Structure controls relevance. Messy accounts waste budget on wrong queries and unlearnable reports.
Hierarchy
- Account — billing, access, shared libraries
- Campaign — goal, budget, locations, language, network settings
- Ad group — tight theme of keywords + matching ads
- Keywords / assets — queries and extensions
- Ads — RSA headlines/descriptions pointing to one clear landing page
Rule of thumb: One ad group ≈ one intent cluster (e.g. “dental implant cost” separate from “teeth whitening”).
Match types (simplified)
- Exact: tight control
- Phrase: mid
- Broad: discovery — needs strong negatives and Smart Bidding data
Targeting & budgets
- Locations: country/city/radius; exclude irrelevant regions
- Schedules: align with call-center/WhatsApp staffing
- Daily budget: cap burn; watch “limited by budget”
- Bids: manual or automated (tCPA/tROAS) once you have conversion volume
Case study: Account build for a Biratnagar law firm
| Campaign | Ad groups | Notes |
|---|---|---|
| Search – Family law | Divorce consult, Child custody | Geo = Province 1 cities |
| Search – Corporate | Company registration, Trademark | Higher bids; B2B hours |
| Brand | Firm name variants | Protect brand SERP |
Negatives shared: jobs, salary, PDF, “what is” pure research if not offering guides.
Budgeting: NPR 2,000/day split 60% non-brand / 30% brand / 10% experiments. Counselors staffed 10:00–18:00 so ads pause overnight.
Lesson: Structure is a teaching and optimization tool — you can see which intents convert, not one blended “law firm” campaign.
