A marketing glossary, including the terms nobody agrees on

Search engine optimisation. Everything done to earn a listing in search results rather than buy one. Usually split into on-page (the content and structure of a page), technical (whether search engines can crawl and index it), off-page (mostly links and mentions) and local (Google Business Profile and map results). For a Dubai service business, local is frequently the highest-return of the four.

Pay per click. The billing model where you pay only when somebody clicks, and now shorthand for paid search generally. Often used interchangeably with SEM, which is why both terms need pinning down in a proposal.

Conversion rate optimisation. Earning more from the traffic you already have, by finding where people leave and testing changes properly. The discipline it is most often confused with is redesign, which changes fifty things at once and teaches you nothing.

Generative engine optimisation. Work aimed at being cited as a source when an AI system composes an answer. About two years old as a named practice, so treat any claim of a proven methodology carefully.

Answer engine optimisation. Winning the direct answer itself — a featured snippet, a voice reply, an assistant's response. Overlaps GEO and is not the same thing, and Google returns entirely different results for the two terms.

Account-based marketing. Choosing the specific companies you want as customers and aiming everything at them, rather than generating leads broadly and filtering. Genuine ABM changes who you target, what you make and how you measure. If none of those changed, it is a campaign with a company name in the subject line.

Product-led growth. Letting people use the product before buying, so usage rather than a salesperson drives expansion. Works for products someone can get value from alone, and badly for anything needing implementation before it does anything.

Lifetime value is what a customer is worth over the whole relationship. Payback period is how long it takes to earn back what you spent acquiring them. Payback is usually the more useful of the two, because it is a cash question and LTV involves predicting the future.

Impressions count times an ad was shown. Reach counts distinct people. Frequency is impressions divided by reach. On a market the size of the UAE, frequency climbs faster than most advertisers expect and it is the column nobody adds.

Click-through rate. Clicks divided by impressions. It describes the ad rather than the business, and optimising for it alone reliably produces headlines that mislead.

Google's measures of real-world page experience: how fast the main content appears, how quickly the page responds, and how much it shifts while loading. Field data from real visitors is what counts, not a score from a developer's laptop.

A signal telling search engines which version of a page is the original when several URLs show similar content. Quietly one of the most damaging things to get wrong on an e-commerce site.

Code that describes what a page contains in a machine-readable way. It does not improve rankings directly. It makes a page eligible for rich results and easier for AI systems to extract, both of which change how often people click.

Tags telling search engines which language and region version of a page to show. Easy to get subtly wrong and hard to notice — the symptom is the wrong language ranking in the wrong country.

Advertising to people who have already interacted with you. Used interchangeably in practice. Worth capping by frequency and excluding people who have already bought, which is skipped often enough to be worth stating.

An audience a platform builds by finding people similar to a list you supply. The quality depends entirely on the quality of the source list, so a lookalike built from all past leads performs worse than one built from actual customers.

Server-side conversion tracking sent directly from your systems to an ad platform alongside the browser pixel. Recovers a meaningful share of signal lost to browser restrictions, and it is usually the highest-return technical work on an inherited advertising account.

The period after a campaign change while a platform's model gathers enough conversions to perform predictably. Editing a campaign restarts it, which is why accounts that get adjusted twice a week never stabilise.

In chatbots, the share of conversations resolved without a human. Worth treating carefully as a target, because a high rate achieved by refusing to escalate is a worse outcome than a low one.

Time from an enquiry arriving to a genuine first contact attempt. One of the strongest predictors of conversion there is, almost never measured, and almost never anybody's responsibility.