The series started out with a fairly simple photographic exploration: “I wanted to capture the energy of Benin through its light, landscape, and people”, Victor shares. But during the photographers time there, wide open landscapes and streetside shots turned into intimate portraits of the people he met along the way. A sincere attempt at documenting somewhere, to simply have a visual record to bring back, instead became much more about ‘being’.
For the first time in a while, Victor allowed himself to explore what came into frame in a different way, capturing that which felt real and uncomposed. He did this by greeting people and pausing often. “Some of the most meaningful portraits came from chance encounters”, he shares, “what I learned is that the most powerful images sometimes come when you’re not trying too hard, when you’re simply open.”
An image that stands out to the photographer is from his encounter with a young woman carrying two of her children – one on her shoulder and the other tied to her back, “all while wearing the biggest, most radiant smile”, he says. “There’s something deeply human and quietly heroic about that moment. It captures the resilience and softness that runs through the whole series. It speaks to care, strength, and presence, the kind of everyday grace that often goes unseen but deserves to be honoured.”
A gentle invitation to feel the warmth of the people of Benin, the project’s title Here, In Light became an apt name for not only the thread of light as a focus of each study, but the feeling of being welcomed and embraced in a place away from home. The collection of images stands as an invitation to slow down to Benin’s pace and find a stillness somewhere in the viewing: “I want the work to evoke warmth, reverence, and recognition, even if you’ve never been,” Victor says.
The project has opened something up for the image maker creatively, and he’s since witnessed a change in the ways he wants to speak visually and how he looks at things: “There are a few ideas already growing quietly in the background, but I’m letting them breathe for now”, he ends. “What I can say is that Here, In Light has reminded me that slowness and presence are non-negotiable in my process.”
Google Search Console now includes a dedicated search appearance filter for discussion forum content, giving publishers new visibility into how user-generated discussions perform in search.
The update applies to pages that use either the DiscussionForumPosting or SocialMediaPostingstructured data types.
In a brief announcement, Google stated:
“Starting today, Search Console will show Discussion Forum rich results as a search appearance in the Performance reports.”
Until now, this type of content was lumped into broader appearance categories like “Rich results” or “Web,” making it difficult to isolate the impact of forum-style markup.
The new filter allows you to track impressions, clicks, and search position metrics specifically tied to discussion content.
This update isn’t about new search capabilities, it’s about measurement. Structured data for forums has been supported for some time, but publishers now have a way to monitor how well that content performs.
The eligible schema types, DiscussionForumPosting and SocialMediaPosting, are designed for pages where people share perspectives, typically in the form of original posts and replies.
Google considers these formats appropriate for traditional forums and community platforms where conversations evolve over time. Pages built around user-generated content with visible discussion threads are the intended use case.
Both schema types share the same structured data requirements, including:
Additional details such as like counts, view stats, or reply structures can also be included. For forums with threaded replies, Google recommends nesting comments under the original post to preserve conversational context.
To qualify for the new search appearance, forum content must follow Google’s structured data guidelines closely.
Google explicitly warns against using this markup for content written by the site owner or their agents. That includes blog posts, product reviews, and Q&A-style content.
If the site’s structure is centered around questions and answers, publishers are expected to use the QAPage schema instead.
Another nuance in the documentation is the recommendation to use Microdata or RDFa rather than JSON-LD. While JSON-LD is still supported, Microdata formats help reduce duplication when large blocks of text are involved.
This update provides a clearer understanding of how forums contribute to search visibility. With the new search appearance filter in place, it’s now possible to:
Google’s decision to break out discussion forum results in Search Console highlights the growing role of user conversations in search. It’s a signal that this type of content deserves focused attention and ongoing optimization.
For publishers running forums or discussion platforms, now’s the time to ensure structured data is implemented correctly and monitor how your community content performs.
Take 5 Oil Change has named Doug Zarkin as its first chief marketing officer (CMO).
The exec joins from wellness and fitness services company Modern Performance and Recovery Brands, where he was chief brand officer since 2023. Before that, he was CMO of Pearle Vision and held marketing leadership roles at fashion companies like Kellwood, Warnaco, and Limited Brands, as well as cosmetics brand Avon.
