first. Small changes to the first two seconds can produce large differences in watch time and completion rate, which flow through to every downstream metric. Test hooks before testing anything else.

After hooks, test format: talking head versus voiceover, creator-style versus polished production, text-overlay-heavy versus dialogue-driven. Then test offers and calls to action. Then test audio (trending sounds versus original audio versus no music). Work through variables systematically rather than changing multiple elements simultaneously, which makes it impossible to know what drove the difference.

Volume matters. TikTok creative fatigue is real and it happens faster than on most other platforms. A creative that performs well in week one may be exhausted by week three, particularly if your target audience is not enormous. Build a pipeline of creative variants rather than a single “winning” ad, and plan for regular creative refresh as a standard part of your campaign management process.

Using TikTok’s Creative Tools

TikTok offers several tools within Ads Manager that are worth using. Creative Center provides trend data on sounds, hashtags, and creative formats that are currently performing well on the platform. It is a useful input for creative briefs, particularly if your team is not native TikTok users. Symphony Creative Studio uses AI to generate video ad variants from product information and existing assets. The output quality varies, but it can accelerate the volume of creative variants you have available for testing. Video Insights shows engagement data at a granular level, including where viewers drop off, which is directly useful for diagnosing hook and pacing problems.

Tools like these, alongside broader social media marketing tools that help with scheduling, analytics, and content planning, can meaningfully reduce the operational overhead of running TikTok campaigns at scale. The caveat is that tools do not replace creative judgment. They accelerate it.

TikTok Advertising for Different Business Types

TikTok advertising is not equally well-suited to every business model, and being honest about that saves a lot of wasted budget.

E-Commerce and Direct-to-Consumer

This is where TikTok advertising has the strongest track record. Impulse-friendly categories, visually demonstrable products, and audiences that skew younger tend to perform well. TikTok Shop has lowered the friction between discovery and purchase significantly, and brands that have integrated their catalogues and invested in creator partnerships are seeing meaningful returns. The challenge is that competition in popular categories is increasing, which is pushing up CPMs. Creative differentiation matters more than it did two years ago.

B2B and Considered-Purchase Categories

TikTok’s audience skews younger and the platform’s nature rewards entertainment and immediacy. That does not make it useless for B2B or considered purchases, but it changes what success looks like. Conversion campaigns optimising for immediate purchase are unlikely to work in a category where the sales cycle is six months long. Awareness and brand-building campaigns that put you in front of future buyers before they are in-market are more defensible.

There is also a genuine audience development opportunity in B2B. Decision-makers are on TikTok. Senior professionals use the platform. The question is whether your category and your creative can earn their attention in a context where they are not in work mode. That is a harder brief to write, but it is not an impossible one.

Local and Regional Businesses

Geographic targeting on TikTok is available and functional, but the platform’s minimum audience sizes for efficient delivery mean that very local campaigns (a single city or postcode area) can struggle to generate enough volume. For local businesses, TikTok organic content often generates better returns than paid, at least initially. Building an organic presence that generates local awareness and then using paid to amplify the best-performing content is a more efficient starting point than running paid campaigns cold.

Common Mistakes That Waste TikTok Ad Budget

After seeing a lot of TikTok campaigns from the inside, the failure modes tend to cluster around a few consistent errors.

Repurposing content from other platforms is the most common. An Instagram Reel cut down from a television commercial, a LinkedIn video with a logo slapped on it, a YouTube pre-roll with the first five seconds trimmed: none of these are TikTok ads. They are content that was made for a different platform and a different audience context, and TikTok users identify them as such within seconds. Native content made for TikTok consistently outperforms repurposed content, even when the repurposed content has higher production values.

Optimising for the wrong metric is the second most common error. Watch time and completion rate are the leading indicators of creative quality on TikTok. Click-through rate is a useful signal but it is influenced heavily by the call-to-action placement and copy, not just the creative quality. ROAS as a primary optimisation metric leads to under-investment in awareness campaigns that are building long-term brand value. Choose metrics that match your campaign objective, not metrics that are easy to report.

Underfunding the learning phase is a structural mistake that is particularly common among brands new to the platform. TikTok’s algorithm needs data to optimise, and data costs money. Running campaigns at very low daily budgets extends the learning phase, produces unreliable data, and makes it impossible to draw conclusions about what is working. It is better to run fewer campaigns at adequate budget levels than many campaigns at budgets too low for the algorithm to learn from.

Neglecting creative refresh is a slow-burn budget drain. Creative fatigue on TikTok happens faster than most advertisers expect. Frequency caps help, but they do not eliminate the problem. Building a regular creative production cadence into your campaign planning from the start, rather than treating it as something you will address when performance drops, is the more sustainable approach.

How TikTok Fits Into a Broader Channel Mix

TikTok advertising does not exist in isolation, and the brands that treat it as a standalone channel tend to get less from it than the brands that think about how it interacts with everything else they are doing.

