Your website gets visitors. What happens next?

Most website reports begin with an arrival: visits, sessions, page views, perhaps a channel. Those numbers are useful. They tell you whether attention reached the door. They do not tell you whether the visitor understood the proposition, found a useful next step, or left with the question they came to answer still unresolved.

That gap matters because traffic can improve while the product journey remains weak. A campaign can bring more people to a page that fails to orient them. A field note can earn serious reading without helping someone discover the related product. A visitor can open an inquiry and stop before they have enough confidence to send it. The arrival moved. The outcome did not.

Field signal/ 01

A visit is evidence of arrival. It is not evidence of understanding, intent, or value.

Begin with the decision, not the dashboard

Before choosing a chart, write down the decision you are trying to make. Do we need a clearer first screen? Is the article helping the right people reach the product? Are visitors exploring a service and then losing the thread? Is an inquiry journey asking for commitment before it has earned it? A precise question gives the measurement a boundary.

Then define the moments that would make the question observable. For an article, that might be a real reading start, meaningful progress, and a move into a relevant product story. For a product page, it might be seeing the core proof, selecting the product, and following the live-product link. For an inquiry, opening a dialog and completing valid details are different moments; neither should be renamed as a qualified lead.

The wording is not cosmetic. A click is a click. A valid submission attempt is not proof that a message was delivered, read, or accepted. A person who has not completed the next step may still be inside the time allowed. Honest names stop observations from becoming conclusions as they travel through a team.

A meaningful action needs a definition

Analytics becomes useful when a business moment has an explicit rule. “Read the article” could mean active time plus sufficient depth, not merely loading the route. “Explored the product” could mean selecting its story, not seeing a card flash into view. The definition should say what happened, what properties were required, and which version of that meaning was used.

This is one of the ideas behind Presagio: raw activity can arrive before a team agrees on its business meaning. In the private beta, retained event names and schemas can be inspected, then promoted into versioned definitions. Published journey revisions stay pinned to the definition versions selected for them; a later wording or rule does not silently rewrite the old journey.

That traceability is less dramatic than a rising chart, but it is what makes comparison possible. If the product changes what “reached value” means, the team can publish a new definition and a new journey revision deliberately. The previous interpretation remains available instead of being edited out of history.

Read the journey as movement

A journey connects ordered moments within a stated window. It asks how many anonymous visitors entered, how many reached the final milestone in order, and how many are incomplete. That last group needs care. Some people have exceeded the allowed time; others are still pending and cannot honestly be labelled as failures.

Imagine 100 visitors begin a setup journey. Forty complete it. Twenty have not completed it but still have time. The completion rate is 40 percent at that observation, yet only 40 of the 60 incomplete entrants have actually passed their deadline. Treating all 60 as drop-offs would turn an unfinished observation into a false diagnosis.

A weak transition is a place to investigate, not a verdict. Open the evidence for the relevant group. Look at the ordered milestones and occurrence times. Compare people who continued with people whose window elapsed. Then bring in customer conversations, usability work, support history, and product context. Behaviour narrows the question. It rarely explains the reason on its own.

Field signal/ 02

The useful question is not only where people stopped. It is what evidence would justify changing the product.

Attribution explains a rule, not a cause

Campaign context belongs in the journey because the first visit is rarely context-free. A tagged LinkedIn post, a search result, an email, and a later direct return may all be part of the observed path. Preserving source, medium, campaign, and content across navigation helps keep that entry context attached to the visit instead of losing it after the landing page.

Attribution is a rule for assigning credit, not proof of causation. First touch can answer which observed source began the story inside the lookback window. Last touch can answer which eligible touch came before the conversion. Last non-direct can avoid handing all credit to a later direct visit. The journey can remain identical while the model changes who receives credit.

That is why the result should remain inspectable: the touch, the conversion event, the lookback, and the rule behind the assignment. An attributed campaign may deserve further investment; it does not, by itself, prove that the campaign caused the outcome. Incrementality needs a different kind of evidence.

Shipped today in private beta

Presagio's current private beta can receive consent-aware browser events, show stored activity, discover event types, publish versioned definitions, compose two-to-five-milestone journeys, and open bounded anonymous evidence behind journey counts. Its traffic and campaign views preserve the distinction between an observed arrival and assigned conversion credit. Optional masked replay has a separate consent choice and is linked to evidence without recording form text or page content.

The boundaries are equally important. An anonymous visitor is scoped to a source; the beta does not claim to merge browsers, devices, or separate sources into one person. Journey evidence is deliberately bounded and the interface labels truncation. Retention and query windows are explicit. Consent withdrawal stops future collection locally, while server-side deletion remains an operator-assisted process rather than a self-service screen.

The public Presagio demo is different from a live customer integration. It uses fictional sample data and simulated replay so people can understand the investigation flow without mistaking a demonstration for production evidence. The product is available through a controlled private beta; broad self-serve onboarding, billing, and general commercial availability are not being presented as shipped.

Configured on The Dev Guys

The Dev Guys is a real first-party integration. On production routes, our analytics client records manual page views and deliberate events such as reading progress, section attention, product discovery, service exploration, outbound product visits, and stages of the project-inquiry journey. It strips query strings from page paths, bounds accepted properties, and does not send names, email addresses, or inquiry text into analytics.

That lets us configure journeys around questions we actually have. One can begin with an article arrival, continue through meaningful reading progress, and end at the related product story. Another can begin with product discovery and continue to an outbound product visit. A third can distinguish opening an inquiry from starting it and reaching a valid ready state. These are configured measurement journeys, not claims that each visitor had a particular intention.

The same care applies to this article's campaign link. Its LinkedIn source, social medium, campaign, and founder-post content tag can be captured on arrival and preserved across the single-page site. We can inspect the observed path from the post into the article and onward to Presagio. We still cannot claim the post caused a later business outcome simply because a model assigned it credit.

This integration grew from the product question described in Why We Built Presagio: reduce the distance between a business signal and the evidence behind it.

What remains an ambition

The ambition is broader than reporting a cleaner funnel. We want a founder to begin with a business outcome, see the customer journey behind the movement, and move into the people, sessions, campaign touches, and product evidence that can sharpen the next decision. That complete founder view is the direction; every part of it still has to earn trust through explicit definitions, privacy boundaries, and verifiable product behaviour.

We are not claiming automated certainty, perfect identity, causal answers, or a generally available commercial platform. Some capabilities have shipped inside the controlled beta, some journeys are configured specifically for TDG, and some of the larger operating vision remains future work. Keeping those three categories separate is not cautious marketing. It is how the product stays accountable to the evidence.

A practical way to start

• Choose one decision, not every available metric.
• Name two to five observable moments in plain business language.
• Write the rule behind each moment and keep its version visible.
• Set a completion window that matches the real behaviour.
• Read completed, pending, and elapsed people separately.
• Inspect a bounded sample of evidence before proposing a change.
• Use research or an experiment when the question is why, not merely what.

The point is not to turn every visit into a lead or every journey into a certainty. It is to make the movement after arrival visible enough that the next product question becomes smaller, sharper, and more honest.