TikTok Movie, Keyword Blocking & Content Measurement

Welcome to The Refresh, a weekly newsletter from AdTechGod and Marketecture. Every Thursday we’ll bring you the latest advertising news, commentary, and memes.

• Rokt and mParticle merge (PR Newswire)

• A comprehensive list of 2024 and 2025 tech layoffs (TechCrunch)

• Meta’s Community Notes won’t apply to paid ads. (WSJ)

• Martin Sorrell predicts the END of the Big Six HoldCos: Who will survive the shake-up? (The Drum)

• Verve and Experian Collaborate to power cross-channel advertising solutions (QCnews)

• Chair of UK competition watchdog steps down after Labour intervention leaving questions about the future of Privacy Sandbox (TheGuardian)

Is Broad Keyword Blocking Still Worth It? A Look at the Debate

Brand safety is one of the most discussed topics in digital advertising. Advertisers want to protect their brands from appearing next to unsafe or offensive content, but how far should that protection go? A recent poll by AdTechGod asked a direct question: Is broad keyword blocking worth losing harmless content (for example, blocking the word “shoot” and accidentally blocking basketball content)?

When the poll closed, the results were as follows:

A combined 75% of those surveyed either advocated refining keyword lists or limiting blocking to only the most harmful words. That suggests that many advertisers and industry professionals see an opportunity-or even a necessity: for more nuanced brand-safety tools.

From these discussions and poll findings, a few key themes emerge regarding the future of brand safety strategies:

Will AI save the world? Probably not, but it will help with:

1. Contextual Analysis Over Keywords
   AI-powered contextual analysis can dissect entire articles, videos, or other media to understand real meaning. This approach promises far fewer “false positives” cases in which legitimate, brand-appropriate content gets erroneously filtered out.

2. Refinement of Existing Block Lists
   Many advertisers have built up massive keyword lists over time. The next step is to prune them, removing words that frequently block harmless content. Instead of listing “shoot” as a universal ban, a refined list may block only if “shoot” appears in combination with certain other terms or sentiments.

3. Hybrid Approaches
   For many brands, a blend of minimal keyword blocking plus deeper AI-based scanning can strike a middle ground. The broadest or most universally offensive terms can still serve as a safety net, with everything else evaluated in context.

4. Transparency and Control
   Advertisers increasingly want to see why particular pages are blocked. By using AI solutions that produce transparent reporting (e.g., showing which words triggered a block and how the system interpreted context), advertisers can feel confident in brand-safety decisions and fine-tune settings on the fly.

   Brand safety is moving away from blanket exclusions and toward intelligent, context-based solutions. This approach may protect advertisers more without sacrificing a large swath of perfectly safe inventory which most industry professionals seem ready to embrace.

A New Tool Puts Content Measurement Center Stage in Streaming

Connected TV advertising has grown rapidly, spurring a widespread push for data and programmatic integrations. Despite sophisticated ad measurement, content-level measurement has received relatively little focus, making it harder for publishers to monetize their own shows and distribution platforms.

To address this gap, a new tool by Comscore has been launched that combines audience and contextual insights across social media, mobile, desktop, streaming, and linear TV. The aim is to help publishers understand how viewers consume their content, where they watch it, and how frequently. With these insights, publishers can plan distribution strategies, decide whether to license content to external services, and price ad inventory more effectively.

Platforms can also use this data to demonstrate their audience value for premium content deals. Ultimately, greater visibility into audience and content patterns leads to stronger partnerships, better monetization, and more efficient use of ad resources.

Ari’s View: Also this week, Nielsen gained accreditation for their hybrid TV panel that uses both the in-home meter and “Big Data” from set-top boxes and OEMs. TV measurement is all drama all the time.

Read AdExchanger

Perplexity AI’s Surprise Bid For TikTok US Sparks Debate

Perplexity AI recently submitted a merger proposal to TikTok’s parent company, aiming to form a new entity potentially worth over $50 billion. The move promises ByteDance investors ongoing equity while integrating more video content into Perplexity’s platform.

Comments ranged from skepticism about merging an AI-driven service with a popular short-video platform, to support for any deal that keeps TikTok accessible in the United States. Some wondered if AI was behind the strategic decision, while others noted that 2025 is already poised to be a big year for ad tech mergers.

Although the deal’s details remain unclear, publishers and media professionals appear intrigued. Questions persist: Would Perplexity’s tech empower TikTok’s ad platform and user growth? Or is this an unlikely partnership solely benefiting Perplexity?

For now, the conversation continues as industry insiders watch closely…

Ari’s view: TikTok seems to be investing its actual capital into political capital with the new administration, so unlikely this long-shot offer has much of a chance.

Read CNBC

From Ban to “Thank You, Trump”: TikTok’s Unbelievable Overnight Flip

I don’t use TikTok very often. I tried it, but to be honest, it’s just not my thing. Maybe I’m a bit too old, or maybe I just prefer spending my time on LinkedIn, Instagram, Threads, Reddit, Slack, BlueSky, YouTube and X. That said, the entire situation surrounding the recent TikTok ban and its abrupt reversal felt strangely orchestrated. I don’t have any insider knowledge, but the way everything played out, from the company’s confidence before the ban to the short-lived “blackout” and the sudden reinstatement, had an almost theatrical quality. (see movie poster above)

Up until the ban took effect, TikTok seemed incredibly self-assured that it wouldn’t stick. Employees were reportedly told not to worry about their jobs, and apart from a brief period when the app went dark, nothing felt truly disrupted. That blackout was shorter than some routine server outages on other platforms. Then, just about 12 hours after the ban supposedly kicked in, it was lifted, almost like someone had flipped a switch. Was it a publicity stunt? A political maneuver? It’s anyone’s guess.

Perhaps the most remarkable part was the in-app messaging itself. Typically, when an app faces regulatory hurdles, you’ll see a standard message saying services are temporarily unavailable. In TikTok’s case, it went through three distinct phases:

1. Initial Shutdown Notification
   The company referenced a new U.S. law, saying services would be forced offline on a specific date. That was already unusual, most apps don’t acknowledge legal battles so directly in pop-up notifications.

2. Apology + Political Reference
   The next message explicitly mentioned President Trump’s willingness to work on a solution. It’s rare for major platforms to name-drop a political figure in that way, unless referencing an official announcement. It gave everything a dramatic, almost staged vibe.

3. “Welcome Back!”
   Within hours of the supposed shutdown, users received a pop-up thanking President Trump and announcing TikTok’s return to the U.S. The whiplash from “sorry, we’re shutting down” to “thanks to the president, we’re back!” was unlike anything most of us have ever seen.

Altogether, these rapid changes felt more like a rollercoaster (i hate rollercoasters) than a standard tech or legal battle. Whatever actually happened behind the scenes, it certainly left a lasting impression. Whether you’re a die-hard TikTok fan or someone who barely opens the app, it was hard not to notice the drama. Maybe it was a genuine legal scramble. Maybe it was a well-planned PR move. In the end, though, TikTok is still here, and the whole affair was an unforgettable glimpse into how politics, social media, and public perception can collide in real time.

Join hundreds of marketers, publishers, and agencies in New York City on March 17th.

We are confronting the chaos ahead for advertising and marketing.

Identity is in flux, AI is emerging, and consumer behavior is rapidly changing.

We are bringing advertising experts and AI innovators to help us get prepared.

Connect with Us

Thanks for reading this weeks The Refresh Newsletter. Marketecture Media wants to help bring you the latest news, opinions and community updates. If you are interested in advertising on this newsletter, our podcasts, The Advertising Forum or our slack community you can by contacting us here.