“I’m here to help lead a talented team into that next chapter—unlocking new relevance and growth, deepening emotional connection, and elevating the role this brand plays in people’s everyday lives,” Zarkin wrote in a LinkedIn post. “We’re not just marketing oil changes. We’re creating a moment of ease in an otherwise chaotic day.”
“Excited to ‘get under the hood,’” he concluded.
As CMO, Zarkin will lead all aspects of the brand’s marketing strategy and execution, with a focus on strengthening its story, deepening customer insights, and advancing its data-driven marketing approach. He will report to executive vice president and group president Mo Khalid.
“We’re thrilled to welcome Doug Zarkin to the Take 5 Oil Change leadership team. Doug’s proven ability to build emotionally resonant brands will be instrumental as we continue to grow and deepen our connection with customers nationwide,” Khalid shared in a statement.
Take 5 Oil Change is a part of Driven Brands’ portfolio of automotive brands, which also includes Maaco, Meineke, and Carstar.
In February, the stay-in-your-car oil change chain launched “Take 5 Clive,” a campaign introducing its enthusiastic new brand character. Created by creative agency Dagger, Clive debuted in a 30-second spot in which he is taken aback by Take 5 Oil Change’s quick services.
“With Take 5 Clive, we’re bringing our differentiated customer experience to life in a way that’s fun, relatable, and showcases the great value of our offering,” Mike DeTrana, vice president of marketing, said in a statement. “Clive is a representation of how we want customers to feel when they discover how simple and convenient their oil change experience can be. Through humor and real-life moments, this campaign highlights what makes our brand and service stand out—speed, friendliness, and simplicity.”
The spot ran nationwide across Take 5 Oil Change’s digital, social, TV, audio, and in-store channels.
The brand also weaved Clive into its recent partnership with F1 The Movie for a pit stop-themed spot. It joined forces with the film for a campaign with exclusive sweepstakes, in-store promotions and discounts, and digital content for both racing enthusiasts and moviegoers.
The CUKTECH power banks may not have the prettiest brand name in tech. And power banks in general are, shall we say, not too aesthetically pleasing at the best of times. But this accessory maker has managed the near-impossible: making a portable power brick that genuinely made me go, “Oh, that looks nice” when I first saw it.
Thankfully for us weirdos who also want things to work properly, we’ve tested both the CUKTECH 15 Ultra and the bigger, more voluminous and even more extravagantly bescreened CUKTECH 20, and came away very impressed by both of them.
A large-scale analysis by Ahrefs of 600,000 webpages finds that Google neither rewards nor penalizes AI-generated content.
The report, authored by Si Quan Ong and Xibeijia Guan, provides a data-driven examination of AI’s role in search visibility. It challenges ongoing speculation that using generative tools could hurt rankings.
Ahrefs pulled the top 20 ranking URLs for 100,000 random keywords from its Keywords Explorer database.
The content of each page was analyzed using Ahrefs’ own AI content detector, built into its Page Inspect feature in Site Explorer.
The result was a dataset of 600,000 URLs, making this a comprehensive study on AI-generated content and search performance.
The data shows AI is already a fixture in high-ranking pages:
Among those mixed pages, usage patterns broke down as:
These findings align with a separate Ahrefs survey from its “State of AI in Content Marketing” report, in which 87% of marketers reported using AI to assist in creating content.
Perhaps the most significant data point is the correlation between AI usage and Google ranking position, which was just 0.011. In practical terms, this indicates no relationship.
The report states:
“There is no clear relationship between how much AI-generated content a page has and how highly it ranks on Google. This suggests that Google neither significantly rewards nor penalizes pages just because they use AI.”
This echoes Google’s own public stance from February 2023, in which the company clarified that it evaluates content based on quality, not whether AI was used to produce it.
While the overall correlation is negligible, Ahrefs notes a slight trend among #1 ranked pages: they tend to have less AI content than those ranking lower.
Pages with minimal AI usage (0–30%) showed a faint preference for top spots. However, the report emphasizes that this isn’t strong enough to suggest a ranking factor, but rather a pattern worth noting.
Fully AI-generated content did appear in top-20 results but rarely ranked #1, reinforcing the challenge of creating top-performing pages using AI alone.
For content marketers, the Ahrefs study provides data-driven reassurance: using AI does not inherently risk a Google penalty.
