Tag: Measurement

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
Prompt-Level AI Visibility: How I Measure What Matters

I do not measure AI search the same way I measure traditional search, because the user journey is no longer built around one query, one ranking page, and one click.

A prospect might ask ChatGPT for the best CRM for manufacturing companies, compare options in Google AI Mode, refine the requirements across several follow-up questions, and build a shortlist without ever visiting a website.

If my company appears in those conversations, I have influenced the buying process. The hard part is proving that influence with a measurement system I can trust.

Prompt-level visibility has become one of the fastest-growing areas of AI search optimization. It is also one of the easiest to misunderstand. I see plenty of promises about complete visibility into AI conversations, but the reality is far more complicated.

Here is how I think about what can be measured today, what cannot be measured reliably, and how I would build useful reporting despite the current limits.

A 5-step framework I use to track AI visibility

1. I accept that AI does not have traditional rankings

The first mistake I avoid is trying to recreate an old SEO ranking report. There is no universal position one inside ChatGPT.

The same prompt can produce different responses depending on conversation history, user location, personalization, follow-up questions, model version, web retrieval availability, and timing.

That means visibility is probabilistic rather than deterministic. Instead of asking, "Do we rank?" I ask, "How often are we included across the conversations that matter?"

That shift changes the entire measurement model.

2. I build a prompt library instead of only a keyword list

Keywords still matter, but I no longer treat them as enough on their own.

Instead of tracking only individual search terms, I build a library of prompts that reflects how real buyers research, compare, validate, and challenge their options.

I usually organize those prompts by intent. Discovery prompts ask for the best platforms in a category. Comparison prompts put vendors side by side. Evaluation prompts focus on specific use cases. Validation prompts ask whether a company is worth the cost. Objection prompts explore disadvantages. Alternative prompts ask what to use instead. Implementation prompts test how difficult a product may be to adopt.

Instead of monitoring 10 keywords, I may monitor 200 to 500 prompts across the full buying journey. That gives me a much more realistic view of AI visibility.

3. I measure prompt clusters, not isolated questions

One prompt rarely tells me enough to make a decision.

For example, "best CRM software" might not mention my company, while "best CRM for manufacturing companies" might. A more specific prompt, such as "CRM for manufacturers with field sales teams," could return a different set of recommendations altogether.

That is why I group similar prompts into clusters. A category cluster might include best project management software, best PM platform, and project management tools. An industry cluster might include best CRM for healthcare, manufacturing, and finance. A feature cluster might include CRM with AI automation, forecasting, or enterprise sales support.

The patterns across those clusters are more reliable than the result from any single prompt.

4. I combine synthetic prompts with real customer questions

This is where measurement becomes more difficult.

Most organizations do not know exactly what customers are typing into AI assistants, so I often start by generating synthetic prompts. That may include expanding keyword research into conversational questions, creating AI-generated prompt variations, and building comparison, objection, and follow-up prompts.

Synthetic prompts are useful because they are repeatable, but I do not treat them as perfect. Generated prompts often sound cleaner and more structured than real user behavior.

A real buyer might ask something much richer, such as: "We are a 250-person SaaS company with a small HR team. We already use Workday but need something better for payroll. Budget is not a huge issue. What would you recommend?"

That is much more useful than a short phrase like "best payroll software."

For the strongest measurement program, I use synthetic prompts for consistent benchmarking and then supplement them with real questions from sales calls, customer interviews, support conversations, community discussions, internal search logs, on-site search, and AI transcripts that customers voluntarily share.

I also assume the prompt library will need to change. Customer language evolves, and the measurement set has to evolve with it.

5. I measure multi-turn conversations

Most AI-assisted buying journeys do not happen in a single prompt. A buyer may start by asking for the best cybersecurity vendors, narrow the list to companies strong in healthcare, ask which ones integrate with CrowdStrike, and then compare pricing.

My company may not appear in the first answer, but it may become highly recommended by the third response.

If I only measure the opening prompt, I miss a large share of meaningful visibility.

That is why I want prompt tracking to evaluate full conversation paths, not just one-shot questions. Multi-turn testing often reveals patterns that single prompts hide.

The AI visibility metrics I care about most

Many traditional SEO metrics do not translate neatly to AI search. Rankings, clicks, and impressions still have value, but they no longer tell the whole story.

I focus on measurements that show whether a brand appears, how it is positioned, and how consistently it is recommended inside AI-generated responses.

