Why we started Search Intelligence.

I did not begin my career in marketing.

I was trained in mechanical engineering, low-current electrical engineering, and high-current electrical engineering. After completing my studies, I moved into information technology and became a Microsoft Certified Systems Engineer.

That background shaped the way I approach problems.

Engineering teaches you to look beneath the visible symptom. It teaches you that systems fail for reasons, that outputs are produced by interacting inputs, and that surface-level explanations are rarely enough. When something does not make sense, you do not simply accept the report. You investigate the system.

In 2009, I entered digital marketing through data analytics and later moved into search engine optimisation.

From the beginning, something troubled me.

The same explanations were being repeated across the industry. The same sources were being consulted. The same tools were being used. The same measurements were being reported. Regardless of whether the company operated in insurance, banking, retail, or another industry, marketing and SEO teams were largely working from the same underlying information.

That led me to a simple question:

If everyone is singing from the same song sheet, how can anyone expect to produce a meaningfully different outcome?

The issue was not that the available tools had no value.

The issue was that they created a narrow view of a much larger system.

Most organisations looked at their own website, their own traffic, their own rankings, and their own historical performance. The website was treated almost like an island, as though its performance existed independently from competitor movement, algorithm changes, technical infrastructure, customer behaviour, market conditions, and the wider digital environment.

But no website is an island.

A search engine does not assess a business in isolation. It compares that business with every other relevant option competing for the same customer, the same query, and the same moment of decision.

Yet much of the industry continued to report only on what had been defined as a KPI.

What was missing was everything outside that narrow reporting frame:

This is where survivorship bias enters the system.

Only the information that fits the existing reporting model survives. Everything else is filtered out, ignored, or recognised too late.

Explore the Reporting Blind Spot

In 2019, I began working with one of South Africa’s major insurance businesses.

By then, I had become convinced that using the same tools and the same data in the same way as every competitor would not create a different result.

So, in 2020, we began developing our own tools.

Instead of relying primarily on retrospective website analytics, we began tracking the entire competitive environment.

In the South African car-insurance category, we identified the keywords being used across the vertical, tracked the competitors appearing for those searches, indexed the results every day, and built a historical record of how the market was moving.

That fundamentally changed the way we understood performance.

We were no longer asking only:

“Did our rankings improve?”

We could ask:

That change removed much of the knee-jerk culture that exists in SEO.

A Monday-morning ranking drop no longer meant the entire strategy had failed. We could see whether the movement affected the whole market, a competitor group, a device, a keyword segment, or only the business itself.

We moved from reaction to interpretation.

From isolated metrics to competitive context.

From reporting what happened to understanding why it happened.

In 2023, the insurer reached an average position of number one in South Africa for the exact search term “car insurance” across mobile and desktop.

Across approximately 300 keywords in the broader car-insurance vertical, the business achieved an average position of five.

In doing so, it displaced Hippo, which had held the leading position for approximately 11 years.

What made this result especially important was what the conventional measures could not explain.

By the common measurements traditionally used to predict success, it should not necessarily have outperformed the market.

But it did.

That was the moment we knew the difference was not simply execution.

The difference was the way the environment was being observed, interpreted, and managed.

We were not optimising only for what was visible in a standard report. We were building a more complete understanding of the category, its technical foundations, competitive movements, search behaviour, customer intent, and structural risks.

The model later demonstrated predictive value as well.

In June 2024, we identified a smaller competitor building enough momentum to move into a leading position in the car-insurance category. We could see the risk developing approximately six months before it became apparent through conventional reporting.

That changed the role of search data.

It was no longer only historical.

It became an early-warning system.

Read the Full Evidence Story

As our approach evolved, we repeatedly found issues outside the normal SEO reporting stack.

In one case, rankings initially remained stable while lead generation began to decline.

The standard technical indicators did not explain the change.

Page-speed reporting did not reveal the full problem. Conventional SEO reports did not show an obvious failure. But after deeper investigation, the cause was traced to ISP latency affecting the user experience and, over time, search performance.

That issue had consequences beyond organic search.

The same landing pages were being used in paid campaigns. If a page responded slowly or failed technically, both organic and paid performance suffered.

This reinforced a principle that has become central to Search Intelligence:

Technical integrity is the foundation. Everything else is built on top of it.

Today, there is significant pressure on businesses to appear in AI-generated results. But before investing heavily in AI visibility, organisations should first understand whether the digital foundation on which that visibility depends is stable, accurate, technically sound, and governable.

AI did not make the fundamentals irrelevant.

It made them more important.

Explore Technical Integrity

By 2025, the growth of AI search made the limitation of the old SEO model impossible to ignore.

Search was no longer only a list of ranked links.

Customers were encountering brands through search results, reviews, comparison platforms, publisher content, AI answers, technical experiences, and competitor narratives.

The business was being interpreted across an entire ecosystem.

Yet most organisations still had no formal way to understand or govern the risks created by that environment.

Executives could be held responsible for digital risk without having the visibility required to understand what risk existed.

That became unacceptable to me.

Search had become too commercially important, too technically interconnected, and too influential in customer decision-making to remain a reactive marketing activity.

It needed to become a governed business function.

That is where Search Intelligence emerged.

But as the intelligence and governance layer between executive leadership and execution.

Search Intelligence does not perform SEO on behalf of an organisation.

We sit in the space between leadership and the teams responsible for execution.

We are agency-agnostic and team-agnostic.

Traditional reporting usually focuses on the outputs that were requested.

We seek to report on the whole observable environment:

This is why the operating model is:

SEOLens detects. Human Intelligence interprets. Search Governance controls.

Technology gives us breadth, consistency, speed, and evidence.

But technology alone is not enough.

AI can detect patterns. It can accelerate analysis. It can process extraordinary amounts of information.

What it does not possess is lived commercial experience.

It does not inherently understand the nuance of a market, the history of a client, the consequences of a technical decision, the politics of execution, or the difference between a theoretical risk and a materially important one.

Human judgement remains essential.

Data does not reduce risk.

Correct decisions reduce risk.

Explore Human Intelligence

Over time, the data also revealed something more important than rankings.

Customers do not move directly from seeing a brand to buying from it.

They move through a sequence.

Most organisations invest heavily at the edges.

They invest in visibility through media and acquisition.

They invest in demand through conversion activity.

But the decision is often made in the middle.

Trust and influence determine whether visibility becomes revenue, or whether the customer chooses someone else.

Search is one of the primary environments in which those middle layers are formed.

This is the fundamental belief behind Search Intelligence:

Before customers choose, they search for confidence.

Search Intelligence exists because the moments in which customers make decisions should not be left unmanaged.

We believe organisations should understand:

We believe leadership should be able to see the risks for which it is responsible.

We believe agencies and internal teams should be judged in the context of what they can control.

We believe search reporting should create clarity, not noise.

We believe activity is not the same as progress.

We believe visibility is not the same as trust.

And we believe being found is not the same as being chosen.

Search Intelligence is built around a simple promise:

Govern the moments customers decide.

Our purpose is not to make organisations chase every fluctuation.

It is to help them identify the signals that matter.

Our manifesto is equally simple:

Make search worthy of customer trust.

Because customers should be able to trust what they find.

And organisations should be able to govern the environments in which that trust is formed.

That is why Search Intelligence exists.

See Where Search Intelligence Sits