The Signal Stack. What Does Your Organization Actually Know?
Framework 02 of the Design Intelligence system.
Most organizations are not suffering from a lack of data. They are suffering from an inability to distinguish between what matters and what is merely loud. The Signal Stack is the discipline that closes that gap — and in African markets, where formal data is thin and assumptions travel fast, it is not optional precision. It is foundational.
Where We Left Off
In the last issue, we introduced the Friction Map — the first framework in the Design Intelligence system. The Friction Map is the discipline of finding where resistance is structurally embedded in a product, organization, or user journey before that resistance becomes a retention problem or adoption failure. It begins with the user's world, not the product, and produces a ranked picture of where intervention will have the greatest return.
But the Friction Map surfaces something that creates its own problem. Once you know where friction lives — once you can see the entry barriers, the trust gaps, the infrastructure breaks — you are faced with a harder question: what do you actually know about why this friction exists, and what evidence do you have that your proposed intervention will address it?
This is where most organizations stall. They have gathered data. They have conducted research. They have feedback from users, metrics from their analytics stack, observations from their team. The problem is not the volume of information. The problem is that they cannot tell which of it is decision-worthy and which of it is noise dressed as insight.
Acting on the wrong signal at this stage does not just waste resources. It authorizes a direction. And in a compressed delivery environment — where a wrong decision can be realized in weeks — the cost of misreading your evidence is not discovered slowly. It arrives all at once.
The problem is not that organizations lack information. The problem is that they cannot tell what their information is actually telling them.
What the Signal Stack Is
The Signal Stack is the second framework in the Design Intelligence system. It is a five-level hierarchy that classifies information by its decision-making value — from noise at the base to threshold at the apex. Its purpose is not to help organizations gather more data. It is to help them read the data they already have with enough precision to know what it should and should not authorize.
The discipline behind the Signal Stack rests on a single observation: most organizations respond to information based on its volume and urgency rather than its decision-making value. A complaint that arrives five hundred times in a week is treated as a signal. A behavioral pattern observed in three user sessions is treated as anecdote. But volume and urgency are poor proxies for relevance.
You do not need a better dashboard. You need a better signal vocabulary.
The Five Levels
The Signal Stack organizes information into five levels. The levels are not a ranking of importance. They are a ranking of decision-making value.
01. Noise
High volume. Low decision-making value.
Noise is the information that surrounds every organization — social media mentions, support ticket volume, casual user feedback, vanity metrics, anecdotal observations. It is not worthless. It can point toward patterns worth investigating. But it should never authorize a decision on its own. The most common mistake organizations make is treating noise as signal — responding to what is loud rather than what is directional. In African markets, this error is amplified by the fact that the loudest feedback often comes from the most digitally active users, who are frequently not representative of the broader market the product is trying to serve.
02. Pattern
Recurring behavior suggesting something systemic.
A pattern is noise that repeats with enough consistency to suggest there is a structural cause. Drop-off at the same point in the onboarding flow, across multiple user cohorts, over multiple weeks, is a pattern. Complaints about the same feature from users in different geographies is a pattern. Patterns are worth investigating. They are not yet worth acting on. The error at this level is treating pattern recognition as diagnosis — seeing that something is happening without understanding why.
03. Tension
Where user behavior and organizational assumption are misaligned.
A tension is the most important level in the Signal Stack, and the most frequently overlooked. It is the point where what you observe users doing diverges from what you assumed they would do — or need. Tensions are where the Friction Map and the Signal Stack intersect most directly. The Friction Map surfaces where resistance is embedded. The Signal Stack identifies when that resistance contradicts the organizational model of how the product works. Tensions require a response that goes deeper than a design change. They require a revision of the underlying model.
04. Signal
A validated, directional insight that warrants a decision.
A signal is a tension that has been examined, tested, and verified with enough rigor to justify a concrete response. It is not a feeling or an observation. It is a finding — something that has been stress-tested against alternative explanations, confirmed across multiple sources, and assessed for its implications. Signals are rare. Most organizations believe they have more signals than they do.
