Start here
This site is 34 resources and 57 real files, which is too many to browse if you do not already know what you want. So pick the situation below that matches yours and do the four steps in order. Each step is a working file, not an article — you should finish step one with something on your screen you can use.
-
You are setting up marketing for a business from scratch
- Check the fundamentals exist
Before anything clever, confirm the boring things are true: conversions tracked once, one hostname, a form somebody receives.
- Decide the direction, on one page
Four worked examples for different business types. The field that matters is the one naming what you will not do.
- Test the target with arithmetic
Turns a customer target into the traffic it needs. This is where an ambitious goal meets reality, cheaply.
- Write the plan and the budget
Activities, owners, dates. Last, not first — it is the document that changes most.
- Check the fundamentals exist
-
You have inherited a setup somebody else built
- Score it before you change it
45 items across eight weighted sections. It ranks weaknesses by points lost, so you fix the expensive things first.
- Work out what the numbers mean
Inherited reports use metrics inconsistently. Sessions are not users and reach is not impressions.
- See who has been gaining on you
A scorecard rather than a feeling, so you can tell your predecessor’s decisions from the market’s.
- Start reporting a trend, not a month
Twelve months of rows makes a bad month readable as a wobble instead of a crisis.
- Score it before you change it
-
You are freelancing or running an agency and need clients
- Write up one piece of work properly
Constraints and what you would do differently are what make it credible. Ask permission at the start of an engagement, not the end.
- Decide your price before the call
With the margin calculated. Deciding during a call is how a project becomes unprofitable.
- Send a proposal that is mostly assembly
Faster to write and more convincing once the proof and the price already exist.
- Agree terms before starting
Including the confidentiality carve-out that lets you publish the next case study.
- Write up one piece of work properly
-
You are applying for a digital marketing job
- Rewrite the CV around results
Most marketing CVs list duties. The worked examples show the same experience written as outcomes.
- Build something to show
What to include per discipline, and what to do when you cannot share client figures.
- Learn the vocabulary you will be tested on
Interviewers ask what CAC is and how it differs from CPA. The confused-pairs section is the one to read.
- Know what tools the job expects
Grouped by the job each does, with an explicit label for how well I know each one.
- Rewrite the CV around results
What this site is not
It is not a course, and it will not teach you how Google Ads or Meta’s auction works. There is no certificate at the end. Every page here assumes you already have a task in front of you and gives you the file for it — which is genuinely useful if you are doing the work, and the wrong resource entirely if you are trying to learn the field from zero.
If you are starting from zero, the free first-party training is better than anything I could write and it is free: Google Skillshop for Ads and Analytics, and Semrush Academy for SEO fundamentals. Come back here when you need to produce something.
Five things beginners spend money on too early
- Ads before conversion tracking works. Spending against broken measurement does not produce a smaller result, it produces an unreadable one — so you cannot tell whether to continue. This is the single most expensive ordering mistake.
- A redesign instead of a positioning decision. Design is rarely why a site produces no enquiries. A homepage that does not name a specialism is.
- A fourth channel before the first one is working. Three channels is the ceiling for most teams, and two done properly beats four done thinly.
- Tools before process. A reporting tool will not decide what your primary metric is, and buying one is a comfortable way to avoid deciding.
- Content volume before clustering. Publishing weekly into a space you have not researched produces pages that compete with each other.
Or browse by what you are doing
- Planning — Plans, strategy, budgets and the arithmetic that says whether a target is reachable.
- Win work — Proposals, pricing, contracts and the proof a buyer asks for before saying yes.
- Channels — Per-channel audits and checklists — search, paid, social, email, content, conversion and AI search.
- Reporting — Audits that tell you what is true, and reports that show everyone else.
- Tools — What to buy, what to skip, and where AI genuinely helps.