Inclusion rate

If I could track only one AI visibility metric, I would start here.

Inclusion rate measures the percentage of tracked prompts where my brand appears in the AI response. If I monitor 500 prompts and my company appears in 185 of them, the inclusion rate is 37%.

That number is useful as a benchmark, but it becomes more valuable when I segment it by buying stage, product category, industry, geography, or AI model. Those slices often reveal opportunities that a single overall average would hide.

Position within the response

Being mentioned is not the same as being recommended.

Old search marketing tools give way to a faster, connected future, with data streams, AI icons, and a glowing search hub symbolizing SEO innovation and community growth.

I want to know whether my brand appears as the first recommendation, one of the first few options, a late mention, or merely an alternative. If the AI response includes a comparison table, I also want to know where my company appears there.

AI answers do not have traditional rankings, but prominence still matters. A top recommendation is more likely to shape a buyer’s perception than a passing mention several paragraphs later.

Brand framing

Visibility tells me whether my brand is included. Brand framing tells me how it is described.

There is a meaningful difference between an AI system describing a company as "widely considered an enterprise leader" and describing it as "best suited for smaller teams." Both may sound positive, but they position the brand very differently.

I look for recurring themes around strengths, weaknesses, differentiators, pricing, ideal customer profile, and competitive comparisons. Over time, those patterns can expose messaging gaps in my own content or show how the broader web is shaping AI’s understanding of the brand.

Sentiment and confidence

Sentiment is more than a simple positive-or-negative label. I also want to know how confidently the AI system presents my brand.

"Company A is generally considered the strongest option" carries a very different level of conviction than "Company A may be worth considering."

Neither statement is negative, but they do not create the same buyer impression. Tracking confidence, uncertainty, caution, skepticism, and strong endorsement gives me a more nuanced view of how AI systems present the company to prospective customers.

Competitive share of voice

My own visibility is only part of the picture. I also need to know how often competitors appear alongside me or instead of me.

If my inclusion rate stays at 40% month after month, that may look disappointing. But if every major competitor dropped by 20 percentage points after a model update, the story changes.

On the other hand, if one competitor jumps from 35% inclusion to 70% while everyone else stays flat, I would want to investigate what changed.

Competitive share of voice helps me separate category-wide movement from changes that are specific to my brand.

How I view the AI visibility tool landscape

The market for AI visibility platforms has grown quickly. Each product approaches the problem differently, but most are trying to answer the same core questions: does my brand appear, how often does it appear, which AI models include it, which competitors show up, and how is the brand described?

Many platforms now include prompt libraries, competitive benchmarking, citation tracking, answer monitoring, and trend reporting. These features can reduce the manual work required to test hundreds or thousands of prompts on a recurring basis.

Still, I have to be clear about what these tools are and are not measuring.

No tool has access to every AI conversation happening in the wild. Most rely on controlled prompt libraries, repeatable testing environments, or sampled interactions to create a representative view of visibility.

That is useful, but it is not the same as observing every real user interaction.

What I still cannot reliably track

This is the part I do not want to gloss over.

Even though AI measurement is improving quickly, some data is still not observable. I cannot comprehensively track every prompt where my brand appeared, every conversation that influenced a purchase, every recommendation made inside ChatGPT, every citation shown to every individual user, or exactly how personalization changed a response.

I also cannot see every multi-turn conversation across every AI platform or know how often someone acted on an AI recommendation without clicking a link.

The underlying AI platforms do not expose that level of data. If a vendor claims it can see every AI conversation involving my brand, I would ask exactly how that information is being collected.

The practical framework I would build

Rather than chasing perfect attribution, I focus on building a repeatable measurement system that I can track consistently over time.

For visibility, I would track inclusion rate, competitive share of voice, prompt coverage, and model coverage.

For response quality, I would track position within the response, brand framing, sentiment, and message consistency.

For technical signals, I would track citation frequency, content retrieval success, entity consistency, and freshness.

For business outcomes, I would look at AI referral traffic, assisted conversions, branded search lift, direct traffic trends, and pipeline influenced by AI discovery.

No single metric tells the full story. Together, these signals give me a more complete picture of how the brand is showing up and how it is being perceived across AI-assisted research.

The goal is not perfect measurement

Prompt-level visibility is not as mature as keyword tracking became over the past two decades.

