Are You Spying on Your Competitors' Native Ad Campaigns?

Our spy tools monitor millions of native ads from over 60+ countries and thousands of publishers.

Get Started

Advertising platforms are becoming more automated. Machine-learning systems increasingly influence bidding, audience discovery, placement selection, and creative delivery, while generative AI is making it easier to produce and adapt advertising assets at scale.

For marketers, this creates a new challenge in competitive research. A competitor's advertising strategy may no longer be represented by one fixed creative or a single campaign configuration. The messages, formats, and combinations visible in the market can change over time.

That does not mean advertising networks have become large language models. Modern ad platforms combine different technologies, including predictive models, recommendation systems, automated bidding, and generative tools. Some AI systems work with language, but the broader advertising stack is more complex than an LLM.

The practical implication is that competitive research needs to become more systematic. Instead of saving a few screenshots and assuming they represent a competitor's entire strategy, marketers should monitor creative patterns, messaging changes, landing-page experiences, and the wider advertising environment.

Tools such as Anstrex can contribute to this process by helping marketers investigate visible advertising activity across supported networks. The objective is not to reverse-engineer a competitor's private algorithm. It is to understand the market signals that can be observed—and turn those observations into better creative and campaign hypotheses.

The New Reality — Ad Networks Are Becoming AI-Driven, Not Just Inventory Platforms

Advertising technology has evolved beyond a simple system of choosing placements and adjusting bids.

Modern platforms use automated systems to help advertisers reach business goals. Depending on the platform and campaign type, these systems can influence bidding, budget allocation, placement selection, audience expansion, and creative combinations.

The advertiser may provide a conversion goal, budget, audience guidance, and creative assets. The platform then uses its available data and optimization systems to determine how to deliver the campaign.

An overview of AI in AdTech by illumin discusses how artificial intelligence is being applied across advertising workflows, including optimization, targeting, and creative operations.

This is an important shift, but it is worth distinguishing the technologies involved. Automated bidding, predictive audience models, recommendation systems, and generative AI are related parts of the modern advertising ecosystem; they are not interchangeable terms.

From manual campaign management to automated optimization

Traditional media buying often involved frequent manual decisions:

  • Adjusting bids
  • Changing budgets
  • Selecting placements
  • Refining audience settings
  • Rotating creative assets
  • Reviewing performance reports

Today, many advertising platforms automate some or all of these tasks.

Advertisers increasingly define objectives and constraints while platform systems handle parts of the delivery and optimization process.

This can make campaign management more efficient, but it also makes the underlying decision process less transparent. Advertisers may see the campaign results without having access to every signal or decision that influenced delivery.

Creative is becoming part of the optimization system

AI is also changing how advertising creative is developed and delivered.

Instead of treating an advertisement as one fixed file, some platforms allow advertisers to provide multiple headlines, descriptions, images, videos, and calls to action. The system can then assemble or select combinations according to its supported features and optimization logic.

This is connected to Dynamic Creative Optimization (DCO), which can test or deliver different creative combinations.

The illumin AI in AdTech overview discusses the broader role of AI in advertising and creative optimization.

However, not every DCO system uses an LLM to rewrite creative in real time. Some systems select among pre-supplied assets or use predictive models to optimize combinations. Generative capabilities vary by platform and product.

For competitive researchers, the important development is that the creative visible at one moment may not represent every variation a competitor is testing or delivering.

Conversational discovery adds another advertising environment

Consumer discovery is also changing as people use AI assistants and conversational search experiences to research products and services.

A PPC-focused analysis from Real FiG Advertising & Marketing discusses how AI chat experiences may affect traditional search behavior.

This creates a new research question for marketers: where are commercial messages appearing as AI-powered discovery develops, and how can those placements be measured?

The answer depends on the platform and its advertising products. Not every AI answer contains an ad, and advertising availability, formats, and eligibility can change.

A report about Similarweb's AI Ads dataset, covered by Search Engine Watch, describes efforts to measure brand and advertising visibility in AI-powered experiences. Any reported coverage figures should be checked against the original dataset, its definitions, and the specific date of measurement before publication.

The wider industry is adapting to these changes. AdExchanger's coverage of marketers getting used to AI in the ad stack discusses how AI is becoming part of the advertising workflow.

The strategic takeaway is not that every ad network is an LLM. It is that automation and AI are increasingly involved in how advertising is created, delivered, and measured.

