Marketing Attribution Is Getting More Complicated. Are We Measuring What Really Matters?

Marketing Attribution Is Getting More Complicated. Are We Measuring What Really Matters?

Marketing attribution just got a little more complicated. But should we really be surprised?

There has been a lot of discussion recently about changes to how ChatGPT sends traffic to websites. Marketers are reporting that ChatGPT is no longer consistently passing referral information or UTM parameters, which are used to help analytics platforms identify where website visitors came from. That means visitors who previously showed up in Google Analytics as coming from ChatGPT may now be showing up as Direct traffic instead.

Think about what that means for a business trying to understand whether its marketing is working. Your marketing could be performing just as well as it was last week, but your analytics might suddenly tell a completely different story. A channel that appeared to be generating traffic could look like it has stopped performing, even though the problem might actually be how that traffic is being reported.

And this brings me to something I’ve been talking about for a while. We need to stop treating marketing attribution as though it tells us the complete story.

At Elevated Marketing Solutions, we’ve always believed that marketing should be evaluated by the business problems it helps solve, not simply by the activity individual marketing channels generate. As customers continue changing how they discover, research, and choose businesses, understanding the difference between what we can attribute and what is actually contributing to business growth is becoming increasingly important.

What Changed With ChatGPT Referral Traffic, and Why Does It Matter?

In October 2026, marketers began reporting changes in the amount of website traffic they could identify as coming from ChatGPT. Research published by AI visibility company Profound on October 14 found that many links appearing in ChatGPT’s newer interface were missing tracking parameters that had previously helped websites recognize ChatGPT as the source of a visitor.

When someone clicks a link from ChatGPT and the necessary referral information is not passed along, Google Analytics may classify that visit as Direct traffic rather than attributing it to ChatGPT. Direct traffic doesn’t necessarily mean someone typed your website address into their browser. It can also include visits where the analytics platform simply doesn’t have enough information to identify the source.

Imagine that your website typically receives 100 visits from ChatGPT during a reporting period. Suddenly, your report shows only 30 visits, while Direct traffic increases. Did ChatGPT stop recommending your business? Are fewer people clicking through to your website? Did your AI search optimization efforts become less effective? Or did something simply change in how those visits were being tracked?

The answer isn’t necessarily obvious, and that’s exactly the problem. We cannot assume that every decline in reported ChatGPT traffic is caused by missing tracking information, just as we cannot assume that a decline means fewer people are discovering the business through AI search. We need to investigate what changed before drawing conclusions.

This matters because businesses make real financial decisions based on marketing reports. If you believe a channel has stopped producing results, you might reduce its budget, change your strategy, or stop investing in something that is still influencing customers. When our measurement tools cannot tell the whole story, we have to be much more thoughtful about how we interpret the information they provide.

What Is Marketing Attribution, and What Does It Actually Tell Us?

Marketing attribution is the process of identifying which marketing interactions receive credit when someone completes a desired action, such as submitting a lead form, scheduling an appointment, making a purchase, or contacting a business.

For years, businesses have wanted to know which marketing channel is producing the most leads. Is it social media? Google Ads? SEO? Email marketing? Referrals? It’s a fair question. When you’re investing money in marketing, you want to know what’s working, what isn’t working, and where your dollars should go next.

Analytics platforms have helped us answer parts of that question by tracking how people arrive at our websites, what they do once they get there, and whether they complete certain actions. Depending on the tracking technology and attribution model being used, we can also understand some of the interactions that happened before a conversion.

But there is a limitation that we don’t talk about nearly enough. Marketing attribution can only report on the information it can observe and connect. It cannot automatically account for every interaction that influenced someone’s decision, and it certainly cannot tell us everything that happened before someone became a customer.

Even when tracking works exactly as designed, it doesn’t necessarily tell us why someone chose your business. And the reality is that the customer journey isn’t nearly as simple as clicking one advertisement and immediately becoming a customer.

Why Is It So Difficult to Give One Marketing Channel Credit for a Sale?

