Consent is the first receptor your marketing signal passes through. Everything downstream inherits its calibration.

Your CAPI events, attribution models, audience syncs, and bid optimization are all conditioned on what happens at one gating layer most teams never measure. When that receptor is miscalibrated, every measurement the organism takes is taken through a distorted lens; and the system has no way to know.

Diagnose your receptor Read the membrane frame
What you're seeing in the data.

These are the surface-level patterns that bring teams to Signal Quality. Each one is downstream of something further up: usually upstream of the analytics conversation entirely.

Modeled conversions are climbing while observed conversions stay flat.
Consent Mode v2 is doing its job: translating denied-consent into a statistical approximation. But if your primary receptor is miscalibrated, the model is dutifully transducing distorted input into modeled output your bidding layer treats as ground truth.
CPA is drifting up. Your agency says campaigns look fine.
From their dashboard, they're not wrong. The dashboard inherits whatever the receptor let through. Performance Max and Advantage+ are optimizing against the signal shape your membrane permitted, not the one you designed.
Every platform claims credit for the same conversions, and none of them match your CRM.
Each platform is grading its own homework against a different fragment of the same underlying signal. Identity fragmentation, deduplication failure, and parameter decay all compound whatever happened at the receptor.
Your "Direct" traffic is suspiciously high.
When 30–40% of traffic shows up with no source, it doesn't mean people are typing your URL. It means attribution context is being stripped between click and the first downstream receiver: often at the consent layer, before the rest of the stack ever sees it.
Your CMP is deployed. You're still showing up in California enforcement patterns.
Todd Snyder, Disney, Ford, Tractor Supply, Healthline. None of those fines were for lacking consent infrastructure. They were for having infrastructure that looked correct and didn't function. Same diagnostic terrain.
A cell membrane is not a wall. It's a selective interface.

Things pass through it, but only through specific, regulated mechanisms: channels, pumps, receptors. The membrane's job is not to keep everything out. It's to determine what gets in, under what conditions, and in what form.

Consent is the first gating receptor at your marketing membrane. It determines whether any signal is permitted to cross the boundary at all, and which properties of that signal (identity, context, value) travel with it. Everything downstream of that receptor is conditional on what happened at it.

The marketing data field has built sophisticated tools for the nucleus: warehouses, semantic layers, attribution models, BI. It has built almost nothing for the membrane. That gap is where the silent corruption lives.

Every visit resolves to one of four states.

The first three are designed-for. The fourth is invisible to the business, and it's the one regulators are fining and ML systems are silently training on.

STATE 01
Fully permeable
~100% fidelity
STATE 02
Modeled
~45% fidelity
STATE 03
Blocked
0% · dark matter
STATE 04 · FAULT
Miscalibrated
unknown · distorted

State 4 is the dangerous one. Banner present. CMP deployed. The business believes signal is flowing correctly. But pre-consent firing, server-side updates that never propagate, GPC captured but not applied, vendor contracts silent on data sharing: each produces signal indistinguishable from healthy signal until something else breaks.

Read the full membrane frame →

Audit the receptor. Calibrate the membrane. Monitor the boundary.
01 · Audit
A Signal Fracture Audit diagnoses receptor calibration first using only public browser data and a single console capture, no backend access required. It traces inheritance through seven downstream fracture classes: identity fragmentation, deduplication failure, parameter decay, CAPI integrity, sGTM routing, pixel collision, GA4 contamination. You get a clear picture of what your bidding algorithms are actually training on, what it's costing you, and exactly what to address first.
02 · Build
Targeted remediation of the receptor and the transduction layer: consent infrastructure that actually gates, server-side container updates that propagate, GPC signals that reach vendors, deduplication keys that match, hashing that produces match-quality scores above floor. No platform migrations. No six-month overhauls. The membrane works the way the business already thinks it does.
03 · Monitor
Membranes drift. Platforms change APIs. Tags get added. A configuration that was calibrated last quarter degrades silently. Ongoing monitoring catches receptor-state changes and downstream fractures before they show up in performance reports; not after a regulator notices, not after a quarter of distorted optimization.
The Frame
Compliance is the fine-detection layer. Signal Quality is the calibration layer. One catches you when you're wrong. The other makes sure you aren't. The regulatory risk is downstream of the signal quality risk, not the other way around.
The person behind the practice.
Erich Eisenhart

I'm Erich Eisenhart. I diagnose marketing membranes: the gating, transduction, and integration layers between a user's browser and the ML systems your spend trains. The receptor calibration check, the fracture taxonomy, and the deliverable format are mine.

The same patterns keep showing up. DTC brands running Performance Max discovers their CMPs are decorative; every consent category shows update: false regardless of what the user clicks. A sophisticated B2B operator finds GPC captured at the edge but never propagated to vendors. An enterprise advertiser realizes Consent Mode v2 has been transducing a corrupted primary signal into modeled conversions for months, and the bidding layer has been training on it.

These aren't edge cases. Miscalibrated is the default state. Marketing stacks are built tool by tool, integration by integration, over years; nobody steps back to verify the membrane is doing what the organism thinks it's doing. That's the practice. I observe the receptor, document what's inheriting from its calibration, and rebuild the boundary so your downstream measurements are taken through a lens you actually designed.

Diagnose the receptor.
A Signal Fracture Audit takes a few days, requires no platform access or credentials, and produces a single self-contained deliverable: receptor state, downstream fractures, dollar math, and the exact next move. No retainer. No commitment. A clear, honest read on what your membrane is actually doing.
Book a Signal Fracture Audit
Or connect on LinkedIn