The vocabulary, in plain words

Deciding which touchpoint gets credit for a conversion. Every model is wrong; the useful question is whether the same model applied consistently shows a channel getting better or worse.

Total sales and marketing spend divided by all new customers, paid and organic together. Harder to game than channel-level CAC, which is exactly why finance prefers it.

The share of sessions with no second interaction. Frequently misread — a visitor who found the answer in ten seconds and left is a success, not a failure.

An instruction telling search engines which URL is the real one when several show the same content. Wrong canonicals quietly delete pages from the index.

Unsolicited email to someone who has not asked for it. Legal in some markets, restricted in others, and reputationally expensive everywhere if done at volume without care.

Queries your competitors rank for and you do not. The single most actionable output of most SEO tools.

Google's page-experience measurements — loading, interactivity, and layout stability. A ranking factor, but a small one next to actually answering the query.

Clicks divided by impressions. Useful for comparing creative against creative; useless for comparing channels against each other.

Whether your email reaches the inbox rather than spam. Determined by sender reputation, authentication, and how often recipients engage — not by your email tool's marketing copy.

Three DNS records that prove your email is really from you. Skipping them is the most common reason a technically fine campaign lands in spam.

A free tier intended to convert. Distinct from a free trial, which expires, and from free, which does not exist as a business model for long.

Ideal customer profile. The description of who you sell to, written specifically enough that it excludes somebody.

The proportion of available auctions you actually appeared in. Low share with good returns usually means you are under-spending.

Lifetime value — total gross profit expected from a customer. Almost always estimated too optimistically in year one.

The one number a team agrees to move. Its real function is not measurement, it is stopping five people optimising five different things.

Letting the product do the acquiring and converting, with sales arriving afterwards. Requires a product someone can get value from without being sold to.

The share of a cohort still active over time. If it never flattens, you do not have a growth problem — you have a product problem.

Search engine results page. Increasingly not a list of ten links, which is why rank tracking without knowing the page layout is misleading.

Evidence other people already chose you. Specific and attributed beats glowing and anonymous every time.

Tags appended to a URL so analytics knows where a visit came from. Worthless without a naming convention everyone follows — which is what a UTM builder is really for.

Querying data providers in sequence, falling through to the next when one returns nothing. Produces match rates no single vendor can.

A query answered on the results page itself, so nobody visits your site. Rising sharply, and the reason ranking first is worth less than it was.