Tag: Analytics

Why MMM Still Demands Clean Data and Human Judgment
I see marketing mix modeling (MMM) becoming easier to access, but I do not think it has become easy to get right.

After several conversations about MMM adoption, I keep hearing the same concern: “We believe in MMM, but we do not know how to get started.”

My answer is that open-source platforms have lowered the barrier to entry in a meaningful way. What they have not lowered is the level of expertise required to produce results that are trustworthy, explainable, and useful for decision-making.

Open-source MMM has changed the starting point

I am seeing MMM adoption accelerate because marketers need more durable measurement methods. Almost half of U.S. marketers expect to invest more in MMM over the next year, and many now rank it as one of the most reliable measurement approaches available.

The open-source shift is real. Three production-grade libraries now give teams a practical way to approach MMM across a wide methodological spectrum.
- Robyn (Meta, R): I see this as the most approachable starting point because it includes automated hyperparameter search through Nevergrad, Pareto frontier model selection, decomposition, and response curve plots. It is also the one I use most often because it is highly customizable.
- Meridian (Google, Python/TensorFlow): I view Meridian as a more rigorous option, especially because it uses Bayesian inference, geo-level priors, and principled uncertainty quantification. The tradeoff is a steeper learning curve.
- PyMC-Marketing (PyMC Labs, Python): I consider this the most flexible path. It offers a full probabilistic model that comes closest to academic-grade Bayesian MMM, but it also demands the most statistical fluency.
This generation of tools has removed the old $150,000 to $500,000 consulting gate that used to be the primary path into MMM. A team with R or Python expertise and reasonably clean historical data can now run a model in-house.

Open-source R and Python tools have lowered the cost of starting with marketing mix modeling, but the expertise needed to produce trustworthy, actionable MMM remains the real ceiling.

The caveat I always make explicit is this: “free tool” does not mean “free model.” The software may be free, but the domain expertise needed to configure it correctly is not. That expertise is a major part of the value.

The vendor landscape is crowded and complicated

I also see a fast-growing SaaS layer built on top of open-source MMM. To evaluate it clearly, I find it helpful to separate vendors into a few practical groups.

Data-layer-first vendors

Platforms like Rockerbox and Northbeam started with attribution and data collection, then added MMM. Their advantage is usually pipeline speed and data access, not deep modeling flexibility or customization.

Measurement-first vendors

Platforms such as Measured, Analytic Partners, Ekimetrics, and Nielsen Gracenote tend to offer more rigorous modeling and enterprise-grade capabilities, usually at a higher price point.

Google Meridian and GA360

I think Google’s decision to open-source Meridian is both a generous contribution to the field and a strategic move. When a walled garden funds and packages a measurement methodology that can be used to evaluate its own channels, I believe it is worth maintaining healthy skepticism about priors, defaults, and assumptions, even when the code is transparent.

Open-source MMM tools now span a clear trade-off: Robyn offers the most approachable starting point, Meridian adds Bayesian rigor, and PyMC-Marketing pushes deepest into statistical flexibility.

The practical vendor question I keep coming back to is simple: who owns the data layer, and does that ownership create conflicts in the modeling layer?

Challenge 1: Data access can quietly break MMM

I think data access is the most underappreciated MMM implementation blocker. A well-specified model needs more than a quick export from a dashboard.
- I usually want two to three years of weekly data as a baseline, so the model can capture at least two full seasonality cycles and enough spend variation to learn from.
- I need consistent channel-level spend granularity, not just a broad “digital” bucket. Search, social, display, video, and other channels need to be separated.
- I need offline channels such as TV, OOH, radio, events, and direct mail, even though they often live in different systems, belong to different teams, and use incompatible time periods.
- I need external covariates, including macro indicators, competitor activity, pricing data, and product launch calendars.
- For B2B, I often need even more history because longer sales cycles and lower conversion volumes make the data requirements more demanding.
In practice, I often find that the real blocker is the six-week data archaeology project that happens before modeling begins. Finance owns revenue. The brand team owns TV. The agency owns digital spend. A spreadsheet from 2021 may be the only record of trade promotions.

The model is only as good as the data archaeology behind it, and that is rarely the part anyone highlights in a vendor demo.

Challenge 2: I still have to roll up my sleeves

AI assistants have lowered the syntax barrier. They can scaffold a Robyn run, generate a Meridian configuration, or help debug a PyMC model. What they cannot reliably do yet is make the judgment calls that determine whether an MMM is credible.