Some signals are still emerging. Others remain inaccessible because AI platforms do not expose the underlying data. At the same time, user behavior is changing almost as quickly as the technology itself.

That does not mean measurement is impossible. It means the objective has changed.

Instead of trying to reconstruct every AI conversation, I focus on building a representative prompt library, tracking visibility consistently, benchmarking against competitors, and understanding how my brand is being framed.

Those trends are far more actionable than chasing a level of precision the current ecosystem cannot support.

The organizations making the most progress in AI search are not waiting for perfect attribution. They are establishing baselines, watching for meaningful movement, and adapting as both AI models and user behavior continue to evolve.

Inspired by this post on Search Engine Land.

July 6, 2026
The AI Mention Effect: Measuring Real Browse Behavior

I’m measuring downstream web browsing after AI brand mentions, focusing on what happens once a brand shows up in an AI-generated answer or recommendation.

For me, the AI mention effect is about connecting visibility inside AI experiences with real user behavior afterward, especially whether those mentions lead people to search, click, browse, and engage beyond the original AI response.

Inspired by this post on Try Profound Blog.

July 3, 2026
Why I’m Watching Google’s New YouTube Measurement Tools

I’m seeing Google expand its measurement capabilities for YouTube brand campaigns, and the goal is clear: advertisers are getting better visibility into how video ads influence engagement, brand interest, and downstream business outcomes.

What’s new: I’m paying attention to two updates in particular: Shorts Ad Actions for Video View Campaigns and Attributed Branded Searches.

Shorts Ad Actions for Video View Campaigns: When advertisers run Video View Campaigns that are opted into YouTube Shorts, they will now automatically benefit from Shorts Ad Actions in budget optimization. Google is also adding new reporting columns so advertisers can measure these interactions more clearly.

Attributed Branded Searches: Now available globally in Google Ads, this reporting metric measures branded Google searches that happen after someone sees or views a YouTube ad. I see this as a useful way to understand how awareness campaigns may influence purchase intent before a direct conversion takes place.

Why I care: It has always been difficult to connect upper-funnel YouTube campaigns with measurable business outcomes. These updates give marketers stronger signals that link brand advertising to engagement and search intent, which can make it easier to justify brand investment and improve campaign decisions.

By the numbers: According to Google, YouTube Shorts ads that generated more than 10 seconds of watch time and a like delivered 15% higher brand consideration and 20% higher brand favourability.

Google also says every additional branded search generated is associated with an average $31 increase in sales, which gives advertisers another way to connect brand activity with business impact.

Between the lines: I see Google continuing to blur the distinction between brand and performance marketing by introducing metrics that connect awareness campaigns with downstream actions. Attributed Branded Searches, especially, gives advertisers another way to show that YouTube campaigns can influence high-intent behaviour before a conversion happens.

The bottom line: Google’s latest measurement updates help advertisers better prove the value of YouTube brand campaigns by linking video engagement and branded search activity to business outcomes. For me, the bigger story is that upper-funnel advertising is becoming easier to measure in ways that matter to performance-focused teams.

Inspired by this post on Search Engine Land.

July 3, 2026
How I Measure AI Search Leads Before Optimizing

For the past two years, I have heard marketers ask the same urgent question: How do I show up in AI search?

I have seen plenty of conversation around AI optimization, visibility, and the way large language models decide which businesses to recommend. But I believe the more practical question is now becoming harder to ignore: How do I measure whether AI search is actually sending customers my way?

That is the challenge I wanted to understand more clearly.

After analyzing nearly 30 million inbound leads, I found that AI platforms are already shaping how customers discover businesses and decide to make contact. AI-generated leads still represent a small share of total volume, but they are growing steadily enough that I think marketers should start watching this channel closely.

In other words, the conversation is moving from visibility to measurement.

AI search is becoming a new attribution challenge

Traditional attribution models were built for channels like organic search, paid search, direct traffic, and referrals. AI search introduces a different discovery path, and I do not think most reporting systems are fully prepared for it yet.

A customer might ask ChatGPT for the best local HVAC company, use Perplexity to compare law firms, or ask Gemini to recommend a nearby dentist before picking up the phone.

From a marketer’s perspective, those customers may show up as direct traffic, or they may not be attributed at all. That creates a real blind spot.

If AI platforms are influencing customer discovery, I need a way to measure whether those recommendations are turning into real business outcomes.