For competitive intelligence, that means a single screenshot is an incomplete representation of a campaign. Teams need to study patterns over time and distinguish visible evidence from assumptions about the systems behind it.

Why Old-School Competitive Spying Fails in an AI-Driven Advertising Stack

Many marketers still approach competitor research as if advertisements were static objects.

They save screenshots from social platforms, inspect search ads, review auction insights, and collect examples from video and display placements.

These activities remain useful. The problem is assuming that a small collection of visible advertisements reveals the competitor's complete campaign strategy.

AI-driven advertising makes that assumption increasingly unreliable.

The three assumptions behind traditional ad spying

Traditional competitor research often relies on three assumptions:

  1. Creative assets are relatively fixed. A screenshot is treated as a representative version of the campaign.
  2. Targeting logic is visible through manual settings. Audience selections, keywords, and placement choices are assumed to explain who sees the ad.
  3. Optimization happens on a human schedule. Campaign changes are expected to occur through daily or weekly media-buyer decisions.

Modern automated advertising can challenge all three assumptions.

1. The creative you see may not be the only creative being delivered

Some advertising systems allow multiple creative assets and combinations to be tested or selected automatically.

A competitor might provide several headlines, images, and descriptions, while the platform determines which combinations to deliver in different circumstances.

Generative AI can add another layer where supported, but the capabilities vary across platforms.

As discussed in illumin's AI in AdTech overview, AI is being applied to creative and optimization workflows.

This means that the advertisement captured in a public library or research tool may be only one visible example. It should not automatically be treated as the campaign's only execution.

2. Audience and delivery decisions are harder to observe

Platforms such as Google and Meta increasingly offer automated campaign products, including Performance Max and Advantage+.

These products can use advertiser inputs and platform signals to make delivery decisions. Depending on the campaign, advertisers may provide creative assets, conversion goals, audience guidance, and other controls while the system handles more of the optimization.

MarTech's analysis of how AI is making creative a new targeting signal discusses the relationship between creative and automated audience delivery.

Creative can help communicate the intended audience and influence engagement. But it does not reveal the platform's complete targeting logic.

A competitor's headline, visual, or script may suggest the customer situation they want to address. It cannot tell you exactly which audience settings, behavioral signals, or model predictions caused the ad to be delivered.

3. Optimization can happen faster than manual research cycles

Automated advertising systems can adjust delivery based on ongoing performance signals.

The illumin AI in AdTech overview discusses how AI can support optimization and decision-making within advertising systems.

This makes competitor campaigns more difficult to reconstruct from occasional observations.

An advertiser may test new creative, adjust budgets, change placements, or modify campaign settings while a researcher is still collecting examples.

However, visible changes do not prove that an autonomous agent made the decision. A human marketer may also have introduced the variation.

Conversational placements make the research problem more complex

Traditional ad research often focuses on recognizable environments: search results, social feeds, video placements, and display inventory.

AI-powered discovery adds more varied experiences, including conversational answers and recommendation interfaces.

If sponsored placements appear within an AI-generated experience, competitive research may need to consider the prompt, the generated answer, the sponsored unit, and the destination page.

The Search Engine Watch coverage of Similarweb's AI Ads dataset discusses measurement of advertising visibility in AI-powered experiences.

This is different from traditional search-engine results-page monitoring. The format and availability of sponsored placements depend on the specific platform and product.

Static screenshots are snapshots, not campaign blueprints

A screenshot can tell you what was visible at a particular time and in a particular context.

It cannot independently reveal:

  • The advertiser's complete creative library
  • Private targeting settings
  • Bid and budget decisions
  • The platform's internal optimization logic
  • The conversion performance of that specific variation
  • Whether the creative was part of an automated or manual test

That does not make traditional ad research obsolete. It means the research process should become more careful.

Instead of trying to reconstruct an entire hidden campaign from a few visible outputs, marketers should track observable creative and landing-page patterns over time, then use those patterns to develop testable hypotheses.

Creative Is an Important Optimization Signal — And a Valuable Source of Competitive Intelligence

Creative is not only the message a customer sees. In many automated advertising systems, the assets supplied by an advertiser can also influence how the platform interprets and delivers a campaign.

This makes creative an important research signal.

As advertisers provide different headlines, images, videos, benefits, and calls to action, those assets communicate which customer problems and outcomes the business wants to emphasize.

MarTech's analysis of creative as a new targeting signal explores this changing relationship between advertising creative and automated delivery.