At the same time, the rarity of pure AI content at the top suggests human oversight still matters.
The report suggests that most successful content today is created using a blend of human input and AI support.
In the words of the authors:
“Google probably doesn’t care how you made the content. It simply cares whether searchers find it helpful.”
The authors compare the state of content creation to the post-nuclear era of steel manufacturing. Just as there’s no longer any manufactured steel untouched by radiation, there may soon be no content untouched by AI.
Ahrefs’ findings indicate that content creators can confidently treat AI as a tool, not a threat. While Google remains focused on helpful, high-quality pages, how that content is made matters less than whether it meets user needs.
Google Analytics 4 (GA4) made a lot of headlines when it replaced Universal Analytics two years ago. Marketers, upset about losing a familiar tool, moaned loudly about GA4’s “problems.” Since then, however, the criticism has died down. Users now understand Google’s move to an events-based system and, more importantly, how to get the most out of it.
That isn’t to say the grumbles have stopped altogether. That’s a good thing, said Steve Ganem, director of product management at Google Analytics. “Customer feedback is crucial to how we develop and evolve GA.”
Ganem walked us through GA4’s most recent developments and some changes users can expect to see soon.
The most significant change is expanding GA4 from only measuring web and app performance into a tool that can measure the entire business.
“We’re evolving to help marketers measure the entire customer journey and enable measurement across your entire business performance,” he said. “This means providing insights across all full-funnel touchpoints, helping you build your first-party data strategy, and integrating data from any source through Data Manager for Google Analytics.”
That means moving beyond last-click attribution with new ways to connect GA4 to non-Google platforms like TikTok, Snap and Pinterest. Ganem said this came about in direct response to customer feedback about understanding the full impact of their media investments.
“This means you’ll be able to see conversions that happen after someone sees your ad — even if they don’t click it — on Google and other platforms,” he said. “This helps you better understand how effective your ads are and if you’re getting a good return on your investment.”
Dig deeper: 3 ways GA4 is way better than UA
To bridge the gap between last click and AI-powered attribution, GA4 updated path reporting and assisted conversion reports. They will now categorize touchpoints into assists and last touch to better see how different media contribute to conversions earlier in the journey.
It won’t be a surprise that Ganem and his team are embedding AI into the platform.
“We’re developing features like projections, which use smarter forecasting so users can see projected KPI performance by channel, enabling real-time optimization decisions and quick adjustments for in-flight budgets,” he said. “We’re also introducing scenario planning to help you create media plans that drive the highest return across your advertising channels based on your defined spend.”
He said adding an AI chat agent is a direct response to feedback for quicker, more straightforward navigation and data understanding.
Because AI is only as good as the data it operates on, GA4 now has features to aid with data quality, with more on the way. These include:
“We aim to offer a complete customer view by bringing together web, app and offline interactions, along with first-party and other business data,” Ganem said. “The audience builder in Google Analytics already easily connects with Google’s tools, and we’re working to make it even simpler for marketers to activate their audiences across various channels.”
Looking ahead, expect more flexible and customizable dashboards, designed to increase the control and access within a Google Analytics property. “This shifts beyond static reports, allowing you to truly visualize your data,” he said.
Email:
MktoForms2.loadForm(“https://app-sj02.marketo.com”, “727-ZQE-044”, 16293, function(form) {
// form.onSubmit(function(){
// });
// form.onSuccess(function (values, followUpUrl) {
// });
});
The post What’s new and what’s next in GA4 appeared first on MarTech.
Let’s get straight to it. You can definitely rebuild foundational knowledge later in a career. No doubt. Perhaps you have even more chance of doing so due to your head start in understanding the subject before you begin, so don’t lose hope.
Saying that, I don’t believe mastery or expertise in anything comes quickly, or easily. To build the confidence, or poise that you’re speaking of takes years (if not decades) to truly develop – it won’t be a case of finding the right online tutorial, or evening class.