Rows of illuminated data cabinets and paper files stretch into the distance, capturing the pressure on marketers to turn fragmented customer data into a smarter performance engine.
- I still have to decide where to land on a Pareto frontier across hundreds of model solutions, balancing NRMSE against DECOMP.RSSD tradeoffs.
- I still have to know whether Nevergrad’s optimizer has meaningfully converged or simply landed in a local minimum.
- I still have to configure adstock transformation parameters, including Weibull shape and scale or geometric decay, so they reflect realistic channel behavior.
- I still have to diagnose why a model gives a channel an implausible contribution and decide whether the fix is a prior, a data correction, or a variable exclusion.
In other words, if I try to vibe code my way into MMM, I may end up with a model that appears to work but is wrong in ways I will not catch. The scripting is not the hardest part. The real work is validating the output, including using channel-specific incrementality experiments to calibrate the model.

Challenge 3: Human expertise is not optional

Even if the tools mature enough for AI to run a competent default MMM, I still see human expertise as essential. The irreplaceable work is encoding business context that no model can infer from the data alone.
- Adstock and carryover context: I need to know whether a TV buy carries over for four weeks, paid search carries over for three days, or a brand awareness campaign decays over months. That knowledge usually lives with channel experts, not inside the dataset.
- Saturation curve shape: I need to recognize when a channel is probably approaching diminishing returns before the model says so, and I need to question the model when it suggests something implausible.
- Guardrails and anomaly handling: I need to explicitly model or flag COVID troughs, product launches, pricing shifts, and macro disruptions as structural breaks. AI does not automatically know a client had a pricing crisis in Q3 2022.
- Interpretation sanity checks: If a model assigns 40% of contribution to TV for a brand spending $2 million on TV, I need the experience to say, “That feels wrong,” and investigate. That intuition is earned, not computed.
- Organizational translation: A technically correct model has little value if I cannot explain why it recommends moving 15% of search budget to CTV in language a CMO and CFO will act on.
I start with the groundwork before the model

The best place to begin is not the model itself. I start by understanding what data is needed, who owns it, and who can help interpret it in the context of real marketing decisions.

None of that is quick or easy, but it is essential if I want meaningful insight from MMM, whether I choose an open-source library or a subscription-based platform.

As a practical first step, I would download Robyn’s demo script and experiment with sample data before applying MMM to my own business data. That kind of hands-on testing makes the strengths, limits, and judgment calls much clearer.

Inspired by this post on Search Engine Land.
July 9, 2026
Why I Run Each Prompt Once Daily: The Data Behind It

I often get asked why I “only” run each prompt one time per day.

For me, the answer comes down to signal quality. Running a prompt once daily gives me enough consistent data to understand performance without overloading the process with unnecessary repetition.

The statistics show that a single daily run is plenty. It gives me a reliable view of how prompts behave over time, while keeping the workflow focused, efficient, and easier to interpret.

Inspired by this post on Try Profound Blog.

July 7, 2026
Google Search Console Adds Powerful Social Video Reporting

I’m seeing Google Search Console get a useful new reporting layer for social and video content through what Google calls platform properties. This gives me a way to understand how my content on Instagram, TikTok, X, and YouTube is performing in Google Search.

The big change is that I can now connect supported social or video accounts to Search Console and see how people find that content through Google. Instead of only analyzing websites I own or manage directly, I can begin looking at search visibility for content hosted on third-party platforms.

Google said this update makes it possible to track which search terms lead people to Instagram, TikTok, X, and YouTube content in Search, along with how audiences interact with those posts. I’ll be able to review this data inside the performance report, insights report, and achievements sections of Google Search Console.

A Google Search Console dropdown highlights the new platform property flow, with the rustybrick X profile appearing as a selectable property for reporting.

In the performance report, I can review total clicks, impressions, and other key metrics. I can also filter and sort the data to see which posts and queries are driving the most traffic, and if I want to analyze it somewhere else, I can export the data.

In the insights report, I can get a higher-level view of recent traffic trends, top-performing posts, and the ways people are discovering my account through Google Search.

A Google Search Console platform property view shows how an X profile appears in Search, pairing 28-day click and impression trends with the queries driving visibility.

The achievements section adds another useful angle by helping me track growth milestones, such as reaching a new threshold for total clicks from Google Search over the last 28 days.

This feels similar to the social channel details that previously appeared in Search Console insights, but platform properties look like a more direct way to verify and analyze these accounts.

A Google Search Console Insights view highlights how YouTube posts are gaining visibility in Search, with 17.8K clicks and traffic broken down by web, video, Discover, and image search.

To set this up, I need to verify a platform property inside my Google Search Console account. I can start by opening Search Console, going to the Search Console verification page, or using the property selector dropdown anywhere in Search Console and choosing “Add property.”

From there, I select one of the currently supported platforms: Instagram, TikTok, X, or YouTube. Then I follow the onscreen verification steps to securely authorize the connection.

A glowing Google search bar cuts through streams of digital data, capturing the fast-moving world of search, shopping visibility, and SEO innovation.

