Search is the one channel where people tell you what they want

Google will happily spend a budget on searches that were never going to convert, and the default settings help it. The expensive part of most accounts is not the bid, it is the breadth: broad match reaching for anything loosely related, no negative list, and a search terms report nobody has read since launch.

So the first work is subtraction. Pull the search terms report, see what you actually showed for, and build the negative list from reality rather than from a template. In Dubai property that single exercise routinely removes a third of the spend without touching a bid, because a large share of the traffic is brokers, students and people researching a city they are not buying in.

Someone searching your company name has already decided. Someone searching your category has not. Running both in one campaign means the cheap, high-converting brand clicks flatter the whole thing, the category terms look better than they are, and you keep funding something that is not working because the blended number looks fine.

Split them and the picture becomes honest. It also stops you paying full price to appear above your own organic result, which is a cost worth measuring rather than assuming. For some businesses bidding on your own name is defensive and correct, because competitors bid on it. For others it is a tax you are paying to Google for traffic you already had.

Performance Max is given a budget and a goal and reports back very little about where the money went. That is manageable once you know which terms and audiences convert, and close to unmanageable before. Running it first is how accounts end up spending well with nothing learned.

When it does go in, brand searches are excluded. Without that exclusion it quietly absorbs the people who were already looking for you, counts their conversions as its own, and reports a cost per acquisition that cannot be true.

Click prices here are high in several sectors because you are bidding against portals and agencies with budgets you will not match. Property is the clearest example: the portals occupy the first page and outbid almost everyone on the obvious terms.

That is not always a problem to be solved. Sometimes the honest answer is that a term is not worth buying at any price, and the budget belongs on narrower searches with less competition and more intent. I would rather tell you that in week one than take a management fee for losing an auction every day.

Sending paid traffic to a homepage wastes a meaningful share of what you paid for. The page should match the ad, load fast on a phone, and ask for the minimum. Most of the gap between two accounts with the same budget is not in the bidding, it is in what happens after the click.

This is where building and marketing in the same pair of hands actually pays. The page is built to the campaign rather than handed over as a brief, and when the data says the form is the problem, the form changes that week rather than entering someone else's backlog.

Before a dirham is spent I want conversion tracking firing on a real enquiry rather than a page view, and server-side tracking through the Conversions API where the browser alone no longer reports reliably. Without that the optimisation is guesswork dressed as data.

What you should get monthly is cost per qualified enquiry, which searches produced them, and what changed. Impressions and click-through rate are diagnostics in that report, not the headline. If the account cannot produce that, fixing it comes before anything else.