The distinction is important: creative can provide context to an optimization system, but it does not reveal the exact audience the platform has selected or the private logic behind its delivery decisions.

Creative variations can suggest changing campaign priorities

Imagine a competitor that begins emphasizing "simpler onboarding" more frequently.

That change could suggest several possibilities:

  • The company is testing an implementation-focused message.
  • Customer feedback has made onboarding a more important theme.
  • The advertiser is targeting a different use case.
  • The marketing team is refreshing its positioning.
  • An automated system is selecting among previously supplied assets.

The visible change is a useful observation. The reason behind it remains a hypothesis until supported by additional evidence.

This is where systematic creative tracking becomes valuable.

Look for patterns across creative dimensions

Instead of examining each ad in isolation, organize competitor creatives around several dimensions.

Problem framing

Which customer problems appear repeatedly?

Examples might include wasted advertising spend, manual reporting, slow onboarding, or inventory management challenges.

Implied audience

Who does the creative appear to address?

Examples include founders, procurement teams, IT leaders, agency owners, or first-time buyers.

These are interpretations of the message, not confirmed audience-targeting settings.

Outcome language

Which benefits are being emphasized?

  • Save time
  • Reduce costs
  • Improve efficiency
  • Reduce risk
  • Increase revenue
  • Simplify implementation

Offer mechanics

How is the advertiser presenting the offer?

  • Free audit
  • Trial
  • Bundle
  • Guarantee
  • Financing
  • Consultation
  • Discount

Grouping ads by these dimensions helps reveal how competitors communicate their offers and which themes appear across their visible advertising.

Format changes are worth investigating

Creative formats can also provide useful research signals.

For example, a competitor might shift from polished studio photography to founder-led videos, screen recordings, or customer-style demonstrations.

Possible explanations include:

  • A new creative testing direction
  • A different campaign objective
  • A change in the intended audience
  • A seasonal campaign
  • A response to changing platform formats
  • A creative refresh

Do not assume that a format change proves the platform discovered a more profitable audience. That information is not available from the creative alone.

Conversational discovery makes message clarity important

As consumers use AI assistants and conversational search to research products, clear and specific messaging becomes increasingly valuable.

The Real FiG PPC analysis of ChatGPT's impact on traditional search discusses changes in the relationship between informational and commercial search behavior.

For advertisers, this creates an opportunity to review whether their messaging clearly explains:

  • What problem the product solves
  • Who it is designed for
  • Which benefits it provides
  • What makes it different
  • What evidence supports its claims

Where advertising is delivered in conversational or AI-assisted environments, the available formats and requirements depend on the specific platform.

The Search Engine Watch report on Similarweb's AI Ads dataset is an example of the developing measurement landscape for sponsored visibility in AI experiences.

Creative intelligence is not mind-reading

A competitor's creative can reveal the themes and offers it chooses to communicate publicly.

It cannot establish:

  • The competitor's private audience composition
  • Which customer segment produced the best conversion rate
  • The platform's internal model weights
  • Whether a particular hook was selected by AI or a human
  • The advertiser's actual return on investment

The strongest approach is to treat each visible creative change as a research clue.

Capture it, categorize it, compare it with earlier observations, and investigate whether it represents a meaningful shift in the market.

How to Research AI-Optimized Competitors in Practice

Researching AI-optimized competitors requires a shift from collecting isolated ads to building a repeatable record of visible market activity.

The goal is not to reverse-engineer the private media plan or reconstruct a platform's internal model. It is to understand how competitors communicate, which advertising environments they appear in, and how their visible creative changes over time.

A practical workflow can be built around five steps.

Step 1: Identify the advertising environments you want to monitor

Start by listing the platforms and advertising environments relevant to your market.

Depending on your industry, this could include:

  • Search advertising
  • Social advertising
  • Native advertising
  • Display advertising
  • Video advertising
  • AI-assisted search and recommendation experiences

The availability of public ad information varies by platform. Some provide searchable ad libraries or transparency resources, while other environments may offer limited visibility.

The Search Engine Watch report on Similarweb's AI Ads dataset discusses one approach to measuring sponsored visibility in AI-powered environments.

Use such sources to understand what can currently be measured. Do not assume every AI assistant or answer experience supports advertising or exposes the same information.

Step 2: Build a consistent prompt and observation set

If you are researching AI search experiences, create a standardized set of questions relevant to your category.

Include different levels of commercial intent.

