The growth hacking agencythat diagnoses, ships, kills losers, compounds, retainsyour funnel, not your vanity metrics.
A growth hacking agency that runs the AARRR funnel diagnostic before touching a channel, ships 2 to 4 experiments per week priorized by ICE, weighs retention equal to acquisition, kills vanity metrics at week 1. The cleanest growth math we know.
ActiveCampaign
Adalo
AdCreative.ai
Ahref
Airtable
Allo (The Mobile First Company)
Anthropic
Apify
Apollo.io
Attio
Base44
Baserow
Brevo
Bright Data
Browse AI
Bubble
CaptainData
ChatGPT
Claude
Claude Code
Claude Cowork
Clay
Clickup
Cursor
Deepseek
Dust
ElevenLabs
Fillout
Flutterflow
Folk CRM
Freepik Spaces
Gamma
Gemini
Glide
Grok
HiggsfieldGrowth that actually compounds stands on 4 pillars.
Most growth retainers run paid ads, declare victory on impressions, and miss the leak entirely. The stack we deploy in 2026 diagnoses the funnel first, prioritizes experiments by ICE, weights retention equal to acquisition, and reviews monthly so the team learns where the math actually moves.
- Diagnostic
AARRR funnel audit
We pull your numbers on Acquisition, Activation, Retention, Revenue and Referral, identify the stage that's leaking the most, and stop you from spending another euro on the wrong layer. Most teams that ask for 'more leads' actually need retention work first — the math says so before the strategy does.
How we diagnose - Experiments
Weekly backlog, priorized ICE
Every growth idea gets scored on Impact, Confidence, Ease — the ICE framework. We ship 2 to 4 experiments per week, kill the losers fast, double down on the winners. The team works on the next test before the last one's results are even cold. Pace beats perfection.
See the experiment backlog - Acquisition stack
Paid, cold, content, viral loops
Channel mix per ICP and per CAC ceiling. Paid Meta + LinkedIn Ads when ROAS proves out. Cold email + LinkedIn outreach with multi-touch sequences. Content engine wired into SEO + AI SEO. Viral loops built into the product when the math allows. We benchmark CAC by channel monthly.
See the acquisition stack - Retention + revenue
LTV, churn, expansion
Acquisition without retention is a leaky bucket. We engineer onboarding sequences that hit activation faster, automated nurturing that surfaces upsell triggers, churn-prevention workflows that catch at-risk accounts before they cancel. Every CAC dollar amortized over more months. The cleanest growth math we know.
See retention work
What a growth retainer in motion actually delivers.
- −50%CAC after the AARRR diagnostic
Most teams come in spending on channels their funnel can't convert. Stop the leak first, then double down on what works. Directional benchmark, your starting CAC sets the amplitude.
- 2-4Experiments shipped per week
On a retainer mission, we run 2 to 4 experiments weekly priorized by ICE. The team works on the next test before the previous one is statistically conclusive. Velocity is the moat.
- +30%LTV from retention work
Most growth pitches forget retention. We don't. Onboarding sequences hitting activation faster, automated nurturing surfacing upsell, churn-prevention workflows. Cohort LTV typically lifts 20-40% on B2B SaaS missions.
Our 4-step build, from diagnostic to compounding.
Same shape regardless of starting baseline. We diagnose the leak, prioritize experiments by ICE, ship weekly, iterate monthly. The team learns the operating mode within the first month, owns the dashboard from week 2.
- Diagnose · AARRR funnel audit, identify the leak, stop spending on the wrong stage
- Prioritize · experiments scored on ICE, top 10 picked, sequenced into a 12-week plan
- Ship · 2-4 experiments per week, results read in the dashboard the day they conclude
- Iterate · winners doubled down, losers killed fast, backlog refreshed every 4 weeks
We measure revenue, not impressions.
Most growth retainers report on impressions, followers, social engagement — metrics that look great in the deck but don't pay rent. We kill them at week 1. Every experiment ships with a revenue-driving primary metric and a kill criterion. Velocity beats polish: 4 experiments per month at 25% win rate beats 1 experiment per month at 80%.
- ICE scoring on every experiment — no idea ships without a confidence rating
- Vanity metrics killed at week 1 — impressions and follower counts don't pay rent
- Weekly experiment cadence with results read in a shared dashboard
- Retention work weighted equal to acquisition — leaky buckets get patched first
We run the AARRR diagnostic, you leave with the leak.