Google said platform properties will roll out gradually over the coming weeks, so I may not see the option in my account right away. For setup details, Google points users to its help center documentation. The help document had briefly appeared a few weeks earlier before being removed, so this release makes the feature official.

This is also different from Google’s search profiles feature, which has its own analytics.

What stands out to me is the access this gives marketers, creators, and SEOs. Google has not traditionally given us a clear way to see how our content performs on domains or properties we do not own. With platform properties, I can finally start seeing how my social and video content performs in Google Search, even when I do not have developer access to those platforms. That opens up a much better view of search-driven visibility beyond my own website.

Inspired by this post on Search Engine Land.

July 7, 2026
Why ChatGPT Brand Recommendations Drive High-Intent Visits

When I look at Similarweb’s findings, the message is clear: users who saw a brand recommended by ChatGPT were much more likely to visit that brand’s website within a week.

What happened. I found the biggest takeaway in the behavior shift. On average, users were 2.5 times more likely to visit an AI-recommended brand than a direct competitor, based on Similarweb’s study of U.S. desktop activity across finance, travel, and beauty.

Similarweb tracked users who asked ChatGPT industry-relevant questions, received a specific brand recommendation, and then visited either that recommended brand’s website or a competitor’s site within seven days.

To keep the data focused, the study excluded users who had visited the brand’s site in the prior four weeks or had named the brand directly in their prompt.

Recommendations shifted traffic. I saw the same pattern appear across all three industries Similarweb analyzed, which makes this more than a one-category trend.

In finance, after an American Express recommendation, 7.2% of users visited American Express, compared with 3.1% who visited Capital One. After a Capital One recommendation, 14.2% visited Capital One, compared with 3.8% who visited American Express.

In travel, after a Skyscanner recommendation, 9.5% visited Skyscanner, compared with 7.6% who visited Kayak. After a Kayak recommendation, 12% visited Kayak, compared with 3.4% who visited Skyscanner.

In beauty, after a Sephora recommendation, 7.9% visited Sephora, compared with 3.3% who visited Ulta. After an Ulta recommendation, 7.6% visited Ulta, compared with 4.6% who visited Sephora.

AI demand showed up in search. What stands out to me is that most AI-influenced visits did not appear as AI referral traffic. ChatGPT may shape the user’s brand choice, but the later website visit often shows up in analytics as search traffic instead.

Similarweb found that 55.9% of AI-influenced visits came through search, compared with 40.4% of non-AI-influenced visits.

Direct traffic told a different story. It accounted for 19.9% of AI-influenced visits, compared with 38.8% of standard visits.

Recommended users stayed longer. I also think the engagement data matters. AI-influenced visitors viewed 12 pages and spent 11.8 minutes on site, on average, compared with 6.5 pages and 5.6 minutes for non-AI-influenced visitors.

That deeper engagement suggests these users may have already narrowed their options during the AI conversation before they ever reached the brand’s website, Similarweb said.

Why I care. AI visibility can drive meaningful visits even when referral reports miss the original source of influence. I need to understand whether ChatGPT is creating demand for my brand or sending that demand to a competitor.

About the data. Similarweb used its opted-in U.S. desktop web panel to track user journeys from July through December 2025. The report focused on finance, travel, and beauty brand pairs with competitive overlap.

The report: The Downstream Impact of AI Visibility (registration required).

Inspired by this post on Search Engine Land.

July 3, 2026
Channel Strategies: Broad Approaches vs. Focused Commitment

When I first started looking at budget allocation, I was tempted to believe that every marketing channel followed the same path: spend a little, get a lot, but with diminishing returns.

Visually, it’s easy to assume all channels mimic this pattern.

The typical log-shaped curve illustrates that the first dollar you spend is often the most productive. With this mindset, spreading the budget across numerous channels seems like the go-to strategy.

However, I quickly learned not all channels conform to this model. Some require much more than just a sprinkle of funds to be effective. These channels start with a less efficient spend but eventually pay off if given time to warm up. This condition shifts away from the usual ‘test small, scale the winners’ strategy many marketers follow.

At the core of this difference lies a fundamental question: Is the response curve C-shaped or S-shaped?

Understanding the shape of the response curve can drastically change how I conduct channel testing and measurement, especially with Google’s increasing inclination towards S-shaped campaigns.

Let’s delve into what these two curves signify and why they are crucial.

Response curves plot conversions or revenue against spend. Typically, we encounter two main types in marketing.

A C-shaped curve means diminishing returns kick in from the first dollar spent. Meanwhile, an S-shaped curve starts slow, becomes steep at the inflection point, and finally leads to saturation.

This insight is crucial for allocation because the marginal curve—the derivative—guides budget decisions. Here, shapes diverge with significant implications.