My guess is that the designers you’re looking up to and comparing yourself to probably felt exactly the same as you about their education – but they were dogged and unrelenting in the honing of their craft. I also bet those same designers you revere still regularly lose confidence and doubt themselves, no matter what they put on social media. So know you’re just on the same path as them, and be under no illusion it came easy for anyone…
Practically speaking, you can get the education in so many different ways now; through reading books, listening to podcasts, watching online tutorials, seeking out new courses or influences – but that bit for me is irrelevant. There will be no golden shortcut. The actual task at hand here is being resilient and dedicated enough to go to your desk every day, practice and perhaps most importantly, get out there and make sure you’re getting the right experience.
I think I learnt 99% of my design skills and knowledge after I left university. From being curious, excited, open to failure and seeking out opportunities to ask advice from those I admired. Email your favourite designer, ask for the chance to sit next to them on a project, at a conference, to show them you work, or other ways to pick up little tips here and there.
So my answer is annoying really (sorry): you just have to keep going. Keep practicing. Keep sharing your work. Keep asking questions. Keep reaching out to those you admire. Keep seeking honest feedback from people who will tell you the truth. Keep celebrating the wins. Keep reflecting on the challenges. Keep on, keeping on.
Which in itself, especially at moments of economic uncertainty like the industry is facing now, is no small ask – but you were curious enough to write to us, so just because of that, I know you’re clearly more driven than 99% of the other people in your position. You got this.
There’s no one-size-fits-all answer when it comes to deciding how much of your marketing budget should go toward SEO versus PPC.
But that doesn’t mean the decision should be based on gut instinct or what your competitors are doing.
Marketing leaders are under more pressure than ever to show a return on every dollar spent.
So, it’s not about choosing one over the other. It’s about finding the right balance based on your goals, your timelines, and what kind of results the business expects to see.
This article walks through how to think about budget allocation between SEO and PPC with a focus on what kind of output you can reasonably expect for your spend.
When you spend money on PPC, you’re buying immediate visibility.
Whether it’s Google Ads, Microsoft Ads, or paid social, you’re paying for clicks, impressions, and leads right now.
That cost is largely predictable and better to forecast. For example, if your cost-per-click (CPC) is $3 and your budget is $10,000, you can expect about 3,300 clicks.
PPC spend can be directly tied to pipeline, which is why it’s often favored by performance-driven teams.
With SEO, you’re investing in long-term growth. You’re paying for content, technical fixes, site structure improvements, and link acquisition.
But you don’t pay for clicks or impressions. Once rankings improve, those clicks come organically.
The upside is compounding growth and reduced cost per lead over time.
The downside? It can take months to see meaningful impact, and the cost-to-output ratio is harder to predict.
It’s also worth noting that PPC costs often increase with competition, while SEO costs tend to remain relatively stable over time. That can make SEO more scalable in the long term, especially for brands in high-CPC industries.
If you need leads or traffic now, PPC should probably get the bulk of your short-term budget.
Launching a new product? Trying to meet quarterly goals? Paid search and social can give you the volume you need pretty quickly.
But if you’re trying to reduce customer acquisition cost (CAC) in the long run or improve visibility in organic search to support brand awareness, SEO deserves more attention. It builds value over time and often pays dividends past the life of your campaign.
Many brands start with a 70/30 or 60/40 split favoring PPC, then shift the mix as organic efforts gain traction.
Just make sure you set clear expectations: SEO is not a quick fix, and over-promising short-term gains can backfire when the board wants results next quarter.
If you’re rebranding, expanding into new markets, or supporting a product launch, a heavier upfront PPC investment makes sense. But brands that already rank well organically or have strong content foundations can afford to rebalance the mix in favor of SEO.
One emerging challenge for organic marketing is the rise of AI Overviews in Google Search. More brands are seeing a dip in organic traffic even when they maintain strong rankings.
Why?
Because the search experience is shifting. AI-generated summaries are now answering questions directly on the results page, often pushing traditional organic listings further down.
That means your SEO strategy can’t just be about rankings anymore. You need to invest in content that earns visibility in AI Overviews, featured snippets, and other enhanced search features.
This may involve rethinking how content is structured, focusing more on schema markup, FAQs, and direct-answer formats that AI models tend to surface.
In practical terms, your SEO budget should now include:
This shift doesn’t mean SEO is no longer worth it. It means you need to be more strategic in how you spend.
Ask your SEO partner or in-house team how they’re adapting to AI search changes, and make sure your budget reflects that evolution.
