Why website analytics tell an incomplete story of buyer decisions

Marketing dashboards still reward visits, click-through rates, and engagement as if attention proved demand. For B2B companies, a website visit rarely leads directly to a signed engagement. Buyers consult colleagues, review evidence, ask for referrals, and return months later. Automated traffic complicates the picture further.

When I rebuilt jaymandel.com using ChatGPT Sites, I wanted to test a different approach — the website should be an active source of business information, evidence, and decision support, rather than a static brochure or traffic generator. Web tools make content iteration fast and accessible, but speed alone doesn’t build trust. Without strategic clarity about the buyer’s decision process, publishing at scale merely generates more noise rather than meaningful outcomes.

The business relationship is the unit of analysis

Jacob Sanders of the Marketing Accountability Council described a pattern in his work: “I get a lot of traffic from people that are already in the pipeline.”

Those visits may help an existing prospect prepare for a meeting, circulate a recommendation, or check a proposal’s credibility. They may never appear as a new lead conversion.

Moni Oloyede described another form of influence: “When there is a need, I am a preference.” Her audience may encounter her work long before they need to buy. As she put it, “I don’t have to give you a click every time I see you.”

Clicks are observable actions. Confidence, preference, and recommendation are harder to observe. Neither a click nor its absence proves a buying decision.

What I changed on my own website

My previous site emphasized capabilities — marketing, strategy, leadership, data, and execution. I reorganized it around four problems clients bring me — clarify a consequential problem, make a decision, build team capability, and move work forward.

I added clearer explanations of the work, fit criteria, case studies, pricing signals, articles, and practical evidence. The Carousel of Truth makes gaps between promises and delivery concrete. A prospective client should be able to ask: Does this person understand my problem? Can I examine the reasoning? Can I defend this choice to my colleagues?

The website should also help a journalist understand my perspective and a partner accurately explain my work. That’s useful even without a tracked conversion.

Recommended Articles

The benefits of server-side tracking — and what it takes to get them

Oct 13, 2026

Why HIPAA compliance in martech goes beyond BAAs

Oct 9, 2026

20 years and billions of dollars later, CX still has the same problem

Oct 7, 2026

Design around buyer decisions

Working with Wendy Louise Wilder introduced me to a more disciplined way of thinking about the decisions buyers need to make. Wilder developed the BASE3 methodology, a framework that maps customer decision stages against the information and communication channels available to support them.

That framework helped me reconsider the site from the buyer’s point of view. Someone defining a problem needs a different explanation from someone comparing consultants or preparing to recommend one internally. The service pages, evidence, glossary, articles, and conversation interface address those different information needs.

SingleFig, Wilder’s CRM and project-management system, provided another perspective — a website interaction is one event in a relationship that continues through conversations, follow-ups, referrals, and actual work. The goal is to connect what buyers need to know with what happens next, rather than assume every visit starts a new funnel.

My rebuilt website demonstrates the method in practice, but it doesn’t prove that the method increased revenue. The broader lesson is that website design should begin with customer decisions, and measurement should follow those decisions across channels.

(Disclosure: Wilder is my business partner. I have a professional interest in BASE3 and SingleFig.)

Machines aren’t the same as buyers

Automated requests aren’t all equivalent. 

• Search crawlers help index content.
• Social preview bots retrieve page information.
• Monitoring tools check availability.
• AI systems may crawl or retrieve information.
• Malicious scrapers and credential attacks can consume resources or distort server logs.

Google Analytics 4 excludes known bots and spiders, but its reporting isn’t a definitive census of humans. Some automated activity can still appear in analytics, while many crawlers never execute the analytics tag. A zero-second visit, a particular country, or a data-center location is a clue, not proof of bot activity. Conversely, an engaged session isn’t proof of a qualified buyer.

Relying heavily on simple attribution metrics can lead to severe misattribution issues. Practices such as cookie stuffing disingenuously manipulate conversion tracking to claim credit for deals or actions that would have occurred anyway, further skewing analytics and misdirecting marketing strategy.

Information from a website may reach a prospective buyer through search results, AI-generated answers, or a colleague’s summary without a conventional website session. A crawler request demonstrates access, not recommendation or commercial influence. Those claims require separate evidence.

This makes accurate, current, and clearly structured service information an ongoing responsibility. When an offer changes, the website must change with it.

Measure the decision the site helps support

For my business, meaningful outcomes include qualified conversations, proposals, paid engagements, repeat work, referrals, and relevant media opportunities. I’d connect those outcomes to what buyers say they encountered and used — not claim that the last click caused the sale.

The practical questions are:

• Are suitable buyers considering us? Examine introductions, qualified inquiries, and opportunity records.
• Do they understand the offer? Review inquiry quality, meeting feedback, and repeated questions.
• Does the evidence help them decide? Ask what case studies, explanations, or examples they used.
• Does interest become profitable work? Track proposals, wins, contribution profit, and repeat engagements.
• What needs to change? Record confusion, missing proof, lost opportunities, and broken handoffs.

Website analytics can support these questions. It can’t answer them alone.

A practical measurement reset

• Ask every new prospect how they heard about you, what they reviewed, and what gave them confidence. Don’t force a single source when several influenced the decision.
• Capture useful content interactions in the CRM alongside opportunity stage, sales conversations, and outcomes. Avoid interpreting tracked document views as proof that a particular person read or believed the material.
• Segment suspected automated traffic cautiously, using analytics, server-side evidence, and known bot signatures where available. Keep infrastructure monitoring separate from marketing performance.
• Review wins and losses. Ask customers and prospects which website materials clarified the offer, which raised questions, and what was missing.
• Compare commercial outcomes with the full cost of building and maintaining the site. Distinguish observed contribution from incremental profit attributable to the investment.

How to talk about website ROI

Website ROI discussions should start with what the site is supposed to accomplish:

“We built this site to help suitable buyers understand your offer, evaluate evidence, and advance decisions. We’ll examine its contribution alongside qualified opportunities, revenue, and customer feedback. We’ll track the cost of keeping the information useful. A precise ROI figure requires evidence of incremental profit attributable to the investment. Until then, we’ll report observable outcomes, supported contributions, and gaps.”

That answer is less flashy than a traffic-growth chart. It’s also more accountable.

The work continues after launch. Questions from buyers should change the copy. New evidence should replace weak claims. Failed handoffs should be fixed. The website is a maintained business system, not a one-time deliverable.

The question changes

My rebuild made my business easier to examine. I haven’t yet established that it produced a paying client, and I won’t claim otherwise.

The question is: When someone has a problem I can solve, does my website help them understand why they should consider me — and give them enough evidence to decide?

That’s the standard I intend to measure.