For a C-shaped curve, the highest marginal return is from the first dollar, decreasing thereafter. Conversely, for an S-shaped curve, the initial return is low, increases up to a peak, and then declines.

This aspect of increasing marginal returns is pivotal. It’s what differentiates channels with productive small budgets from those that seem inefficient but could perform better when scaled correctly.

Mainstream marketing campaigns exhibit this principle clearly. For instance, if your CPA goal is $50, the way the S-shaped channel behaves under scaling tells a critical story.

A preliminary $10,000 test may misleadingly suggest failure, but at $20,000-$25,000, the channel might be your most cost-effective choice. Small trials in the warm-up phase mislead the eventual conclusion.

This common misconception arises as many automatically rely on ‘test small, scale what works’. Yet, without sufficient testing past the warm-up phase of an S-curve, we risk dismissing channels that could have been game-changers.

For allocation logic, in C-shaped channels, going wide is beneficial. One global optimum dictates that spreading your budget thinly across many channels generally works.

But with S-shaped channels, a small budget is inadequate. Either commit enough budget to surpass the inflection point or don’t invest at all. There is a true minimum budget to ensure viability.

In marketing, determining whether a channel requires breadth or depth is critical. Channels historically leaned towards a concave shape, although modern platform dynamics have blurred these lines.

The differences are increasingly relevant with AI-driven campaigns. For example, ‘AI Max’ necessitates sufficient conversion data to learn effectively, affirming the concave-to-sigmoid shift. Campaigns like PMax blend both response types, initially concealing inefficiencies through promising headline numbers.

The key is recognizing the harvest versus create dichotomy. Harvest channels, like branded searches, display fast saturation and diminishing returns. Still, creating new demand—especially through platforms like Meta or YouTube—demands investment beyond superficial trials for truly incremental growth.

In conclusion, understanding whether to expand broadly or concentrate deeply in a specific channel can transform the efficiency of a marketing strategy.

Inspired by this post on Search Engine Land.

June 23, 2026
Unlocking SEO ROI: Boost Revenue with These 3 Strategies
When I dive into SEO attribution, it often feels like navigating a maze. Unlike paid search, organic search doesn’t offer the same level of tracking precision. Plus, there’s a delay between the work done and the observable results, largely because of factors like fluctuating rankings that are beyond our control.

And just when I think I’ve got a handle on it, new challenges present themselves. With AI-generated answers monopolizing SERPs and LLMs that might not link back to our content, SEO attribution has become even muddier. But at the end of the day, businesses only care about one thing: tangible returns on their marketing investments.

Here’s the silver lining: It’s still within my reach to craft a compelling ROI story through SEO. It requires nuanced thinking, deep data analysis, and more complex mathematics than ever. Let me guide you through the essentials to consider while building your next SEO ROI narrative.

Let’s start with the tried-and-true formula we’ve always used for SEO ROI:
- ROI = ((Incremental organic revenue − SEO costs) / SEO costs) x 100
This formula is simple and executive-friendly, having served its purpose well before AI’s interference in search. But with the rise in zero-click searches and attribution challenges from LLMs, our traditional models are less effective.

Organic traffic trends might seem stagnant or declining, yet visibility could be growing through impressions or AI enhancements. We need a fresh approach to authentically represent SEO’s value. Here are my three strategies for building a more comprehensive ROI model.

1. Acknowledge All Organic Revenue, Not Just Incremental Gains

With 60% of searches ending without a click—and that figure is growing—it’s crucial to see SEO as a defensive strategy as much as anything. Think of our efforts as protecting web traffic that might otherwise fall off the map.

Consider the analogy of judging a goalkeeper by goals scored; it’s more about preservation. Likewise, good SEO means defending existing traffic as much as chasing new clicks. Rather than focusing on new achievement only, remember the entire spectrum of organic revenue SEO helps secure.

Segment Brand vs. Non-Brand Clicks

Giving SEO credit for all organic revenue may seem dishonest if brand-led growth is driving results. Brand traffic can fluctuate due to multiple factors, from PR campaigns to word-of-mouth, and aren’t solely SEO’s doing.

Since we can’t achieve a neat split in Google Analytics, my workaround is to extract branded versus non-branded data from Google Search Console. Here’s an example with real-world data:

In this scenario, to fairly distribute credit, if 70% of traffic is branded and 30% is non-branded, we would attribute a portion (e.g., 10% for branded, 100% for non-branded) based on their respective impact.
- (70% brand x 10% weight) + (30% non-brand x 100% weight) = 37% blended attribution weight
With this model, a site generating $100,000 in monthly organic revenue translates to $37,000 credited to SEO, adequately recognizing its broader scope.