Let’s put this into numbers. Say you have a $100,000 annual digital marketing budget.
Putting $80,000 toward PPC might get you 25,000 paid clicks and 500 conversions (based on a fictional $3.20 CPC and 2% conversion rate).
The remaining $20,000 on SEO might buy you four high-quality articles a month, technical clean-up work, and backlink outreach.
If done well, this might start showing traction in three to six months and bring in sustained traffic over time.
The key is to model your budget around what’s actually possible for each channel, not just what you hope will happen. SEO efforts often have a longer lag time, but PPC campaigns can run out of gas as soon as you turn off the spend.
You should also budget for maintenance and reinvestment. Even strong SEO performance requires fresh content and updates to keep rankings.
Similarly, PPC campaigns need regular optimization, creative testing, and bid adjustments to stay efficient.
You should also plan for budget allocation across different campaign types: brand vs. non-brand, search vs. display, and prospecting vs. retargeting.
Each serves a different purpose, and over-investing on one without supporting the others can limit growth.
For example, allocating part of your PPC budget to retargeting warm audiences can drastically improve efficiency compared to cold prospecting alone.
While branded search often delivers low-cost conversions, it shouldn’t be your only area of investment if you’re trying to scale.
Leadership wants to know two things: how much are we spending, and what are we getting in return?
A mixed SEO and PPC strategy gives you the ability to answer both.
PPC provides short-term wins you can report on monthly.
SEO builds long-term momentum that pays off in quarters and years.
Explain that PPC is more like a faucet you control. SEO is more like building your own well. Both are valuable.
But if you only have one or the other, you’re either stuck renting traffic or waiting too long to see the impact.
Board members and non-marketing executives often prefer hard numbers. So, when proposing a budget mix, include projected costs per acquisition, estimated traffic volumes, and timelines for ramp-up.
Make it clear where each dollar is going and what kind of return is expected.
If possible, create a model that shows various scenarios. For example, what a 50/50 vs. 70/30 SEO/PPC split might look like in terms of conversions, traffic, and cost per lead over time.
Visuals help ground the conversation in data rather than preference.
One challenge with mixed-channel budget planning is deciding which key performance indicator (KPI) to prioritize.
PPC is easier to measure in terms of direct return on investment (ROI), but SEO plays a broader role in business success.
For PPC metrics, you may want to focus on KPIs like:
For SEO metrics, you may want to focus on:
When reporting to leadership, show how the two channels complement each other.
For example, paid search might drive immediate clicks, but your top-converting landing page could rank organically and reduce spend over time.
Your initial budget allocation isn’t set in stone. It should evolve based on performance data, market shifts, and internal needs.
If PPC costs rise but conversion rates drop, that could be a cue to pull back and invest more in organic.
If you’re seeing strong rankings but low engagement, it may be time to shift some SEO funds into conversion rate optimization (CRO) or paid retargeting.
Seasonality and campaign cycles also matter. Retailers may lean heavily on PPC during Q4, while B2B companies might invest more in SEO during longer sales cycles.
Set quarterly review points where you re-evaluate performance and make adjustments. That level of agility shows leadership you’re making informed decisions, not just sticking to arbitrary ratios.
Some companies go all-in on SEO, expecting miracles. Others burn through paid budgets with nothing left to sustain organic efforts. Both approaches are risky.
A healthy mix means budgeting for:
Don’t forget to budget for what happens after the click: landing page development, CRO, and reporting tools that tie it all together.
Another mistake is treating SEO as a one-time project instead of an ongoing investment. If you only fund it during a site migration or a content sprint, you’ll lose momentum.
Same goes for PPC: Without a proper landing page experience or conversion tracking, even high-performing ads won’t deliver meaningful results.
There is no universal perfect split between SEO and PPC. But there is a perfect mix for your goals, stage of growth, and available resources.
Take the time to assess what you actually need from each channel and what you can realistically afford. Make sure your projections align with internal timelines and expectations.
And most importantly, keep reviewing your mix as performance data rolls in. The right budget allocation today might look very different six months from now.
Smart marketing leaders don’t choose sides. They choose what makes sense for the business today, and build flexibility into their strategy for tomorrow.