What 30 million leads tell me

The data shows me that AI platforms are already generating measurable inbound leads for businesses. It also shows that this activity is growing over time and appearing across multiple industries, not just one category or use case.

One platform currently accounts for most AI-attributed calls, while other platforms contribute smaller shares that continue to change as customer behavior evolves. The data also reveals which industries are receiving more AI-driven calls than others.

At the same time, I have to be clear about what this dataset can and cannot measure. It does not explain why customers chose one AI platform over another, what prompts they used, or why a specific business was recommended. What it does measure is more concrete: when customers identify an AI platform as part of the journey that led them to contact a business.

That distinction matters. There is no shortage of opinion about AI search. What I need now is evidence that it is influencing customer acquisition.

Measurement should come before optimization

I understand why marketers are eager to optimize for AI search. But before investing in new tactics, I think it is worth answering a simpler question first: Is AI already driving customers to my business?

Without measurement, it is difficult to know whether greater visibility is translating into meaningful business results.

As AI search becomes another customer acquisition channel, I want to measure it the same way I measure other demand sources, including paid search, organic search, referrals, and social.

The goal is not to replace existing attribution models. The goal is to make sure those models evolve as customer behavior changes.

From visibility to measurement

The first wave of AI search focused on visibility. I believe the next wave will focus on proving business impact.

For marketers, that means moving beyond questions like, “Can customers find us?” and toward more outcome-focused questions like, “How many leads did AI actually generate?”

The businesses that answer those questions first will be better positioned to understand how AI fits into their marketing mix and where to invest as customer discovery continues to evolve.

Don’t just watch the shift. Start measuring it.

As AI search keeps evolving, I am focused on giving marketers the attribution they need to connect AI discovery with real customer conversations.

Try CallRail free at CallRail.com.

Inspired by this post on Search Engine Land.

July 3, 2026
Why Technical SEO ROI Is So Hard to Prove and Fund
Six months ago, a core update could have crushed my website. But it did not.

It did not because my team had already fixed canonicals, redirect problems, duplication issues, and JavaScript rendering gaps eight months earlier. It was the kind of unglamorous technical work that often lands with an engineer or developer because the ticket has been sitting at the bottom of the list.

And I do not really have proof. What I have is experience from years in SEO and the ability to recognize that the site had the same warning signs I have seen on sites hit hard by similar updates.

Traffic could have been cut in half. It was not.

There is no parallel internet timeline where I skipped the work, so there is no clean way to confirm what would have happened. There is no record of the disaster that never arrived.

That is why technical SEO ROI is so hard to prove. I see it as an inference problem with no control group, even though the industry often treats it like a reporting problem we can solve with one more tool.

The internet doesn’t stop

When I work in digital, I am working inside at least two open systems: the internet and the market. I could add a third if I count the maturity and expectations of internet users. I could add a fourth if I count my own website infrastructure. In reality, there are even more moving parts than that.

The point is simple: the environment I am trying to measure is always shifting, expanding, shrinking, and changing shape. There is no fixed “before” state I can pin down, and there is no clean way to model what would have happened if I had done nothing. Bayesian forecasting and similar methods can help, but they are still educated guesses.

A technical change might improve visibility today. If I make that same change six months later, it might do very little. That could happen simply because Google changed its crawl budget behavior or adjusted how it reads websites.

Cause and effect do not always stay close together in SEO. Google recrawls and reindexes on its own schedule, so the impact of a technical fix may land long after the release. By then, the result is spread across a recrawl cycle and the clean before-and-after comparison I would want for a proper test has already blurred.

As with SEO overall, there is a lot I cannot control. If I tried to track every change across the web that might influence my site, I would end up with sleepless nights and a lot more gray hair.

Technical SEO adds another layer because these changes rarely ship in isolation. It is almost never, “I made one change to the website.” It is more often, “Thirty fixes from five teams are going live on Thursday so we still have people around on Friday if something breaks.” Please do not ship on Fridays.

A lot of technical SEO also keeps the site above water. I am managing technical debt, staying current with regulations, and adapting to new releases of codebases, platforms, and frameworks. True enhancements matter, but even those can be difficult to isolate.

Technical work is closer to insurance or public health than a standard growth campaign. I usually realize how important it was only when it stops working. Much of technical SEO is disaster prevention, not new-city construction. I cannot invoice for an earthquake that did not happen.