Let’s think about how someone might actually choose a business. A potential customer sees your company mentioned on LinkedIn. A few days later, they notice your business sponsoring a community event. They aren’t ready to contact you yet, but now your company name is familiar. A couple of weeks later, they have a need for your services and decide to ask ChatGPT about businesses that might be able to help.

Your company is included in the response, so they visit your website, read about your services, and look at some of your client reviews. They’re interested, but they’re still not ready to make a decision. Another week goes by, and they search Google for your company name, return to your website, and finally call you.

So which marketing channel gets credit for that lead?

Was it social media because that’s where they first encountered your business? Was it the community event that made your company recognizable? Was it ChatGPT because it helped them research their options? Or was it Google because that’s how they returned to your website before calling?

Depending on the attribution model, tracking technology, and information available, your marketing report might assign credit to one of those channels while overlooking the others. But would that customer have contacted you if they hadn’t experienced several of those interactions along the way? We don’t know for certain, and that is the point.

Marketing channels influence people in different ways, often over an extended period. Some introduce your business to potential customers. Others educate them, build trust, reinforce your credibility, or make it easier for them to take the next step. Someone might encounter your company six or seven times before deciding to reach out, and not all of those interactions will happen in a way that analytics can track.

When we try to reduce that entire journey to one source, we risk misunderstanding what’s actually generating business. More importantly, we risk making decisions that weaken the very marketing activities that helped create the opportunity in the first place.

Does This Mean Marketing Attribution No Longer Matters?

Absolutely not. I want to be very clear about this because I’m not suggesting that we stop tracking marketing performance or abandon attribution. Quite the opposite. I believe we need to be more intentional about measurement, understand what our reports are telling us, and recognize where they have limitations.

Every marketing channel has its own metrics that help us understand whether it’s doing its job. If we’re running Google Ads, we need to know whether those ads are generating qualified inquiries and whether the cost makes sense for the business. If we’re investing in SEO, we need to understand whether our visibility is improving for searches that matter to our customers. If we’re sending emails, we need to evaluate whether people are engaging with those messages and taking the next step.

The same applies to social media, video marketing, events, sponsorships, public relations, and every other marketing activity. Each one has a purpose within the larger strategy, and we should be evaluating its performance against that purpose. The mistake is expecting every individual channel to receive direct credit for a sale in order to prove that it has value.

We also need to remember that different types of marketing serve different stages of the customer journey. A social media campaign introducing your business to people who have never heard of you shouldn’t necessarily be evaluated the same way as a Google Search campaign targeting people who are actively looking for your services. Both can contribute to business growth, but they are doing different jobs.

The problem begins when we treat attribution reports as the final authority on marketing success. Attribution is one source of evidence about what’s happening with your marketing. It isn’t the complete picture, and it shouldn’t be the only information guiding your marketing decisions.

What Should Businesses Be Measuring Beyond Marketing Attribution?

This is where I think we need to change the conversation about marketing performance. Instead of looking at marketing as a collection of independent channels and asking which one deserves the most credit, we need to evaluate how those channels work together to move people through the customer journey and ultimately support business growth.

That doesn’t mean ignoring individual channel performance. It means connecting those measurements to the larger questions we should be asking about our businesses.

Are the Right People Finding Your Business?

Before someone can become a customer, they need an opportunity to discover your business. That might happen through Google, ChatGPT, social media, paid advertising, community involvement, industry publications, referral partners, or traditional media. Each of those channels can help introduce your business to potential customers, but not every interaction will immediately result in a website visit or a lead.

We should be evaluating whether our marketing is reaching the audiences most likely to need our products or services. Search visibility, branded searches, relevant website traffic, audience reach, referral activity, and engagement can help us understand whether we’re becoming more visible to the right people.

We also need to recognize that someone can become familiar with your business without immediately taking an action we can measure. A person might see your company mentioned several times before they ever search for you directly. That exposure may influence their eventual decision, even if we cannot connect it to a specific conversion.

Are We Building Enough Trust for Someone to Consider Our Business?

