Performance Marketing
The whole funnel, one model
Channel-by-channel optimisation eventually stops working, because the constraint moves somewhere else: to the offer, the landing experience, the lifecycle, or the margin.
Performance marketing treats acquisition as an economic system: what a customer costs, what they are worth over time, how quickly the cost is recovered, and which lever moves that relationship most. Channels are instruments within that system, not the system itself.
We build the model with your finance and product teams, identify the binding constraint, and run a programme of experiments against it, across paid channels, landing experiences, offers and lifecycle communications.
Problems we solve
What usually brings people to this work.
Channel silos
Each channel optimises its own metric, and nobody owns the number that decides whether the business grows profitably.
Growth that costs more than it returns
Acquisition scales, payback stretches, and cash gets tighter as revenue rises.
Experiments that never conclude
Tests run without a hypothesis, a success threshold or enough traffic to produce a readable result.
Attribution paralysis
Platform numbers disagree, analytics disagrees with finance, and decisions stall waiting for a certainty that will not arrive.
Methodology
How the engagement runs.
- 01
Model
Build a shared view of acquisition cost, conversion rates by stage, retention and contribution margin. Agree which numbers the programme is accountable for.
- 02
Diagnose
Identify the binding constraint, whether that is traffic quality, landing experience, offer, onboarding or retention, instead of optimising whatever is easiest to change.
- 03
Experiment
A prioritised backlog of tests, each with a written hypothesis, a required sample size and a decision rule agreed before launch.
- 04
Systemise
Wins are documented and rolled into the baseline; losses are documented too, so the same idea is not rediscovered in six months.
Deliverables
- Unit economics and funnel model
- Measurement and attribution approach, with its limits stated
- Prioritised experiment backlog
- Channel and campaign execution
- Landing page and offer testing
- Lifecycle and retention recommendations
- Experiment log with decisions and rationale
Benefits
Growth you can afford
Scaling decisions are made against payback and margin, not against a traffic chart.
Compounding learning
A documented experiment log turns individual campaigns into institutional knowledge.
Fewer stalled decisions
Agreeing decision rules in advance removes the argument about what the data means after the fact.
Typical objectives
- Improve payback period or contribution margin
- Scale acquisition without destroying unit economics
- Increase repeat purchase or subscription retention
- Build a repeatable experimentation programme
Questions
Frequently asked.
Paid media management optimises campaigns within a channel. Performance marketing decides whether the channel is the right place to work at all, and extends the work into landing experiences, offers and retention.
Experimentation needs volume to produce readable results. With lower traffic we focus on larger structural changes and qualitative evidence rather than incremental tests, and we say clearly when a test cannot be read reliably.
We use platform data, analytics and, where volume allows, incrementality testing, and we report the disagreements between them rather than hiding behind one number.
More answers on the FAQ page.
Considering performance marketing?
Tell us your objective, your current setup and what you have already tried. You will get a view on scope, sequence and whether this is the right place to start.