The control group was never there

Another reality is that many technical changes, whether SEO-led or not, are sitewide because they have to be. There is no control group. Render pipelines, crawl budget, and site speed touch everything at once, so there is no untouched slice of the site left to compare against.

Two examples make this clear.
- Sunsetting 301 redirects more than a year old: The server stops reading every redirect line on every page load. The benefit is crawl and resource efficiency, but that benefit is mostly invisible in analytics.
- A migration done right: The win condition is “we did not lose traffic.” Maybe the line stays flat. Maybe it ticks up slightly. Migration work usually becomes obvious only when it fails.
My only comparison is the past, and the past existed under different external conditions. Time becomes the problem. I can compare relative movement, incremental change, and long-term trends, but the outcome shifts based on which metrics I choose and which assumptions leadership brings into the conversation.

When I can, I want to run a proof of concept. In practice, that means something close to SEO A/B testing: choose a segment, make the change there and nowhere else, measure the result, and decide what to do next. But that is not always possible, and it requires a different kind of buy-in.

I am also working in a search environment where LLMs make more things probabilistic. Answers are personalized, discovery paths are less predictable, and many of the measurements I have relied on are less deterministic than they used to be.

So I keep it relative

There are two levels of relative thinking I come back to: how I prioritize technical work and how I measure its impact.

The way I prioritize the work helps determine the impact I am trying to create.

When I prioritize technical SEO, I start with impact. How much of the website does the issue affect? How much of that impact lands on priority sections or priority pages? After that, I move into the usual scoping and grooming conversations with development teams.

For me, impact is the anchor.

Measurement and reporting are harder. A lot of the SEO industry, myself included, is now rethinking how we measure almost everything, not just technical SEO. LLMs have accelerated that shift and left many of us in an uncomfortable middle ground.

I do not have a perfect “what would have happened if…” comparison for my own website. But I do have competitors. Watching how competitor sites respond to global events, especially Google updates, is probably the closest I can get to that missing counterfactual in technical SEO. It is ROI by proxy, sitting close to share of voice.

And the funding

Technical SEO is infrastructure. It is insurance. If I am struggling to get it done or funded, I need to look closely at how I am framing the work.

At its core, I see technical SEO as insurance against the shocks of an open system. I should treat it that way. It is not always a direct revenue driver.

Yes, technical SEO can produce meaningful improvements and help the line move up and to the right. But the workhorse, the 80%, the majority of the discipline, is keeping the engine running. The work does not always promise upside. It lowers the odds and the cost of getting hit. The core update that did not sink the site is the claim that paid out.

That is why I recommend talking to finance. I want to understand how finance teams quantify, value, and evaluate insurance, security, and infrastructure.

Then I can start looking at technical SEO that way. More importantly, I can start talking about it that way.

Technical SEO is growth resilience. It is the foundation my flywheel cannot move without, not an investment I should be apologizing for.

Inspired by this post on Search Engine Land.
July 3, 2026
How I Measure Paid Social’s Real Impact on Paid Search
I’ve learned that generating demand is one of the hardest jobs in digital marketing. Measuring where that demand actually started can be even harder.

For years, I’ve seen paid search and paid social treated like separate worlds. Paid search usually gets evaluated through clicks, conversions, and ROAS, while paid social is often judged by platform-reported metrics and attributed conversions.

The challenge is that people don’t move through the buying journey in neat, channel-by-channel steps.

Someone might first discover a brand through a Meta ad, ignore it, see another ad a few days later, and eventually search for the brand or product on Google before adding something to the cart and converting. In most reports, paid search gets the credit because it captured the last click. But I don’t think that tells the full story if search didn’t create the demand in the first place.

As privacy rules, platform tracking, and attribution limits keep changing, I need better ways to understand how paid social influences search behavior. These are the practical signals and measurement methods I use to connect the two.

Signs I Look For When Paid Social Influences Search

Paid social’s impact on search is not always obvious inside attribution reports. I usually see it show up first in performance trends. These indicators help me understand whether social campaigns are building awareness that later turns into search activity and conversions.

Branded Search Volume Starts Rising

One of the clearest signs I watch for is an increase in branded search queries.

When people see a relevant, compelling social ad on Meta, TikTok, LinkedIn, or another platform, they often do not click right away. Instead, they may come back later and search for the brand name, product name, founder, or another branded term.

For example, after launching a new Meta Ads campaign, I might look for increases in searches like these:
- Brand name.
- Brand + product category.
- Brand + reviews.
- Brand + pricing.
- Brand + competitor comparisons.
I monitor these branded searches over time because they can reveal whether paid social is creating awareness that later becomes search behavior.

