Record the baseline
Current performance by campaign, keyword and device. Without it you cannot prove the restructure helped.
PPC Strategy · Delhi NCR & remote
Bid strategies get all the attention and can only work within the structure they are given. An account organised badly cannot be rescued by a clever automated bidding setting.
Short answer
PPC campaign management is the structural work inside an advertising account: how campaigns and ad groups are organised, which bid strategy suits your conversion volume, how budget is allocated, and how negative keywords are maintained. Structure comes first because it determines what any bidding approach is able to optimise.
Four details. A real reply the same day, from me.
01 The premise
Most accounts I inherit are organised the way the business thinks about itself: one campaign per product line, or per department, or per page of the website. It feels tidy and it is the wrong axis. The platform does not care what your departments are. It allocates budget within campaigns, so the campaign boundary should follow the thing you want controlled separately.
In practice that usually means intent and margin. A campaign for people ready to buy should not share a budget with one for people still researching, because the second will consume the first’s money at a better click-through rate and a worse outcome. Similarly, a high-margin service and a low-margin one competing for the same daily budget means the platform optimises towards volume and away from profit.
The second structural decision is granularity, and both extremes fail. Too broad and the ad cannot speak to the search; too narrow and no single ad group gathers enough conversion data for automated bidding to learn anything. The right level is where keywords genuinely share a meaning and there is enough volume to measure — which is a judgement, not a rule.
02 Bidding
The main variable is how many conversions you actually record. Choosing above your data volume is the common error.
| Strategy | When it is the right choice |
|---|---|
| Manual or enhanced CPC | Low conversion volume, or a new account with no history. You keep control while data accumulates. Unfashionable and frequently correct. |
| Maximise clicks | Almost never for lead generation. Useful briefly to gather traffic data on a genuinely new campaign, then changed. |
| Maximise conversions | Once you have steady conversion volume. Good when you want as many leads as the budget allows and cost per lead is secondary. |
| Target CPA | When you know what a lead is worth and have enough history for the target to be realistic. Set too low, it simply stops serving. |
| Target ROAS | Retail with revenue values passed back correctly. Requires accurate transaction data or it optimises towards nothing. |
| Maximise conversion value | Where different conversions are worth genuinely different amounts and you can supply those values. |
03 Decisions
Each one is cheap now and expensive to unpick later.
Free structure review
Most are organised around the website menu or the product catalogue. Give me view access and I will show you where budget is being allocated against your own interests.
Four details, and a real reply today.
04 Restructuring
Rebuilding an account risks losing learning. This sequence protects it.
Current performance by campaign, keyword and device. Without it you cannot prove the restructure helped.
Proven keywords and their history move across rather than being recreated from nothing.
New structure created before the old is paused, so there is no gap in coverage.
Moved over in stages rather than all at once, so a problem affects part of the account rather than all of it.
Automated bidding relearns after significant change. Judging performance in the first fortnight will mislead you.
— PPC Strategy
These sit next to campaign management and are usually bought with it. Same person doing the work in each case.
— Questions
Straight answers, including the ones that cost me work. If yours is not here, ask it — I reply the same day.
Around what you want to control separately — usually intent and margin, with brand always separated from non-brand. Structuring by product catalogue or website menu feels tidy but hands budget allocation to the wrong logic.
Once you have steady conversion volume, generally yes — it processes signals no human can. Below that threshold it has nothing to learn from and manual bidding while data accumulates is the better choice. Choosing a strategy above your data volume is the common mistake.
Few enough that one ad can speak to all of them, which usually means a small number sharing a genuine meaning. The counter-pressure is that excessive fragmentation starves each group of the conversion data automated bidding needs.
Always. Brand searches convert cheaply because those people already know you. Mixed into the same campaign, they subsidise and disguise the performance of prospecting, and you lose the ability to see whether new-customer advertising works.
Temporarily, yes — automated bidding relearns after significant changes, so expect an unsettled fortnight. Migrating proven keywords rather than recreating them preserves history, and switching in stages limits the disruption.
Rarely, if it was right. Structure is a foundation and constant rebuilding prevents anything from accumulating learning. Ongoing work is negatives, budgets, ads and bids — structural change should be a deliberate project, not a habit.
Free consultation · No obligation
Give me view access. You will get an honest read on whether your account is organised around your commercial interests or around your website menu.
Name, number, email, what you need.