Exploratory prompts

  • Best project management tools for agencies
  • How can a small business improve inventory management?
  • What should a first-time buyer look for in a home security system?

Comparison prompts

  • Compare project management platforms for a 20-person agency
  • Which inventory tools offer better reporting?
  • What are the alternatives to [Competitor]?

Commercial prompts

  • Buy project management software for a 20-person agency
  • Affordable overnight flower delivery
  • Best local service provider for an urgent repair

These are research examples, not guaranteed to trigger sponsored placements.

Record the exact wording, platform, date, and available response details. Where a sponsored unit is actually displayed and can be lawfully observed, record the creative and destination.

Avoid collecting private customer conversations or personal information without appropriate authorization and privacy safeguards.

Step 3: Capture and categorize visible creative

For every relevant ad or creative example, record:

  • Brand or advertiser
  • Date observed
  • Platform or network
  • Headline and message
  • Creative format
  • Customer problem addressed
  • Implied audience
  • Main benefit or outcome
  • Offer structure
  • Landing-page destination

Then group the examples by recurring themes.

For instance, a collection of competitor ads may show repeated emphasis on:

  • Faster implementation
  • Reduced manual work
  • Lower costs
  • Better reporting
  • Specialist support

This allows the team to study message patterns without assuming that every creative represents a separate campaign or a proven winning variation.

Step 4: Track visibility consistently

Traditional auction insights can help marketers understand competitor overlap in particular search environments.

For AI-assisted experiences, the research question may be different: which brands appear for a standardized set of commercial prompts, and how does that change across repeated observations?

If the platform and available data support it, record:

  • Which brands appear in sponsored placements
  • Which brands appear in organic answers
  • Which sources are cited
  • What message is displayed
  • Which destination page is linked
  • Whether the same prompt produces a different result over time

A simple research measure can be calculated:

Observed appearance rate = Number of tracked observations containing the brand ÷ Number of eligible observations

This should be described as an observed appearance rate for your particular prompt set and testing conditions—not universal AI share of voice.

Results may vary by location, account, platform, date, personalization, and prompt wording. Keep those conditions consistent where possible.

Step 5: Investigate the wider advertising infrastructure carefully

Modern advertisers may use different combinations of analytics, measurement, identity, creative, and media-buying systems.

Public announcements and industry coverage can help identify developments in this ecosystem.

For example, AdExchanger's Q2 2026 coverage of AI media scaling discusses developments in AI-related advertising and media infrastructure.

When reviewing a competitor's public advertising, look for observable clues such as:

  • Consistent landing-page tracking conventions
  • Product feeds or structured product information
  • Comparison tables and FAQs
  • Repeated message themes across channels
  • Clear connections between ads and destination pages

These observations may suggest a coordinated marketing approach, but they do not prove which vendors, integrations, or AI systems a competitor uses.

The goal is to understand the visible customer experience and the consistency of the advertiser's messaging—not to guess its private technology stack.

Turning Anstrex Into an AI-Aware Creative Intelligence Radar

Anstrex is most useful when it is treated as a competitive advertising research tool rather than a simple screenshot collection.

The objective is to study visible advertising patterns across supported sources and use those observations to understand how competitors communicate their offers.

In an AI-driven advertising environment, this means looking beyond one apparently successful creative.

A more useful approach is to monitor how headlines, hooks, angles, formats, and landing pages change over time.

Anstrex does not reveal a competitor's private AI models, internal optimization decisions, or actual campaign profitability. It can, however, help marketers investigate visible creative and destination-page examples.

1. Move from isolated screenshots to structured sampling

Instead of saving only the ads that look interesting, build a consistent research collection.

Where the available Anstrex features and export options support it:

  • Review competitor creatives across different observation periods.
  • Compare recently observed ads with earlier examples.
  • Tag creative by promise, proof, visual style, audience language, and offer.
  • Group ads by problem or apparent buying stage.
  • Save relevant landing-page examples for comparison.

For example, a creative could be tagged with:

  • Promise: Save time
  • Proof: Product demonstration
  • Visual: Screen recording
  • Audience language: Agency owners
  • Offer: Free consultation

This makes it easier to identify recurring patterns across a competitor's visible advertising.

MarTech's article on AI-native advertising provides broader context on the use of AI in creative and campaign workflows.

Do not assume that a competitor's visible ads are generated by AI. The advertiser may be using conventional creative production, automated asset selection, generative tools, or a combination.

2. Analyze the creative patterns, not imagined model decisions

Once the examples are categorized, examine how the messages change.