2. Consider Assisted Conversions and First-Click Influence

I’ve always considered last-click attribution as limiting for SEO insights. Organic is often the gateway to a consumer’s journey, and its role is foundational—even if there’s no direct click indicating it.

It’s vital that we recognize when organic assists a conversion, despite another channel closing the deal.

GA4, albeit less straightforward than Universal Analytics, allows us to look at fractional credit using data-driven attribution to prop up the assist role SEO plays.
- 1,345.69 (early) + 687.34 (mid) = 2,033.03 in conversion credit
For illustrative purposes, calculating the value is as simple as multiplying these credits by $100, yielding $203,303 in attributed revenue, well above what SEO alone would capture via last-click metrics.

3. Assess SEO Content’s Cross-Channel Impact

The byproduct of our work on organic-optimized content is often overlooked in metrics. When SEO-led articles and research translate into usable material for ads or campaigns, it’s an extension of our influence across channels.

I noticed a client benefiting from fresh articles and content updates within a mere month, catalyzing conversions on unrelated channels.

Even modest figures, like 29 calls and five qualified leads, spell opportunity for growth and recognition of SEO’s extended value.

Adopting a system to track pages that have been utilized across multiple platforms is one way to give attribution where due:
- 500 conversions (paid search) x $100 (conversion value) x 5% (from SEO pages) = $2,500
This approach, despite more complex math, highlights SEO’s role in a bigger revenue picture. Always account for these values when quantifying SEO contributions.

The Do’s and Don’ts of SEO ROI

SEO’s impact shouldn’t be restricted to merely counting revenue leaps. Tailor your approach, collaborate with analytical thinkers, and make sure to:
- Thank all organic performance, avoiding credit for every branded effort.
- Consider varied attribution models; don’t confine yourself to the organic silo.
- Value when SEO content is reused by others; track its downstream impact.
- Try innovative angles to crack the ROI code without being bound by old metrics.
The primary ROI model isn’t incorrect, merely lacking in scope. As search landscapes evolve, so must our methods of measuring success.

Inspired by this post on Search Engine Land.
June 22, 2026
Master Google Analytics with New Source Grouping & Filters

I’m excited to share that Google Analytics is introducing significant updates aimed at streamlining our data analysis efforts. The introduction of cleaner source attribution and enhanced filtering controls is set to make evaluating cross-channel performance much simpler.
With these updates, I’m finding it easier to manage fragmented traffic source reports, enhance cross-channel performance analysis, and minimize noise in the analytics data we rely on.
What’s New. The new Source Group reporting dimension consolidates different traffic source variations into one cohesive category.
For example, instead of seeing scattered source names like “facebook,” “fb,” and others, all Facebook-related traffic can now be grouped under a single identifiable value.
At the same time, Google’s improvements to the Source Platform field ensure classifications align consistently across advertising channels, providing us with clearer data insights.
Why We Care. This cleaner source classification allows me to perform more accurate attribution analysis and cross-channel reporting. Instead of dealing with traffic fragmented by inconsistent labels, I can better understand which platforms truly drive conversions and where our budgets are yielding the best performance.
Including AI traffic sources like ChatGPT and Perplexity in this analysis offers a standardized way to measure these emerging channels alongside traditional ones. New hostname filters further refine data quality by making sure that only approved domain traffic enters our reporting.
The Big Picture. As we manage campaigns across multiple platforms, inconsistent source naming complicates attribution and budget analysis. This new reporting structure is a breath of fresh air, simplifying these comparisons and enhancing our strategic decision-making.
Between the Lines. This update extends source standardization beyond Google’s properties to platforms like TikTok, Pinterest, and Amazon, while also including support for emerging AI-driven traffic sources such as ChatGPT and Perplexity.
Also New. Google has added hostname filters in the Admin section, allowing us to exclude events from unapproved domains before reporting, enhancing data accuracy.
This feature helps prevent unwanted traffic from skewing our analysis, ensuring that our data remains precise and actionable.
What Advertisers Get. The updates provide standardized source reporting, retroactive access to historical source group data, cleaner attribution analysis, and more control over which domains contribute to reporting.
The Bottom Line. Google is equipping us with new tools to maintain reporting consistency, improve attribution analysis, and keep datasets cleaner as our traffic sources continue to diversify.

Inspired by this post on Search Engine Land.

June 12, 2026
Transform SEO Reports into Actionable Insights

When I work on SEO reports, I know they often include comprehensive research. This might involve keyword data, technical analysis, competitor insights, content gaps, and actionable recommendations. Yet, the challenge arises when stakeholders finish reviewing the report but remain unsure about the immediate next steps.