To do that, I review data from Google Ads, Microsoft Advertising, Google Analytics, Google Search Console, Google Trends, and any third-party SEO tools available.

I also compare trends before, during, and after major paid social launches or budget changes. If branded search volume keeps rising as paid social investment increases, I take that as a strong directional sign that social is helping generate demand.

That does not mean every increase in branded search comes from paid social. My goal is not to prove perfect causation. My goal is to find a meaningful relationship I can use to make better decisions.

I also account for other factors that can lift branded search volume, including:
- Influencer partnerships.
- Email campaigns.
- Public relations coverage.
- Seasonal demand.
- Product launches.
- Highly engaging organic social activity.
Search CTR Improves

Another signal I watch closely is click-through rate. If paid social is increasing brand familiarity, people may be more likely to click a search ad from that brand instead of choosing a competitor.

For example, someone might see Instagram video ads for two weeks and later search for a related topic on Google. When several ads appear, they may be more inclined to click the brand they already recognize.

I see the same concept reflected in brand recognition surveys that Meta and LinkedIn sometimes show in user feeds. I often find myself recognizing brands I have never purchased from simply because I have seen their ads repeatedly on social media.

That basic familiarity can still matter. It can help lift CTR on branded search campaigns, improve CTR on non-branded campaigns, and potentially lower CPCs over time.

Whenever I launch a new paid social campaign or make a significant adjustment, I compare paid search CTR before and after the change to see whether search engagement improves.

Search Conversion Rates Improve

Brand familiarity can also affect conversion rates. When people have already seen or engaged with a brand, they may arrive on the website with more trust and confidence than a completely cold visitor.

Because of that, I look for improvements in search conversion rate, lead quality, search CPA, and revenue per visitor after periods of strong paid social activity. This effect can be especially noticeable for products or services with longer consideration cycles and multiple touchpoints before purchase.

For me, conversion efficiency is one of the most useful signs that paid social is influencing downstream search behavior.

How I Validate Paid Social’s Impact on Search

The signals above give me directional insight. When I need stronger evidence, I use more structured measurement methods to evaluate whether paid social activity is actually influencing paid search performance.

Pre- and Post-Campaign Analysis

One of the simplest ways I evaluate the relationship is with a pre- and post-campaign analysis.

Before a paid social campaign launches, I benchmark key paid search metrics. Then I compare those numbers with performance after the campaign goes live.

The metrics I usually measure include:
- Branded search impressions.
- Branded search clicks.
- Search CTR.
- Search CVR.
- CPA.
- Total search conversions.
This analysis will not prove causation on its own, but it can show whether increased social activity may be influencing search performance. When I run this type of analysis, I account for seasonality, compare similar time periods, and watch for changes in competitor activity.

Geotargeted Holdout Testing

When I need stronger evidence, I consider a geotargeted holdout test. In this setup, I run paid social in selected geographic markets while withholding it from comparable control markets. Then I compare paid search performance across both groups.

For example, instead of running paid social everywhere, a nationwide advertiser could split markets into two groups:
- Test market(s): Paid social campaigns are active.
- Control market(s): Paid social campaigns are paused or excluded.
I would run the test for several weeks and monitor the same core metrics in both groups:
- Branded search volume.
- Search CTR.
- Search CVR.
- Leads.
- Revenue.
If the test markets show meaningfully stronger search performance than the control markets, I have a better basis for isolating the impact of paid social.

I like geotargeted tests because they reduce attribution bias. They let me evaluate business outcomes across similar populations instead of relying only on platform-reported conversions, which can be limited by privacy changes and tracking gaps.

If I run a holdout test, I choose comparable markets, set aside enough budget, and give the test enough time to produce statistically meaningful results. This approach usually works best for larger advertisers running regional or national campaigns. For smaller brands, I would usually start with pre- and post-campaign analysis.

Why I Measure Influence Across Channels

The relationship between paid search and paid social is often stronger than reporting platforms make it appear. I try not to evaluate these channels in isolation because they often play different roles in the same customer journey. Search captures demand, while paid social can help create it.

By digging into the data, I can find better ways to invest, build future demand, and drive conversions across platforms. Monitoring branded search, CTR, conversion rates, and structured test results gives me a clearer view of how paid social contributes to business growth.

