Why this feels harder than it is
The Google Ads interface asks for things in the order that suits Google, not in the order that makes sense to someone building their first campaign. You get asked for a bidding strategy before you have written a single ad. You are offered settings whose consequences only become visible three weeks later.
So people fill it in top to bottom, accept the defaults on anything unfamiliar, and end up with a campaign that technically runs. It runs badly for reasons that are invisible from inside the interface, which is how 36% of accounts end up below a Quality Score of 4.
The structure itself is not complicated. There are four levels and each one answers a different question.
The account level: do this before anything else
Billing, and conversion tracking. Conversion tracking is the one to be rigid about. An account without it can tell you how many clicks you bought but not whether any of them were worth buying, and every decision after that is a guess with a chart attached.
Set it up before you spend, not after the first disappointing month, because it cannot backfill.
The campaign level: four decisions
Type. Google lists Search, Display, Shopping, Video, App and Performance Max. Search shows text ads to people actively searching for what you sell, which is why it is almost always the right first campaign. Display shows image ads to people browsing, a different job with a different expectation. Shopping needs a product feed.
Budget. Daily, and Google may spend somewhat more on a busy day and less on a quiet one. Pick a number you can lose for a month, because the first month buys information rather than customers.
Locations. Where you actually trade. This one is cheap to set and expensive to get wrong, because nothing about the interface tells you that you have been advertising in a city you do not serve.
Bidding. Without conversion data, automated strategies have nothing to optimize toward. Starting with manual or a maximise-clicks approach and switching once you have conversions is the honest path, though plenty of accounts run automated from day one and survive.
The ad group level: the one that decides everything
This is where accounts are made or ruined, and it comes down to one rule.
One ad group covers one thing. Tight enough that a single ad can honestly mention what everyone in that group searched for. If you cannot write one headline that fits every keyword in the group, the group is too broad.
A plumber does not have a "plumbing" ad group. They have "boiler repair", "bathroom installation", "emergency callout". Each gets ads that say those words and a landing page that delivers them.
The reason is not tidiness. Google builds Quality Score from expected click-through rate, how closely the ad matches the intent behind the search, and how useful the landing page is. A loose ad group fails all three simultaneously, and Quality Score is what you are priced on.
Keywords and match types, without the hours
Google names three match types. Exact match shows on searches with the same meaning or intent as your keyword and gives you the most steering over who sees the ad. Phrase match shows on searches that include the meaning of your keyword. Broad match, in Google's own words, may show on searches that do not contain the direct meaning of your keywords.
The usual advice is to start with a huge keyword list. I think that is backwards for a first campaign. A short list on tighter match types tells you something clean within two weeks. A long list on broad match tells you that you spent your budget, and not much else.
Where the keywords come from is the part that eats the hours, and it is also the part most people overthink. The words are already on your website. You wrote them when you described what you sell. The job is extracting them, grouping them by theme, and deciding a match type for each, which is mechanical rather than creative.
The ads, and the words that must repeat
Use the ad group's words in the headlines. If the group is "boiler repair", the headline says boiler repair. This looks almost too obvious to state, and it is the single most commonly skipped thing, because writing fifteen headlines is tedious and generic ones are faster.
Then negative keywords, which belong here and are covered properly in setting up without wasting budget on irrelevant clicks. A quarter of accounts have none at all.
The fast path, honestly described
Everything above is a few days of work the first time, and most of it is mechanical: extracting themes from your own pages, grouping keywords, deciding match types, writing fifteen headline variations that all say roughly the same thing.
AdPilot does that pass. You give it your website URL, it reads the site and researches the market around it, and it produces the whole structure: ad groups themed properly, keywords with match types, negative keywords, headlines and descriptions, budgets, locations and extensions. Every field is editable, and each choice has its reasoning next to it, so approving a setting does not require already understanding it.
Campaigns push into your Google Ads account paused. You read the whole thing and switch it on yourself, which is the only sensible order when real money is on the other side.
- ✓Ad groups themed from your own pages rather than guessed
- ✓Match types chosen per keyword, not left on the default
- ✓Negative keywords included before launch
- ✓Every setting explained in plain language where you make it
- ✓Campaigns arrive paused so nothing spends unreviewed
Questions people actually ask
How many ad groups should a campaign have?
How do I do keyword research without spending hours?
Which match type should I start with?
Do I really need conversion tracking before launching?
Should my first campaign be Search or Performance Max?
Citations (5)showhide
- 1How Google Ads accounts are structuredGoogle Ads Help
- 2About keyword matching optionsGoogle Ads Help
- 3Choose the right campaign typeGoogle Ads Help
- 4About Quality Score for Search campaignsGoogle Ads Help
- 5Our Biggest Google Ads Performance Study Yet: What We FoundWordStream
