Growth Marketing How-To Guides: Operator Playbooks
Step-by-step guides to the work itself — paid media audits, AI search citations, deliverability fixes, tracking setups and more, written the way operators run them.
On this page
A growing library of how-to guides written the way operators actually run the work — audits, setups, fixes and builds across paid media, AI search, email, measurement and automation, each with the concrete sequence, the numbers that define done, honest timelines, and the free tool that automates the hardest step.
Audit first: find out what's true
The diagnostic guides that reveal where the money is leaking, and what the rest of the library should fix in your case:
- How to run a paid media audit — the 40-point method: measurement, structure, creative, efficiency
- How to run A/B tests that don't lie — sample-size math and runtime discipline before the next test launches
Win the AI search shift
The citability motions, in execution order — the strategy context lives in our State of AI Search report and the SEO vs GEO comparison:
- How to get cited by ChatGPT, Perplexity & AI Overviews — the full citability playbook
- How to implement schema markup — priority types first, validation always
Fix the email machine
Deliverability is physics plus discipline; these cover both, with the Email Deliverability Checker grading your domain before and after:
- How to warm up a sending domain — the ramp schedule, week by week
- How to build a welcome flow that pays for the list — five sends, correctly ordered
Make the numbers trustworthy
- How to set up UTM tracking — taxonomy first, governance always
- How to measure content marketing ROI — beyond pageviews, with honest timeframes
These pair with the measurement cluster in the glossary — attribution, incrementality and friends — and with our UTM Builder for the taxonomy work. Building this muscle is the core of a data and analytics engagement.
Spend and scale intelligently
- How to calculate a marketing budget — three methods with worked examples
- How to scale paid ads without killing ROAS — marginal thinking and pacing discipline
- How to choose a growth marketing agency — the questions that expose weak proposals
- How to automate marketing reporting — from copy-paste to agents, in rungs
The budget and scaling guides lean on the ROAS & Break-Even Calculator and Media Mix Planner; the agency guide pairs with the pricing guides so you negotiate from market data.
How to run a guide well
Two habits make the difference. Do the checkpoint math instead of skimming past it — every guide states numbers that define "done" (complaint rate under 0.1%, test power above 80%, tags resolving on 95%+ of sessions) because finished-looking and finished are different states. And when a guide's diagnosis exceeds your team's capacity, that's information rather than failure: the comparisons collection covers the in-house-versus-agency call, and some of this work simply goes faster with an operator who has run it fifty times.
The sequence, if you're starting from zero
Teams ask for a master order more than any other question, so here is the one we run in engagements. Measurement first — tracking hygiene and UTM setup — because every later decision inherits the quality of the data underneath it. Economics second: the budget math and CAC discipline establish what you can afford before any channel spends it. Efficiency third — audit what's already running before adding anything new, since recovered waste is the cheapest growth available. Only then expansion: new channels, AI-search visibility, automation. The reversed order — expansion first, measurement someday — is the most expensive common pattern in growth marketing, because it scales spend on top of unverified numbers.
What "done" looks like, numerically
The guides share a philosophy worth making explicit: done is a number, never a feeling. Authentication is done when SPF, DKIM and DMARC all pass and complaints hold under 0.1%. A warmup is done when volume reaches target with placement holding above roughly 95% inboxed. Tracking is done when platform-reported conversions reconcile with your source of truth within a tolerance you chose deliberately. A test is done when it reaches the sample size computed before launch, and at no earlier moment regardless of how the graph looks. Writing the number down before starting is the entire discipline — it converts "we worked on deliverability" into "we moved placement from 78% to 96%," which is the difference between activity and progress, and incidentally the difference between reports that get skimmed and reports that get budgets renewed. Every guide opens with its own done-number for exactly this reason: you should know what finished looks like before the first step, and be able to prove it to anyone afterward.
When to run a guide yourself, and when not to
A guide is worth running in-house when the work is well-scoped, the failure modes are recoverable, and the frequency is low enough that building deep expertise wouldn't pay off — a one-time UTM taxonomy, a welcome flow, a single tracking audit. Reach for outside help when the opposite is true: the work is continuous and compounding (paid media management, an always-on citability program), the cost of getting it wrong is high and slow to detect (a botched domain warmup can burn sender reputation for months), or the diagnosis reveals a problem your team simply doesn't have the hours to fix well. The honest read on which situation you're in is itself a decision the comparisons collection and a short strategy call can help you make.