Three useful analyses include:

Message changes

Track when certain benefits, phrases, or customer concerns appear more frequently.

For example, a competitor may begin emphasizing "done-for-you setup" more often.

That could suggest a change in positioning or a new test. It does not prove an AI system identified implementation anxiety as a profitable audience segment.

Format changes

Compare shifts such as:

  • Static image to short-form video
  • Studio photography to product demonstrations
  • Founder-led messaging to customer testimonials
  • Feature lists to problem-focused storytelling

A format shift can be worth investigating, especially if it appears across multiple observations.

However, it may reflect a creative refresh, a campaign launch, a seasonal decision, or a platform requirement—not necessarily an automated optimization result.

Unusual creative experiments

Look for less common customer situations, niche use cases, and unusual claims.

These may reveal a new positioning direction or a creative experiment worth researching further.

Before adopting a similar angle, check whether the message is accurate, relevant to your audience, and supported by your own product.

3. Compare traditional ad creative with AI-search visibility

If your team also monitors AI-assisted search experiences, compare the messages used in visible advertising with the brand information appearing in AI-generated answers.

For example, investigate:

  • Does the same core product benefit appear across ads and public product information?
  • Are competitor comparisons consistent with the advertiser's landing pages?
  • Does the brand communicate a clear use case across different channels?
  • Are there customer questions that competitor messaging does not address?

Search Engine Watch's coverage of Similarweb's AI Ads dataset provides context on the developing measurement of sponsored visibility in AI-powered experiences.

Anstrex should not be described as automatically tracking every ChatGPT or Google AI Mode placement unless that functionality is explicitly supported by the current product.

Instead, use Anstrex for the advertising examples available within its supported coverage and combine that research with separate AI-search monitoring where appropriate.

4. Turn competitor observations into differentiated creative tests

The purpose of competitive intelligence is not to duplicate what another advertiser is doing.

Use the research to identify opportunities to communicate your own value more clearly.

For example:

  • If competitors repeatedly emphasize low prices, investigate whether convenience, service, or reliability is a stronger differentiator.
  • If their ads focus on broad product features, test a message built around a specific customer problem.
  • If the category relies heavily on generic claims, use clearer product demonstrations and verifiable proof.
  • If competitor landing pages are difficult to navigate, test a more direct and useful customer journey.

These are hypotheses. Your own campaign results should determine which ideas deserve further investment.

5. Build a repeatable competitive intelligence loop

An AI-aware Anstrex workflow can follow a simple cycle:

  1. Observe: Collect relevant competitor ads and landing-page examples.
  2. Organize: Tag them by problem, audience language, offer, format, and benefit.
  3. Compare: Identify changes and recurring creative themes.
  4. Hypothesize: Decide which customer needs or positioning opportunities deserve testing.
  5. Create: Develop original creative that reflects your own product and brand.
  6. Measure: Evaluate the results using first-party campaign data.
  7. Refine: Update the research and develop the next set of experiments.

This connects competitor research with actual marketing decisions.

Anstrex helps teams investigate visible advertising activity. Separate analytics and AI-search tools may provide additional information about other discovery environments. First-party campaign data remains essential for determining whether a creative concept produces meaningful business results.

Conclusion — Track the Patterns, Not Just the Ads

AI and automation are changing how advertising campaigns are created, delivered, and optimized.

But that does not mean every ad network has become an LLM or that marketers can see inside a competitor's optimization system.

The more useful shift is in how competitive intelligence is conducted.

A static screenshot provides one observation. A structured collection of creative examples, message changes, landing pages, and repeated patterns can provide a much richer view of visible market activity.

For marketers, the process is straightforward:

  • Monitor relevant competitors and advertising environments.
  • Record creative examples consistently.
  • Identify recurring customer problems and messaging themes.
  • Compare how offers and formats change over time.
  • Use Anstrex to investigate supported advertising sources.
  • Combine those observations with separate AI-search research where relevant.
  • Turn findings into original creative hypotheses.
  • Validate the results through first-party campaign data.

The objective is not to discover a competitor's hidden algorithm. It is to understand what the market is communicating, recognize changes early, and make better advertising decisions.

When creative keeps changing, the advantage belongs to the team that studies the patterns behind the visible ads—not the team that simply collects the most screenshots.

Top converting landing page sample images
Top Converting Landing Pages For Free

Receive top converting landing pages in your inbox every week from us.