Take, for example, a report suggesting improvements in internal linking. It typically fails to pinpoint which specific pages need links, the team responsible for these updates, the timeline for execution, or the expected outcomes. Similarly, identifying a crawl issue without outlining its priority compared to fixing existing content gaps can leave teams confused.

This is a critical juncture where many SEO reports lose their impact. The analysis may be accurate, but the path forward often lacks clarity.

What I strive for in a strong SEO report is to guide readers into understanding the present priorities, their importance to business objectives, and the immediate actions needed. This reduces the need for further interpretation before implementation can commence.

Research Is Useful, But It’s Not the Final Output

The SEO activities I engage in, such as keyword research, SERP analysis, technical crawls, competitor reviews, and content audits unearth many hidden opportunities and risks. However, it’s crucial that these inputs don’t overshadow the final report.

What stakeholders truly need from me are the conclusions derived from this research. They need clarity on which findings are impactful, which improvements can be deferred, and which actions should be prioritized.

As an illustration, while identifying 300 pages with missing meta descriptions, the report should clarify the significance of those pages. If the descriptions are of low-value archive pages, they might not require immediate attention. However, missing descriptions on high-intent service pages demand prompt action.

The same principle applies to keyword gaps; a useful report pinpoints high-opportunity keywords aligned with commercial intent and informs stakeholders why certain issues deserve immediate action.

Tailor Reports to the Stakeholder

In my experience, SEO reports often fail to incite action because they treat all stakeholders the same. Each stakeholder, whether a CEO, marketing lead, developer, or content manager, requires different levels of detail, and presenting information in their context is critical.

For executives, I focus on business opportunities, risks, resources, and expected impacts, while marketing leads need to understand how SEO efforts tie into demand generation and campaign strategy.

Developers require a clear technical path, and content teams need page-specific action plans. My goal is to present findings in a way that each stakeholder can easily act upon.

What a Decision-Ready SEO Report Should Show

In crafting a useful SEO report, I aim to address a concise set of questions that, while varying across stakeholders, consistently serve the purpose of guiding the next steps.

By starting with clearly identifying where SEO can create business value, pinpointing constraints, and defining prioritized actions, I ensure the report supports effective decision-making.

Finally, outlining how progress will be measured ensures stakeholders remain aligned and motivated as the project unfolds.

Turn Every Finding into a Clear Next Step

All significant findings in an SEO report should be immediately actionable. By answering what was found, why it matters, and what action should follow, I enable stakeholders to move forward confidently.

For instance, a discovery of high-traffic pages lacking links to commercial pages should lead to specific steps involving content updates and measurement, ensuring progress is tracked and objectives are met.

What to Cut from SEO Reports

To make SEO reports concise and effective, I exclude unnecessary data such as tool screenshots and extensive keyword exports. While supporting materials are valuable, the main report should focus on clarity and priority.

I also shorten methodologies unless essential for building trust or understanding. Keeping the report streamlined ensures stakeholders are not overwhelmed with information that doesn’t aid in decision-making.

The Best SEO Reports Make the Next Step Obvious

Ultimately, my objective with SEO reporting is to minimize uncertainty. After reviewing the report, stakeholders should clearly understand what requires attention and the direction to proceed.

Although SEO lacks absolute prediction, each recommendation should outline expected impacts and the signals used to measure progress, turning findings into active projects that propel the business forward.

Inspired by this post on Search Engine Land.