Attribution will never be perfect. But when I measure influence across channels, I can make smarter budget decisions and build a more accurate picture of what is actually driving performance.

Inspired by this post on Search Engine Land.
July 3, 2026
Google Demand Gen Gets Gemini Creative and Reporting Boost

I’m seeing Google roll out a new set of Demand Gen updates designed to help advertisers improve creative performance, reach more potential customers across YouTube, and measure campaign results with more clarity.

For me, the bigger story is that Demand Gen is becoming less about manually adapting assets and more about using AI-assisted tools to make creative work harder across Google’s most visual surfaces.

Demand Gen campaigns are built to drive discovery and conversions across Google’s visual placements. With these latest updates, I see Google trying to reduce creative friction while giving advertisers better visibility into what is actually moving performance.

Google says the enhancements arrive as YouTube continues to show value for customer acquisition. The company cited research from Measured showing that 72% of incremental conversions on YouTube come from new customers.

What’s new. I’m watching Demand Gen add expanded video resizing capabilities, giving advertisers the ability to automatically transform creative into more aspect ratios, including vertical-to-square, vertical-to-landscape, and square-to-landscape formats.

That matters because it should make it easier to adapt existing creative for different YouTube placements without having to produce every version manually from scratch.

Why I care. Expanded video resizing can help existing assets fit more YouTube inventory, Gemini can provide AI-powered recommendations before launch, and new web-to-app measurement can give marketers a clearer view of how Demand Gen campaigns influence app installs and return on ad spend.

Gemini joins the creative workflow. Google is also bringing Gemini-powered recommendations directly into the Demand Gen campaign creation process, which makes AI guidance part of the asset selection workflow instead of a separate optimization step.

When advertisers choose image and video assets, Gemini will offer automated suggestions for optimizing creative for YouTube. I see this as a way for marketers to improve asset choices before campaigns go live, rather than waiting for performance data after launch.

Better app measurement. Demand Gen now includes Web to App Acquisition Measurement, allowing advertisers to measure when web campaigns lead users to install an app.

The new reporting gives me a more complete way to evaluate campaign performance because it attributes app installs generated through Demand Gen campaigns. That should help advertisers better understand the full impact of their media spend.

The bottom line. I see Google’s latest Demand Gen updates as a practical combination of AI-powered creative guidance, more flexible video optimization, and broader measurement tools that can help advertisers improve performance while gaining clearer insight into customer acquisition.

Inspired by this post on Search Engine Land.

July 3, 2026
Boost PPC Performance by Measuring Paid Social Impact
I sometimes find it challenging to measure the true impact of my paid social campaigns on PPC performance. Despite not always seeing conversions directly within the social platform, these ads can significantly influence my overall marketing efforts.

To truly understand how paid social affects my other marketing channels, including PPC, I’ve found a few strategies that help me set up and measure effective tests.

Step 1: Determine Your Hypothesis

I always start by clarifying what I want to learn from my tests. Defining a realistic hypothesis that I can evaluate with available data is crucial.

For example, I often use the following hypothesis to measure the influence of social traffic on PPC:
- Search lift hypothesis: Increasing social media spend will boost brand search volume and PPC CTRs.
- Logic:
  - Social ads build brand awareness, prompting more people to search for my brand during research and purchase stages.
  - As more people become familiar with my brand, they tend to click on PPC ads more, regardless of search terms, enhancing both brand and non-brand CTRs.
  - Exposure to my brand boosts trust, potentially increasing conversion rates.
- Measurement:
  - Track impression and click volume for branded terms.
  - Monitor CTR changes for brand and non-brand terms.
  - Observe conversion rate changes for these terms.
My hypothesis varies, sometimes focusing on the lift from social spend or a surge in direct traffic.

Step 2: The Test

Setting up test parameters is my next step. It’s essential to avoid simply comparing results before and after changes due to possible seasonal effects. A geographic split test is typically my go-to method.

In this test, I increase social spend in specific geographies and analyze PPC data from these areas versus others. While selecting geographies, I control for various factors, such as regional televised sports events or confined TV commercials, to ensure my test results are valid.

It’s crucial to compare control and experimental groups by similar factors like income levels and region types. I also ensure my budget can accommodate anticipated increases in social spent, preventing budget limitations from skewing results.

Evaluating the impression share before and after allows me to ensure budget constraints don’t impact my outcomes.