Getting noticed is only one part of marketing. Once someone discovers your business, they need to decide whether you are credible, whether you understand their problem, and whether they should consider you over another company.

This is where your website, educational content, social media presence, customer reviews, case studies, community reputation, and consistent messaging work together. A potential customer may visit your website several times, read a blog post, look at reviews on Google, and check your social media accounts before they feel comfortable reaching out.

We should be looking at how people engage with these resources and whether our marketing is helping them move closer to a decision. Are visitors exploring our services? Are people returning to our website? Are we seeing more searches for our business name? Are our reviews and reputation helping reinforce our credibility?

None of these measurements alone proves that someone will become a customer. But together, they help us understand whether we’re giving people the information and confidence they need to consider doing business with us.

Are We Generating More Qualified Business Opportunities?

This is where I believe more business owners and marketers need to focus their attention. Are more people contacting the business, and are they the right people?

A service-based company might need more qualified calls and estimate requests. A healthcare organization might need more appointments for a particular service line. An educational organization might need more applications and enrollments, while an event might need registrations or ticket sales.

What we’re measuring should be connected to what the business is actually trying to accomplish. Generating more leads isn’t necessarily a success if those leads aren’t qualified, aren’t in the right geographic area, or aren’t interested in the services the business needs to grow.

We also need to follow those opportunities beyond the initial inquiry. If marketing generates 100 leads but only five are qualified, that’s a very different result from generating 40 leads that produce 20 meaningful sales conversations. Looking only at the number of conversions reported in Google Analytics wouldn’t necessarily reveal that distinction.

Are Those Opportunities Turning Into Customers and Revenue?

Generating inquiries is not the same as generating business. What happens after someone contacts you matters just as much as how they found you.

Does your team respond quickly? Are appointments being scheduled? Are estimates turning into approved jobs? Are proposals converting into customers? Are people following through with registrations, appointments, or purchases?

Marketing can influence these outcomes, but it doesn’t control every part of the sales process. That’s why we need to connect marketing reporting with sales data, customer relationship management systems, scheduling information, and actual business results whenever possible.

Ultimately, we want to know whether marketing is contributing to qualified opportunities, customer acquisition, retention, and revenue growth. If those outcomes aren’t improving, we need to understand where the problem is occurring before assuming that more website traffic or advertising will solve it.

How Do You Measure Marketing When You Can’t Track Every Customer Interaction?

This is the practical challenge businesses are going to face more frequently, and I don’t believe the answer is to keep searching for a single attribution model that perfectly explains every customer decision. We should absolutely continue improving our tracking, but we also need a more complete approach to evaluating performance.

The first part of that approach is continuing to measure individual marketing channels. Google Ads, SEO, social media, email, website analytics, and other marketing activities still provide valuable information. We should be evaluating those metrics regularly, comparing results over time, identifying meaningful changes, and making adjustments based on what we learn.

For example, a sudden decline in reported ChatGPT referrals should prompt us to investigate potential tracking changes, Direct traffic patterns, landing-page activity, and other available information before concluding that AI search visibility has declined. The same principle applies to any unexpected change in marketing performance. We need to understand what happened before deciding what to do about it.

The second part is measuring the combined business outcomes. We should be looking at total qualified inquiries, phone calls, appointments, proposals, registrations, new customers, and revenue. Those numbers should be evaluated against business goals, seasonal patterns, marketing investment, and operational capacity.

If your marketing is reaching more qualified people, your sales opportunities are improving, and revenue is growing, those outcomes deserve attention even when attribution reports cannot explain every influence. Of course, we also need to be careful not to assume that all business growth came from marketing. Pricing changes, economic conditions, referrals, market demand, and sales performance can all influence results. Our goal should be to understand marketing’s contribution as accurately as possible, not claim credit for everything that happens.

The third part is something I think businesses overlook far too often, and that’s actually talking to their customers. Ask people how they first heard about your business, what made them decide to contact you, and whether they encountered your company somewhere before they were ready to buy.