June 10, 2026
Google’s Zero-Click Searches Surpass 68%: A 2026 Study Insight

In early 2026, a significant shift unfolded in the world of search engines—68.01% of Google searches ended without a click. I discovered this intriguing fact through a study by SparkToro, which utilized Similarweb clickstream data. This percentage marks a noticeable rise from 60.45% in 2024, a 7.56-point increase over two years.
Fewer searches are resulting in clicks. Between 2024 and 2026, the share of searches generating at least one click fell by 9.51 percentage points, representing a decline of 22.9%. This includes clicks to organic results, paid ads, and Google-owned platforms like Maps and YouTube, excluding follow-up searches within Google.
During this period, I noticed that the share of searches leading to another Google search increased by 7.2 percentage points. This trend demonstrates Google’s growing proficiency in providing direct answers within its search results, encouraging us to refine or continue our searches without leaving the platform.
AI Overviews and the zero-click phenomenon. SparkToro suggests that AI Overviews might be contributing to the rise in zero-click searches, though the study doesn’t pinpoint how much of the rise from 2024 to 2026 can be specifically attributed to these overviews.
According to the research, I’ve observed that AI Overviews now appear in over 20% of Google searches, causing click-through rates to plummet by nearly 60% when they do.
AI Mode and zero-click growth. While AI Mode seemed to play a minor role during the study period from January to April 2026, SparkToro noted that only 0.34% of searches transitioned into AI Mode. However, Google announced during I/O 2026 that AI Mode had attracted over 1 billion monthly users, with query volume more than doubling each quarter, indicating a future increase in influence on search behavior.
Historical perspective on zero-click searches. SparkToro’s long-standing tracking of zero-click searches reveals an upward trend, although constantly changing data sources mean that long-term comparisons might lack precision. Nonetheless, available data consistently indicates an increase in zero-click behavior over time.
Here are some historical insights: In 2019, 49% of Google searches ended without a click, based on Jumpshot clickstream data. By 2020, SimilarWeb data showed that the figure had risen to 64.82%. And in 2024, 58.5% of U.S. searches (59.7% in the EU) ended without clicks, according to Datos data.
Why this matters to us. These findings imply that Google is increasingly meeting user needs internally, which might reduce traffic to external websites. However, direct year-to-year comparisons should be approached with caution due to differing methodologies in SparkToro’s analyses.
The evolving role of SEO. SEO remains crucial, but it’s not the sole solution for regaining traditional levels of Google-referred traffic. Rand Fishkin, SparkToro’s co-founder, advised us to focus on building brand awareness and engagement on platforms where our audience is active, irrespective of the impact on direct site visits.
SEO is still valuable for certain categories, such as branded searches, local business inquiries, and high-intent transactional searches, according to Fishkin.
About the study data. The research utilized Similarweb desktop and mobile web panel data on U.S. Google searches from January through April 2026. SparkToro estimated two-thirds of searches occurred on mobile devices, with the remainder on desktops. Searches within Google’s mobile search app, where zero-click behavior might be higher, were excluded.
To explore these insights further, check out the study titled In 2026, Less than One Third of Google Searches Still Send a Click.

Inspired by this post on Search Engine Land.

June 9, 2026
Mastering PPC: Dynamic Strategies for Budget Success
I’ve realized that chasing the perfect PPC budget split can be a never-ending task. Fixed budget ratios often struggle to withstand real-world scenarios, which is why I’ve learned to assess funnel health and adjust spending as market dynamics evolve.

Most PPC budget discussions revolve around balancing brand awareness with conversion-driven campaigns, but I’ve found that this is often not the ultimate goal.

In my experience, the ideal balance is subject to constant change, influenced by our business stage, market saturation, seasonality, competitive pressures, and revenue goals.

Yet, I’ve noticed that many teams treat funnel splits as fixed decisions—set it and forget it. While it might work today, it could be completely inappropriate in six months.

Budget conversations often lead to debates: should we reduce brand awareness spend since it doesn’t convert directly, or are we risking future pipeline issues if we only focus on conversions?

Both viewpoints have merit, which makes these decisions challenging for us.

The Lower Funnel Case is Simple

When I think about the lower funnel, Shopping, Performance Max, and high-intent Search come to mind.

A term like “buy running shoes new york” signifies a ready-to-purchase mindset. Shopping categorically showcases the right product, while PMax exploits the conversion signals across all Google surfaces. The attributions are clear, ROAS is apparent, and this delights the CFO.

But I understand that these campaigns only capitalize on existing demand—they don’t generate new demand. Each conversion is fed by awareness sparked elsewhere:
- A YouTube pre-roll.
- A friend’s endorsement.
- A social media post.
- Years of brand presence.
I feel like I’m just picking fruit from a tree I didn’t plant.

Search is unique as it serves both ends of the funnel. For instance, a query like “best running shoes for marathon training” is more informational.

The individual is investigating rather than purchasing. With AI Max and broad match expansion, Google Ads pushes Search campaigns deeper into this space, enabling Search to straddle both ends of the funnel based on its configuration and captured queries.

It’s something I regularly review: Is our Search spend closing existing demand, or are we engaging with prospects earlier in their journey?

This strategy holds until it falters, often with slow warnings of decline.

Branded search volumes may stagnate, CPCs soar for core terms, and new customer acquisition rates may plateau as retention remains stable—symptoms of a brand living off existing demand without revitalizing it.

Lower-funnel efficiency is real, yet it counters future growth.

Dig deeper: PPC budget planning: Aligning business goals, ad spend, and performance

The Reseller Trap in Lower Funnel

I’ve encountered issues quite specific to resellers and multi-brand ecommerce that don’t get enough attention.

If I sell branded products not owned by my organization, our lower funnel might perform well short-term.

Shopping and Search campaigns do wonders for established brands since brand owners have taken care of awareness. I’m simply reaping the demand built by major brands like Nike or Adidas.

Yet, I lack control over that demand. If a brand cuts back on marketing, exits the market, or loses relevance, our Shopping and Search performance suffers.

The ability to counter such shifts is hampered by the absent demand to harvest.