The most common interaction is with branded search. TikTok awareness campaigns drive branded search volume, which drives Google conversions, which get attributed to Google. If you are running TikTok campaigns and not monitoring branded search volume as a proxy metric, you are missing a significant part of the picture. A sustained increase in branded search queries following a TikTok campaign launch is strong evidence that the campaign is working, even if the in-platform ROAS looks modest.

TikTok also interacts with email and CRM. Users who discover your brand through TikTok and visit your site but do not convert immediately are retargetable through email capture and through pixel-based retargeting on other platforms. The TikTok ad that introduced them to the brand may have been the first touch in a sequence that ends with a purchase driven by a promotional email three weeks later. Understanding that sequence, even imperfectly, shapes better budget decisions.

A useful framework for thinking about channel mix is to separate channels by their primary function: reach and awareness, consideration and engagement, or conversion and retention. TikTok sits most naturally in the first two categories for most brands. Treating it as a conversion channel and judging it by conversion metrics alone is a category error that leads to premature budget cuts.

For a more complete picture of how social channels work together and how to build a coherent social strategy rather than a collection of disconnected channel tactics, the social media marketing section of The Marketing Juice covers channel planning, content strategy, and measurement frameworks across the major platforms.

Thinking about social media marketing more broadly, resources like Semrush’s social media strategy guide are useful for understanding how to build a channel strategy that connects to business objectives rather than just activity metrics. And if you are thinking about how TikTok sits alongside platforms like Instagram, HubSpot’s overview of Instagram marketing tools gives a sense of the comparative toolset available across platforms.

For brands managing content planning across multiple platforms, Sprout Social’s social media calendar is worth looking at as an operational tool for keeping creative production and publishing on track. And for teams thinking about international expansion, this Search Engine Land piece on international social media marketing raises questions about platform relevance by market that apply directly to TikTok, which has very different penetration rates across geographies.

What a Realistic TikTok Advertising Budget Looks Like

TikTok’s minimum campaign budget is low enough that almost any business can technically run ads on the platform. Whether that budget is sufficient to generate meaningful learning is a different question.

As a rough guide, a campaign budget of less than £50 to £100 per day per ad group is unlikely to generate enough data for the algorithm to optimise effectively in most markets. That does not mean you need a large overall budget, but it does mean you should concentrate it rather than spreading it thinly across many campaigns and ad groups simultaneously.

Creative production is the budget line that most brands underestimate. If you are producing native TikTok content properly, whether through an internal team, a UGC creator network, or a specialist production partner, you need enough creative volume to support ongoing testing and to manage fatigue. Budgeting for media without budgeting adequately for creative is one of the most consistent mistakes I have seen across paid social channels, and it is particularly costly on TikTok where creative is doing so much of the strategic work.

A reasonable starting allocation for a brand entering TikTok advertising seriously might be 60 to 70 percent of the total budget on media and 30 to 40 percent on creative production. That ratio feels uncomfortable for teams used to traditional paid social where creative costs are lower, but it reflects the reality of what drives performance on this platform.

The Honest Assessment: Is TikTok Advertising Right for Your Brand?

Not every brand should be running TikTok ads right now. That is not a popular thing to say in a piece about TikTok advertising, but it is true.

TikTok advertising requires a genuine creative commitment. If your organisation cannot produce native-feeling content at volume, either because of internal capability, brand guidelines that prevent the necessary informality, or approval processes that take three weeks, the platform will consistently underperform for you. The issue is not TikTok. It is the mismatch between what the platform requires and what your organisation can deliver.

It also requires patience with measurement. If your stakeholders need clean, attributable ROAS from every channel to justify continued investment, TikTok will be a difficult sell. The channel’s contribution is real, but it is not always visible in last-click attribution, and building the case for its value requires a more sophisticated measurement conversation than most organisations are used to having.

Early in my career, I would have been the person pushing every client onto every new platform as soon as it showed promise. Twenty years of watching campaigns succeed and fail has made me more selective. The question is not whether TikTok advertising works. It does, for the right brands with the right creative approach and the right measurement framework. The question is whether it is the right investment for your specific business at this moment, given your creative capabilities, your audience, your category, and your measurement maturity. That is a more useful question than “should we be on TikTok?”

For brands where the answer is yes, the opportunity is real. TikTok’s algorithm is genuinely democratic in a way that most advertising platforms are not. A well-made ad from a brand nobody has heard of can reach a million people who needed to know it existed. That is not a small thing. It is, in fact, exactly the kind of reach that grows businesses.

The brands that will look back on this period as a competitive advantage are the ones investing in creative capability now, building testing systems that generate real learning, and treating TikTok as a brand-building channel with measurable downstream effects rather than a performance channel judged by metrics it was never designed to optimise for.

About the Author

Keith Lacy is a marketing strategist and former agency CEO with 20+ years of experience across agency leadership, performance marketing, and commercial strategy. He writes The Marketing Juice to cut through the noise and share what works.

Frequently Asked Questions