More Resources:
Featured Image: Jirapong Manustrong/Shutterstock
For most of the last 50 years since the upfront was an integral part of marketers’ plans to advertise on TV, the buying bonanza usually wrapped up business between TV sellers and buyers before the July Fourth break. That is definitely not the case this year, for a few reasons.
According to three heads of investment at major agency groups, the multi-billion dollar marketplace is being slowed down for three reasons: its increasing complexity, discrepancies with Nielsen’s latest ratings system, and lingering confusion over which agency is responsible for upfront negotiations with clients that have recently moved their business from one holdco to another.
The first reason has been the case for at least a few years now, since the pandemic irrevocably changed TV viewing habits, bringing on the general decline in linear viewership and accelerating the rise of streaming and connected TV viewing options.
“There’s so many different dynamics now, it’s difficult to say that there is kind of one trend that’s driving everything,” said one multi-agency head of investment who spoke on condition of anonymity in order to speak freely.
The buyer cited myriad factors that need to get sorted with streaming and platform deals, from pricing to cancellation options to dollar volume, which all the major streamers are looking to increase. “There’s certain elements of this that are dragging out out because you have very limited supply — and you have an abundance of supply where demand hasn’t caught up in others,” said the buyer.
The situation with Nielsen revolves around Nielsen’s introduction of its Big Data + Panel (BD+P) ratings, which increased the sample size of its audience from a panel of 42,000 homes to 45 million smart TV homes — a significant shift in sample size, to understate the matter. At the beginning of 2025, Nielsen received accreditation from the Media Rating Council — a big step in validation after losing accreditation in 2021.
Certainly some variance in the numbers was expected, and it seemed possible this new system would add wrinkles to the upfront. But the head of investment at a second agency group said the problem is the inconsistency in the numbers — the fluctuation keeps going up and down without making sense.
“It’s an absolute freaking mess, and it is holding up conversations with a lot of different holding companies,” said the buyer. “With Nielsen being Nielsen, the numbers are wonky. By their own admission, Nielsen said that the adults 25-54 numbers are wrong, and most of my deals are done on adults 25-54.”
The buyer pointed out that over the last year, ratings for a major Hispanic network went from 35% higher in BD+P compared to panel-only to 2% lower in the most recent numbers — wildly divergent. While the numbers for a smaller cable network group dropped 23% from panel to BD+P.
With the latter example, “If we did this the correct way, their CPM would go up by 23% then they’re now going to be priced higher than [the parent channel]. So we’re not going to buy any of that network. We’re going to pull the dollars from there, and they’re likely going to go out of business,” said the buyer.
For its part, Nielsen knows its new ratings aren’t perfect, but argued that the problems are sporadic at best. “The vast majority of clients on both the buy and sell sides support and encourage Big Data + Panel as currency,” wrote a Nielsen representative in an email reply. “We’re thrilled that Big Data + Panel is being embraced. We’ve worked hand in hand with clients for years to get to this point and we’re proud to be the first company to earn accreditation for this new frontier in measurement. We remain committed to our clients so that we can help them achieve their goals.”
Meanwhile, the second buyer said they’re hearing from Nielsen’s competitors including Videoamp and Comscore, but that buyer added that the alternatives all have holes in their offerings too, so that buyer will continue to work through BD+P and said it’s just going to take longer to get to the bottom of where the ratings settle.
Finally, the moves by a handful of major clients to shift North American media budgets from one holding company to another — thinking specifically about Coca-Cola’s and Mars’ moves from WPP Media to Publicis as the two biggest of the last six months — have caused a bit of confusion over which agency is actually handling upfront negotiations. But there are others, from Marco’s pizza to P&G’s Oral Care to PayPal — all of which add up to some $1 billion in spend that’s currently in flux from one agency group to another.
“If I’m actually going to have to go back and negotiate the inclusion of those clients into deals that I’ve already struck, that’s going to be a difficult part of the process,” said a buyer.
Publicis Media declined to comment on which group is handling which part of the upfront negotiations, and WPP Media didn’t respond to a request for comment by press time.
So when will the market wrap? One buyer said they plan to get to finalizing business starting today July 7, and could be finished by end of the week.
The first buyer added that it’s going to take whatever time is needed to get the best deal for clients: “If a negotiation is dragging on, it’s it’s because it needs to, to make sure that we’re delivering adequate value across the board.”