You might learn that someone first saw your business on LinkedIn, remembered you from a local event, researched you through ChatGPT, and eventually searched Google to find your phone number. Those conversations can help reveal connections between your marketing activities that analytics alone cannot identify.

Customer feedback isn’t a perfect attribution system either. People don’t always remember every advertisement, article, or interaction that influenced them. But when we combine what our customers tell us with channel performance and actual business outcomes, we develop a much more useful understanding of how our marketing is working.

How Is AI Search Changing the Way We Need to Evaluate Marketing?

I think the recent discussion about ChatGPT referral tracking is just one example of a much larger change happening in marketing. For years, we’ve relied heavily on website traffic to evaluate whether our marketing is generating interest. But now someone can research your company through an AI-powered search experience and learn a significant amount about your business before ever visiting your website.

A potential customer might ask ChatGPT to recommend businesses that provide a particular service, compare several companies, or explain what to look for when choosing a provider. They may receive information about your business, read a summary of your services, and begin forming an opinion about whether you’re a good fit.

Sometimes they click through to your website. Sometimes they don’t. They might remember your company name and search for it later, ask someone they trust about your business, or contact you through a different channel.

Google’s AI Overviews, AI Mode, ChatGPT, and other AI search experiences are changing how people gather information and evaluate businesses. That creates two separate challenges for marketers. We need to understand whether our businesses are appearing when potential customers research relevant problems and solutions, but we also need to recognize that those interactions may not always create measurable website traffic.

A business may influence someone’s decision through AI search without receiving a directly attributable referral. That doesn’t mean the interaction had no value, but it also doesn’t mean that every AI mention is producing meaningful business results.

This is why we need to evaluate AI search visibility alongside traditional search performance, branded search activity, customer inquiries, and business outcomes. We should be asking whether people are discovering our businesses, whether they are considering us, and ultimately whether they are choosing us.

Showing up matters. Being considered matters. Being chosen matters. And our measurement needs to reflect all three.

Could Focusing Too Much on Attribution Actually Hurt Your Marketing Strategy?

Yes, and this is probably my biggest concern about how businesses interpret attribution reports.

When businesses put too much weight on which channel receives credit for a conversion, they can begin investing only in the marketing activities that are easiest to track. On the surface, that might seem like a responsible way to manage a budget. If one channel appears to generate more leads than another, why wouldn’t you put more money toward the channel that’s producing results?

The problem is that the channel receiving credit for the conversion may not be the same channel that originally created the demand.

Think about a Google Search advertisement that generates a lead because someone searches for your company name and clicks the ad. That advertisement may have helped the customer take the final step, but what made that person familiar with your company in the first place?

Maybe they saw your educational content on social media. Maybe they heard you speak at a local event, read about your business in an industry publication, or received a recommendation from another customer. Maybe ChatGPT mentioned your business while they were researching their options.

If we continually move our marketing budgets toward whatever receives the most direct attribution, we risk underinvesting in the activities that create awareness, build credibility, and help people develop a preference for our business before they’re ready to buy.

Over time, that can become a business problem. You cannot rely exclusively on capturing existing demand if you’re not also helping future customers discover, understand, and remember your company.

That doesn’t mean every awareness campaign deserves an unlimited budget or that branding should be exempt from accountability. It means we need to evaluate marketing activities according to the roles they play in the customer journey and the evidence available about their contribution.

A good marketing strategy connects those activities rather than expecting each one to operate independently. And a good measurement strategy needs to do the same.

What Should You Be Asking Your Marketing Agency About Attribution and Results?

If your marketing agency is showing you reports filled with impressions, clicks, website visits, and conversions, that’s a starting point. Those numbers provide useful information about what’s happening in your marketing channels, but they should also lead to a larger conversation about what those results mean for your business.

You should be able to understand whether your marketing is reaching the right audiences, whether qualified opportunities are improving, where potential customers may be dropping out of the buying process, and what your agency recommends doing next.

When something unexpected happens, like a sudden decline in reported ChatGPT referral traffic, your agency should be investigating the possible explanations before recommending major changes. A decline in traffic doesn’t automatically mean your marketing stopped working, just as an increase in traffic doesn’t automatically mean your marketing improved.