Before quoting anything, we spend 60 minutes pulling your numbers across Acquisition, Activation, Retention, Revenue, Referral. We identify the single stage that's leaking the most, score the top 3 experiments to fix it, give you the rough cost and expected lift. Zero pitch, just a diagnostic.
- AARRR snapshot across the five stages
- The single stage leaking the hardest (with the math behind it)
- Top 3 experiments to ship within 30 days, scored on ICE
- Honest take on whether you need acquisition work or retention work first
How we run a growth engagement.
Five steps, in order, no skip. We don't touch a channel before the AARRR diagnostic is signed, we don't ship an experiment without an ICE score and a kill criterion, and we don't bill a retainer before the first 2 experiments are live in production. Every step has a DOD and you approve before we move to the next.
- Step 1 · AARRR diagnostic
Run the AARRR funnel diagnostic before touching a channel
We pull your numbers across the five stages: Acquisition (visitors / leads per channel), Activation (% who hit the first value moment), Retention (cohort curves at 1 / 3 / 6 / 12 months), Revenue (ARPU, LTV, expansion rate), Referral (% of new customers from existing). Then we identify the single stage that's leaking hardest. Most teams asking for 'more leads' actually have an activation or retention problem — the math says so. You walk away with the leak diagnosed and the top 3 quick wins to fix it within 30 days.
- Step 2 · ICE prioritization
Score every experiment on Impact, Confidence, Ease
Growth ideas pile up faster than any team can ship. We score each one on three axes: Impact (how much revenue it moves if it works), Confidence (how sure we are it'll work, based on prior data or industry benchmark), Ease (how fast we can ship it, in days). Sort by ICE total, pick the top 10, sequence into a 12-week plan. The framework forces the team to argue with numbers, not opinions — which is the only way the backlog actually moves.
- Step 3 · Ship + measure
Ship 2 to 4 experiments per week, read results daily
Every experiment ships with a hypothesis written upfront, a primary metric, a target lift, and a kill criterion. We use feature flags or A/B testing tools (Mixpanel, Amplitude, Statsig, GrowthBook) to expose the variant to a slice of traffic, watch the metric, call the result. Winners get rolled out to 100%. Losers get killed within a week — no sunk-cost ride-along. Pace beats perfection: 4 experiments per month at 25% win rate beats 1 experiment per month at 80%.
- Step 4 · Retention engineering
Engineer retention in parallel with acquisition
Most growth retainers focus on acquisition because it's the loudest stage. We weight retention equal. Onboarding sequences engineered to hit the first value moment within the first session. Automated nurturing that surfaces upsell triggers from product usage signals. Churn-prevention workflows that detect at-risk cohorts and fire intervention emails or in-app prompts. Cohort LTV tracked monthly. Every CAC dollar amortized over more months — the cleanest growth math we know.
- Step 5 · Monthly review + backlog refresh
Review the funnel monthly, refresh the experiment backlog
Monthly review: AARRR snapshot vs baseline, experiment scoreboard (shipped / killed / inconclusive), CAC and LTV by channel, leakage shift since last month. We decide what to extend (winners), what to retire (lost relevance), what to migrate (the channel that scaled out of its window). Backlog refreshed with new ideas scored on ICE. Action items written, owners assigned, shipped before the next review. Compounding from month 1.
The same engine, across multiple client funnels.
The frames below come from real monthly funnel reviews with clients running growth retainers: AARRR snapshot vs baseline, experiment scoreboard (shipped / killed / inconclusive), CAC and LTV by channel, leakage shift since last month. Same operational rigor, different industries, all in B2B SaaS, services and ops. Our Trustpilot reviews come from the operators we work with.
- Monthly funnel review with every client running a growth retainer
- Experiment scoreboard updated weekly, no quarterly slide deck
- A losing experiment gets killed within the week, no sunk-cost ride-along
- Trustpilot reviews come from the marketing and CEO teams running the retainer
The 10 questions we get asked on every call.
What's growth hacking and how is it different from marketing?
Growth hacking is data-driven experimentation across the full funnel — Acquisition, Activation, Retention, Revenue, Referral — with the explicit goal of finding scalable, repeatable, low-cost growth levers. Traditional marketing focuses on top-of-funnel (acquisition + brand). Growth hacking weights the five stages equally and uses experimentation instead of campaigns. A growth hacker ships 4 experiments a month and kills 3; a traditional marketer ships 1 campaign a month and defends it. Different operating mode entirely.How much does a growth hacking agency cost in 2026?