05. Threshold
The moment a signal demands a structural response.
A threshold is a signal of sufficient magnitude or urgency that it requires an organizational response beyond a product adjustment. A trust architecture failure that is causing exit across an entire user segment is a threshold. An infrastructure condition that makes the product's core value proposition inaccessible to the majority of the intended market is a threshold. Thresholds are not common — but when they are present and unrecognized, the cost of ignoring them compounds with every iteration that proceeds without addressing them.
The Signal Stack in Practice
In every Design Intelligence engagement, the Signal Stack is applied after the Friction Map. The Friction Map tells us where the system is breaking. The Signal Stack tells us what we actually know about why — and what class of evidence would justify the intervention the Friction Map points toward.
In practice, this means examining the organization's existing research, data, and feedback through the lens of the five levels. What the team believes are signals are almost always patterns. What they believe are patterns are often noise. And the tensions — the places where user behavior contradicts the organizational model — are almost always present and almost always unrecognized.
A pattern from the field: A fintech organization was treating high withdrawal complaints as noise — volume too low, users too vocal. The Signal Stack examination revealed they were behavioral tensions pointing to a trust architecture failure at the point of withdrawal. Users were not complaining about the experience. They were expressing fear about whether their money was safe. The intervention required was not a UX improvement. It was a structural change to how the product communicated security at the moment of highest anxiety. The product team had the data. They did not have the vocabulary to read it correctly.
Why This Matters in African Markets
The Signal Stack is a universal discipline. But its importance is amplified in African markets for a specific reason: the data layer is thinner, the behavioral research is sparser, and the conditions for formal validation are harder to establish. This means that the gap between what organizations believe they know and what they actually know is wider — and the cost of acting on that gap is higher.
In well-documented markets, a wrong signal can be corrected relatively quickly. The behavioral data is abundant enough that contradicting evidence surfaces fast. In African markets, a wrong signal can run uncorrected for months — because the data infrastructure that would reveal the error does not exist, the user feedback mechanisms are not calibrated to the right questions, and the team's confidence in its own reading of the market is not subject to the checks that a richer data environment would provide.
In markets where data is sparse, the discipline of knowing what you actually know is not a research luxury. It is a survival condition.
The Design Intelligence System So Far
The Design Intelligence system is built around three frameworks that work in sequence. Each one answers the question the previous one makes possible to ask.
The Friction Map asks: where is the system actually breaking? It produces a ranked picture of structural resistance. The Signal Stack asks: what do we actually know about why? It takes the picture the Friction Map produces and examines the evidence underneath it. Next week, the Build Threshold asks the final question: are the conditions for success genuinely in place?
The Design Intelligence sequence:
Issue 005 — Friction Map: Where is the system actually breaking?
Issue 006 — Signal Stack: What do we actually know about why?
Issue 007 — Build Threshold: Are the conditions for success genuinely in place?
Each framework answers the question the previous one makes possible to ask. Applied in sequence, they move an organization from assumption to clarity — and from clarity to decisions that are chosen rather than inherited.
The Dispatch
The most dangerous moment in a product organization is not when the data is absent. It is when the data is present but misread — when a pattern is treated as a signal, when noise is treated as direction, when a tension that reveals a fundamental model failure is dismissed as a design problem.
This is not a failure of intelligence. It is a failure of vocabulary. Organizations cannot read what they do not have language for. The Signal Stack provides that language.
Before your next decision, examine the evidence underneath it. Is it noise? A pattern? A tension you have not yet named? A genuine signal? The answer to that question determines whether what you are about to build is grounded in reality or built on assumption.
Want us to run a Signal Stack examination on your organization's current evidence base? Reach out directly to Rey Mungai or Suluhu Studio to start the conversation.
By Rey Mungai
Design Intelligence Dispatch · Suluhu Studio · Issue 006