This predicament requires us to prioritize two strategic imperatives, something often overlooked.
- Own-brand expansion: Allowing us to retain control and invest in independent awareness.
- Enhancing reseller brand: By upping upper-funnel visibility, customers will recognize our name as a destination for all brands we offer.
Both strategies entail upper-funnel spending. Creating our brand necessitates campaigns to elevate product recognition. Building a reseller brand requires enduring efforts in Demand Gen, YouTube, and Display to ensure our brand is integral to the category, beyond individual brands. This applies beyond Google’s ecosystem.

Ultimately, these investments will not manifest in the short-term ROAS report but will signify next year’s resilience in business.

Upper Funnel as Inventory Management

I often see brand awareness spend as the uncertain, tough-to-quantify budget segment, earmarked for leftover funds. This perspective, however, is misplaced.

Investing in the upper funnel is about creating a pool of future converters. Every Demand Gen ad impression on YouTube or Google Display isn’t a wasted effort—it’s a potential high-intent search opportunity in coming weeks, nurturing the top of the funnel for Shopping and Search endeavors to reap later.

Google’s Demand Gen campaigns effectively highlight this throughout a single platform. I use Demand Gen to engage with audiences unfamiliar with our brand, then track Search impression shares and query volumes that surge in subsequent weeks. This lag is both tangible and trackable.

Upper-funnel spending impacts lower-funnel effectiveness the next month, not immediately. This delay prompts cuts when budgets shrink, causing impacts six to eight weeks later rather than instantly.

For effective demand management, I consider upper-funnel campaigns as pipeline investments. The central question isn’t “What is the ROAS on this campaign?” but rather “How much qualified demand is being generated for my Shopping and Search strategies to convert?”

Dig deeper: Paid media efficiency: How to cut waste and improve ROAS

Why Fixed Splits Fall Short

Fixed rules like the 70/30 or 60/40 I often see are merely broad averages seen across different businesses and contexts. They’re decent starting points but poor long-term strategies.

I must account for what affects the optimal split.
- Introducing a new product entails a robust upper-funnel effort given the minimal brand awareness.
- Even mature products in competitive fields require the same, due to shared high-intent search pools with rivals—expanding the pool is the only growth method.
- Seasonal ventures make it essential to complete upper-funnel efforts before peaks, as urgent awareness builds are ineffective in-season.
Conversely, when we face financial constraints or urgent revenue goals, patience for an eight-week upper-funnel maturation isn’t possible. In such cases, focusing on the lower funnel becomes necessary, accepting inevitable drawbacks while planning future awareness investments as pressures ease.

In essence, both choices are appropriate given context. A set split disregards context entirely.

Formulating a Dynamic Budget Split

Rather than adhering to fixed ratios, I advocate establishing criteria that trigger budget adjustments where needed.

Increase upper-funnel focus when:
- Branded search remains static or declines over quarters.
- New customer acquisition costs increase, while retention holds.
- We’re entering new markets or launching new products.
- Competitors significantly amplify brand presence.
- We’re nearing peak season with ample preparation time.
- Reselling top brands with dwindling search interest or decreased active marketing.
Emphasize the lower funnel when:
- Immediate revenue targets cannot wait.
- The upper-funnel campaigns begin showing measurable awareness, indicating readiness for conversion.
- Shopping or Search costs per acquisition fall below target, justifying scaling.
- Demand Gen audience reach saturates, indicating repetitive reach instead of expansion.
Within Google Ads, the necessary data for monitoring this is accessible without additional tools. Trends in branded query and impression share on non-branded terms, along with Demand Gen metrics and customer segmentation data, provide a comprehensive view of funnel health.

Consistent review is as critical as the metrics themselves. I aim for at least monthly funnel split reviews—quarterly rounds are often too infrequent. By the time quarterly evaluations reveal declining branded queries, vital pipeline time has already been lost.

The conversation on funnel balance isn’t typically a matter of analytics—it’s political.

In meetings, lower-funnel spending is easy to defend thanks to visible ROAS and conversion statistics. Conversely, arguing for upper-funnel spending involves creating narratives about future campaign efficacy—a trickier sell under pressure.

Rather than avoiding this justification, I focus on changing the evidence basis.
- Tracking branded search volumes as predictive indicators.
- Ploy a view integrating Demand Gen and Search conversions over time.
- Making lag times distinct, showing evident relationships.
Ultimately, budget allocation isn’t static but a reflection of growth strategies.

Choosing to optimize solely for current ROAS is one decision; investing in future demand drivers another.

For resellers, it also entails whether the business base is self-controlled or rented from brand owners with independent priorities.

I believe the best PPC ventures strike a balance, knowing strategically when to shift focus.

Dig deeper: How to optimize B2B PPC spend when budgets and confidence are low

Inspired by this post on Search Engine Land.
June 8, 2026