No one can deny gaming has had a profound impact on the time spent on different entertainment sources, especially for younger people. Less clear is the impact it’s had on people’s brand choices — since advertising through gaming has had its ups and downs over the last decade. Dentsu recently issued a report on the state of gaming, and here are some stats on what it found:
“There’s a real reckoning coming as consumer habits shift in an accelerated fashion and publishers bury their heads in the sand.”
— Matt Barash, chief commercial officer at Nova, responding to an Atlantic article “The End of Publishing As We Know It,” which was posted on LinkedIn by ex-Netflix exec Peter Naylor.
As more companies race to adopt generative AI tools, some are learning a hard lesson: when used without oversight or expertise, these tools can cause more problems than they solve.
From broken websites to ineffective marketing copy, the hidden costs of AI mistakes are adding up, forcing businesses to bring in professionals to clean up the mess.
Sarah Skidd, a product marketing manager and freelance writer, was hired to revise the website copy generated by an AI tool for a hospitality company, according to a report by the BBC.
Instead of the time- and cost-savings the client expected, the result was 20 hours of billable rewrites.
Skidd told the BBC:
“[The copy] was supposed to sell and intrigue but instead it was very vanilla.”
This isn’t an isolated case. Skidd said other writers have shared similar stories. One told her that 90% of their workload now consists of editing AI-generated text that falls flat.
The issue isn’t just quality. According to a study by researchers Anders Humlum and Emilie Vestergaard, real-world productivity gains from AI chatbots are far below expectations.
Although controlled experiments show improvements of over 15%, most users report time savings of just 2.8% of their work hours on average.
The risks go beyond boring copy. Sophie, co-owner of Create Designs, a UK-based digital agency, says she’s seen a wave of clients suffer avoidable problems after trying to use AI tools like ChatGPT for quick fixes.
Warner tells the BBC:
“Now they are going to ChatGPT first.”
And that’s often when things go wrong.
In one case, a client used AI-generated code to update an event page. The shortcut crashed their entire website, causing three days of downtime and a $485 repair bill.
Warner says even larger clients encounter similar issues but hesitate to admit AI was involved, making diagnosis harder and more expensive.
Warner added:
“The process of correcting these mistakes takes much longer than if professionals had been consulted from the beginning.”
The Danish research paper by Humlum and Vestergaard finds businesses that offer AI training and establish internal guidelines see better (if still modest) results.
Workers with employer support saved slightly more time, about 3.6% of work hours compared to 2.2% without guidance.
Even then, the productivity benefits don’t seem to trickle down. The study found no measurable changes in earnings, hours worked, or job satisfaction for 97% of AI users surveyed.
Prof. Feng Li, associate dean for research and innovation at Bayes Business School, told the BBC:
“Human oversight is essential. Poor implementation can lead to reputational damage, unexpected costs—and even significant liabilities.”
Register for the webinar: The New SEO Playbook: How AI Is Reshaping Search & Content
Kashish Barot, a copywriter based in Gujarat, India, told the BBC she spends her time editing AI-generated content for U.S. clients.
She says many underestimate what it takes to produce effective writing.
Barot says:
“AI really makes everyone think it’s a few minutes’ work. However, good copyediting, like writing, takes time because you need to think and not just curate like AI.”
The research backs this up: marketers and software developers report slightly higher time savings when employers support AI use, but gains for teachers and accountants are negligible.
While AI tools may speed up certain tasks, they still require human judgment to meet brand standards and audience needs.
Related: This Is Why AI Won’t Take Your Job (Yet)
The takeaway for businesses? AI isn’t a shortcut to quality. Without proper training, strategy, and infrastructure, even the most powerful tools fall short.
What many companies overlook is that AI’s success depends less on the technology itself and more on the people using it, and whether they’ve been equipped to use it well.
Rushed adoption may save time upfront, but it leads to more expensive problems down the line. Whether it’s broken code, off-brand messaging, or public-facing content that lacks nuance, the cost of fixing AI mistakes can quickly outweigh the perceived savings.
For marketers, developers, and business leaders, the lesson is: AI can help, but only when human expertise stays in the loop.
Featured Image: Roman Samborskyi/Shutterstock