Step 3: The Measurement

When starting measurement, I keep it simple, comparing platform data to see changes prompted by stopping social spend across all channels like TikTok, LinkedIn, Facebook, etc.

Upon halting social spending, I’ve observed mixed conversion rate results, with some regions showing increases and others decreases, though an overall drop in conversions was common.

Depending on my analytics setup, I delve into more complex analyses, looking at conversion touchpoint differences, visitor overlap rates between social and paid search, or different attribution models.

Before initiating any tests, I ensure that my measurement capabilities are robust enough to understand and interpret results accurately.

Step 4: Evaluation Beyond Test Criteria

While running tests, I measure results against my hypothesis but also look at additional variables that may provide further insight.

In one case, a brand I tested on believed they could cut down on brand advertising without affecting their search volume. However, a drop in common brand terms contradicted this. An evaluation across various factors showed unpredictable results that required expanded analysis.

I rely heavily on my experience to sniff out anomalies and conduct further internal evaluations.

When results seem unexpectedly drastic, I question whether it’s a quirk or if other factors, like recent AI-driven changes, are silently influencing outcomes.

What to Do With Your Social Impact Tests

The test setup is straightforward:
- Define your hypothesis.
- Choose how to test, preferably using a geographic split.
- Ensure you can measure the outcomes appropriately.
- Run the tests and evaluate the hypothesis-related metrics.
- Assess additional metrics for further insights or testing ideas.
For some, social channels like Facebook are top converters, while others see poor outcomes in isolation, necessitating tests to guide budget allocation strategies.

In these scenarios, companies with substantial social media spending reduce to test impact, while others might increase spending to assess performance changes.

Results vary widely across companies, with some seeing significant performance lifts and others noticing minimal changes, underscoring the need for personalized testing.

Conducting geographic split tests can offer incredible insights into how social media campaigns bolster or detract from other marketing channels.

Inspired by this post on Search Engine Land.
April 28, 2026
Google’s New Consent Update: A Simplified Guide for Marketers

I recently discovered that Google is making significant updates to Analytics and Ads consent rules, which are set to take effect this June. This change will prioritize user permission as the key factor in how ads collect and utilize data.

Starting June 15th, the process of data collection in Google Ads will now rely exclusively on the ad_storage consent setting. This alteration removes the previous layer of complexity that came from linked Google Analytics configurations.

Previously, the flow of ad data between Analytics and Ads was governed by both Consent Mode and Google Signals settings within Google Analytics. This often led to confusion among marketers like myself, as many controls were hidden deep within the Analytics settings, rather than clearly visible in consent banners or tag implementations.

Moving forward, Google is streamlining the process. While Google Analytics data collection will still use Google Signals, Google Ads will now focus solely on whether users have consented to ad_storage.

This means that a linked Google Analytics tag will no longer influence Google’s ability to collect or use advertising identifiers.

The new update offers a cleaner, albeit more rigid, consent framework. If ad_storage consent is given, Google Ads can use all available advertising signals, including linking activity to a user’s signed-in Google account when feasible. If denied, Google will only utilize less persistent signals such as URL parameters like gclid.

This change substantially reduces ambiguity—marketers will have a clearer understanding of what drives ads data collection, with fewer options to customize what gets shared.

The primary concern here is that this adjustment makes consent settings more significant for measurement, attribution, and audience targeting. From June, whether Google Ads can leverage identifiers will depend largely on the ad_storage signal, highlighting the importance of correct consent mode setup for optimal campaign performance data.

The update simplifies some of the complexity hidden in linked Google Analytics settings, providing advertisers with more defined rules but less flexibility.

This move by Google underscores a broader strategy to enhance the understanding of consent systems for both advertisers and regulators. Having a single source of truth for ad consent could minimize implementation errors and simplify compliance explanations, but it also demands that brands ensure their Consent Mode is accurately configured.

Should consent updates be delayed or improperly configured, marketers might face gaps in measurement, attribution, and audience targeting.

Marketing teams need to take action before the June deadline by auditing their consent implementation. We should verify that Consent Mode update calls are firing correctly, and that ad_storage settings reflect users’ choices precisely. Brands with Google Signals disabled should be especially vigilant, as they could witness more Ads-linked data under the new setup if users allow ad consent.

The takeaway is clear: streamlined rules are on their way, but getting consent right will be more critical than ever.

Inspired by this post on Search Engine Land.

April 13, 2026