Depends on scope. A focused mission (AARRR diagnostic + 12-week experiment plan + execution on top 5 experiments) runs $12,000 to $30,000. A monthly retainer covering 2-4 experiments per week, retention engineering and monthly funnel reviews starts around $5,000-$10,000/month for SMBs, up to $20,000+ for mid-market with engineering teams to coordinate with. Watch out for agencies that charge by channel (Facebook ads $X, LinkedIn ads $Y) — that's media buying, not growth hacking.What's the AARRR framework?
AARRR (Pirate Metrics, Dave McClure 2007) is the five-stage customer lifecycle: Acquisition (how do they hear about you), Activation (do they get value on first touch), Retention (do they come back), Revenue (do they pay), Referral (do they bring others). The framework forces you to measure all five and identify the leakiest stage — which is rarely the one your team thinks it is. Most teams asking for 'more leads' have an activation or retention problem and don't know it.What's ICE scoring?
ICE scores every growth experiment on three axes: Impact (how much revenue it moves if it works), Confidence (how sure we are it'll work, based on prior data or industry benchmarks), Ease (how fast we can ship it, in days). Score each axis 1-10, multiply, sort by total. The framework forces the team to argue with numbers, not opinions. The top-scoring experiments ship first. Variants like RICE (adds Reach) work fine too — the point is having a framework that prevents the loudest voice in the room from setting the backlog.How long until we see results from growth hacking?
Honest: 4 to 8 weeks for the first measurable wins (a couple of experiments land, CAC moves on one channel, activation rate lifts on the onboarding tweak). 3 to 6 months for the compounding to show up in cohort LTV and blended CAC. Growth hacking isn't a campaign; it's a system that ships experiments weekly. The first weeks are diagnostic + setup, weeks 4-12 are when the experiments start landing and the dashboard starts moving.Will you do cold email and LinkedIn outreach?
Yes, when the AARRR diagnostic says outbound is the right lever. We deploy multi-touch sequences (Lemlist, Smartlead, Instantly, Reply.io, LaGrowthMachine) with AI-personalized first lines, multi-channel cadences (email + LinkedIn + voicemail drop), and deliverability hygiene (warm-up, SPF / DKIM / DMARC, inbox rotation). But we only deploy outbound when the math supports it — if your activation rate is 5%, more cold leads will not save you, retention work will.Will you run paid ads?
Yes, when CAC math supports it. We deploy on Meta, LinkedIn, Google Ads, and TikTok when the ICP and the offer fit the platform. We benchmark CAC per channel monthly and pull spend on losers fast. We don't run paid for branding — every euro spent has a directly measurable conversion target. If your blended CAC is already above LTV, paid is not the answer, retention or pricing is.What tools do you use for growth hacking?
Tool-agnostic, but the typical stack: analytics (Mixpanel, Amplitude, GA4), experimentation (Statsig, GrowthBook, Optimizely), CRM (HubSpot, Pipedrive, Salesforce, Attio, Folk), outbound (Lemlist, Smartlead, Apollo, Phantombuster, Instantly), automation (Make, n8n, Zapier), content + SEO (Semrush, Ahrefs, Otterly), retention (Customer.io, ActiveCampaign, Iterable). We pick per use case, never per fashion.How do you avoid vanity metrics?
Three rules. (1) Every dashboard tracks revenue-impact metrics first (CAC, LTV, conversion rate, cohort retention) and traffic metrics second. (2) Every experiment has a primary metric that's a revenue driver, not an impression. (3) Monthly reviews only debate metrics tied to the AARRR funnel. Impressions, follower counts, social engagement are tracked for diagnostic purposes (they shift before revenue) but never as primary KPIs. The team learns the discipline within 4-6 weeks.How long do we commit for?
Three formats. (1) Audit only: AARRR diagnostic + 12-week experiment plan, flat fee, 2 weeks. (2) Sprint: 12 weeks, we ship the top 10-12 experiments and the retention setup. (3) Ongoing retainer: 6-month minimum (growth compounds; the first 2 months are setup), monthly cancellation after. No forced annual contract, no convoluted exit clauses. If we don't ship and the funnel doesn't move, you stop.
Stop spending on the wrong stage. Find the leak.
A 60-minute AARRR diagnostic, the single leak identified, the top 3 experiments to fix it scored on ICE. If your funnel is already running well, we'll say so and we won't sell you anything.