And when an individual channel receives credit for generating a lead, your agency should be looking beyond that single conversion to understand how other marketing activities might be contributing to the overall results.

The real value of marketing reporting isn’t simply knowing what happened. It’s understanding what those results mean, what we can reasonably conclude from the available information, and how that knowledge should influence our next decisions.

That’s the conversation business owners should expect their marketing partners to be having with them. Not simply whether traffic went up or down, but whether the overall marketing strategy is helping the business accomplish what it set out to do.

A Few Questions You Might Still Have About Marketing Attribution

If ChatGPT Traffic Is Showing Up as Direct, Can I Still Measure AI Search Performance?

Yes, but you need to recognize that the information will be incomplete. You can continue tracking identifiable AI referral traffic, monitoring relevant AI search visibility, evaluating changes in Direct traffic, and reviewing how visitors interact with your website. You can also ask customers whether AI tools influenced their research. None of these measurements can reconstruct every AI-assisted customer journey, but together they can help you better understand how AI search may be contributing to your marketing performance.

Does Direct Traffic Mean Someone Typed My Website Address Into Their Browser?

Not necessarily. Direct traffic generally means that the analytics platform could not identify another source for the visit. While this can include someone typing your website address or using a bookmark, it can also include visits from links that don’t pass sufficient referral information. That’s why an increase in Direct traffic shouldn’t automatically be interpreted as an increase in people intentionally navigating to your website.

Should I Stop Using Google Analytics to Measure Marketing ROI?

No. Google Analytics remains an important tool for understanding website behavior, acquisition patterns, and measurable conversions. The key is recognizing that Google Analytics cannot provide every piece of information needed to evaluate marketing ROI. When possible, you should combine website analytics with customer relationship management data, sales information, customer feedback, and other business performance measurements.

If My Revenue Is Growing, Doesn’t That Prove My Marketing Is Working?

Not necessarily. Revenue growth is an important business outcome, but there are several factors that can influence it, including pricing changes, market demand, existing customer relationships, referrals, and sales performance. Marketing should be evaluated based on its objectives and the evidence of how it contributes to those outcomes. The goal isn’t to give marketing credit for every dollar of growth. It’s to understand whether your marketing investment is supporting the business results you’re trying to achieve.

How Do I Know Which Marketing Channels Deserve More Investment?

Start by understanding what each channel is supposed to accomplish within your overall strategy. Evaluate its performance against that objective, the quality of opportunities it influences, its cost, and how it supports other marketing activities. When possible, use historical comparisons and controlled testing to understand whether changes in investment actually improve business outcomes. Rather than assuming that the channel receiving the most conversion credit is always the best investment, consider what might happen to the entire customer journey if that channel were reduced or eliminated.

Marketing Doesn’t Work in Isolated Channels. Why Do We Keep Measuring It as Though It Does?

Marketing attribution has always had limitations. The recent changes being reported with ChatGPT have simply made those limitations harder to ignore.

As AI continues changing how consumers discover, research, and choose businesses, I think we’re going to have to get much more comfortable evaluating the collective impact of our marketing efforts rather than expecting every lead to come with a perfectly documented journey.

That doesn’t mean abandoning accountability. In fact, I believe it means we need to become even more accountable for understanding what our marketing is actually accomplishing.

We need to know whether the individual channels we’re investing in are performing the jobs they’re supposed to perform. We need to understand how those channels work together to help customers discover, evaluate, and choose our businesses. And we need to connect our marketing measurements to qualified leads, sales opportunities, customer relationships, and business growth.

At Elevated Marketing Solutions, our philosophy has always been to start with the business problem and build the marketing strategy around what the business is actually trying to accomplish. Measurement is a critical part of that process, but the purpose of measurement isn’t simply to assign credit to a channel. It’s to help us understand what’s working, what’s not, and what we should be doing differently to help the business grow.

Attribution still has value. It just shouldn’t be mistaken for the whole picture.