Hypothesis quality over test count
We will not run forty tests a month to justify a retainer. We will run the eight that change a number on your growth model.
Start with an audit
Growth marketing is just marketing, run honestly.
You have read Balfour, Verna, and Lenny. You do not need another framework. You need someone to find the two or three levers actually moving revenue, and run them.
why growth marketing
Why Growth Marketing
Most growth marketing is test-volume cosplay.
CAC is up 40–60% since 2023 (Genesys). Median CAC payback now sits at 6.8 months and is still rising. Two in three experiments fail by default, according to Microsoft's own ExP data. AI broke the loops that worked in 2022 inside eighteen months. Most agencies still ship the 2022 playbook and call it a growth model. We do not.
Get in touchYour accounts. Your data. Yours.
Ad accounts, GA4, Looker, and Mixpanel all sit in your name from day one. You can walk away whenever you like, and the assets walk with you.
North Star, not channel theatre
We report against your North Star, CAC, payback, LTV:CAC, and NRR. These are the numbers your CFO and your board already track. We move them quarterly.
our approach
Our Approach
Find the lever, then pull.
One or two channels in your stack are doing the real work. The rest is noise charged to a media plan. We find the work in 30 days, fund it, and kill the rest.
Thirty days. We pull apart your CAC by channel, your retention curve by cohort, and your funnel by AARRR stage. You leave with a shortlist of the two or three levers that actually move revenue, and the ones to stop funding.
Every test starts as a written hypothesis with a named owner, a baseline, and a kill criterion. If we cannot write the hypothesis, we do not run the test. Most agencies skip this step, which is why their results never accumulate.
Senior specialists run the experiment across paid, lifecycle, CRO, creative, and AI-augmented variant testing. You see every test, every result, and every learning in a live log. There is no once-a-month deck theatre.
The channel mix you need at Series A is not the mix you need at Series B. Every quarter we re-sequence: what to fund, what to retire, what to test next. The organisation evolves, and the growth model evolves with it.
Our Growth
Services
Experimentation, paid, lifecycle, and analytics sit wired together. Senior specialists run them, and we report against the numbers your board already tracks.
Experimentation Programmes
Hypothesis design, test calendar, sequential evaluation, and kill criteria. We run experiments that decide things, not ones that produce slide-deck filler.
Paid Acceleration
Paid social, paid search, programmatic, and AI-augmented creative testing. Each is built around CAC and payback targets, not ROAS theatre.
Lifecycle & Retention
Onboarding, activation, expansion, and win-back. These are the loops that move NRR, because the cheapest pipeline is the customer you already have.
Analytics & Attribution
Multi-touch attribution, marketing-mix modelling, and AI-visibility tracking. We stay honest about what attribution can and cannot tell you.
Fewer tests. Sharper hypotheses.
The growth-hacking era taught a generation of marketers to confuse motion with progress. We do not. We run the smallest number of experiments that can move a real number on your growth model, and we kill them honestly when they do not. The discipline does the work, not the dashboard.
why MxD
Why MxD for Growth
AI broke the old loops. We rebuilt them.
Brian Balfour rewrote the Four Fits in 2025 for a reason. AI ate long-tail SEO, cheap freemium, and most paid-social arbitrage inside a year. Agencies still selling the 2022 playbook will sell you 2022 results. We have rebuilt experimentation, channel mix, and discovery for what works now, including the answer engines your buyers ask before they ever touch Google.
Common
Questions
Isn't growth marketing just marketing with a fancier title?
Mostly, yes. Elena Verna has had to defend the discipline twice in the last year for a reason, and we are not going to pretend the accusation has no weight. What growth marketing actually is, when run honestly, is marketing with experimental rigour, financial accountability, and channel breadth wired in by default. What we do differently is name the hypothesis before the test, name the kill criterion before the spend, and report against your growth model rather than channel-by-channel impressions.
We have been burned by a previous growth agency. How are you different?
Three things are visible by day thirty. You own the ad accounts and the data, with your name on every login, every dashboard, and every pixel; this is the single most common complaint in growth-agency reviews, and we will not let it be a complaint about us. The senior specialist in your weekly is the same one for the duration of the engagement. Our first audit will recommend you cut something, not just add. If it does not, we did not audit honestly.
We are too small for test-and-learn. Will our experiments be statistically meaningless?
At sub-5 million ARR, yes, classical A/B testing rarely produces signal. So we do not pretend it does. Early-stage engagements use qualitative-first, sequential, decision-oriented tests: bigger swings, fewer of them, evaluated against direct revenue rather than statistical confidence. Statistical testing comes later, when traffic justifies it.
What does this cost?
We work in engagement tiers, not hourly rates. A focused single-lever programme sits at one tier, and an integrated cross-channel build sits higher. Pricing is published in the proposal, fixed for the engagement duration, and quoted against scope. There is no retainer that grows when nobody is watching.