When I started managing paid search at Sofyan Corp, I thought optimization was about finding the perfect bid. Seventeen months and IDR 793 million in managed spend later, I've learned that the real game is something far less glamorous — and far more important.

Across three Google Ads accounts — Arvahub, ArvaVO, and Arva Paramania — I managed full-cycle campaigns from setup to scale. The numbers tell one story: 37.9K clicks at a 9.8% average CTR, well above the 2–5% industry benchmark. But the lessons behind those numbers are what actually matter.

1. Budget discipline beats clever bidding.

Everyone obsesses over bid strategy. Manual CPC, target CPA, maximize conversions — the debate never ends. But the single biggest lever wasn't bidding. It was knowing where not to spend. On Arva Paramania, we cut wasted spend so aggressively that CPC dropped to IDR 10.7K — roughly 40% below our other accounts.

The discipline of pausing underperformers fast, reallocating to winners, and never letting a campaign run on autopilot did more than any bid adjustment ever could.

The best optimization isn't finding the perfect bid. It's having the discipline to stop funding what isn't working.

— On managing paid search at scale

2. Quality Score is a compounding asset.

Improving Quality Score isn't a one-time task — it compounds. Better ad relevance lowers CPC, which frees budget, which buys more data, which sharpens targeting further. On Arva Paramania, the combination of keyword refinement and landing page alignment created exactly this flywheel.

What actually moved Quality Score:

  • Tight ad-group themes — one intent per group, never mixing
  • Ad copy that mirrored the exact search query language
  • Landing pages aligned to the ad promise, not the homepage
  • Aggressive negative keyword lists, reviewed weekly
793M
IDR managed across 3 accounts
9.8%
Average CTR (vs 2–5% benchmark)
40%
CPC reduction on best account

3. Reporting is a thinking tool, not a chore.

I used to treat reporting as the boring part — something you do after the real work. That was backwards. Building clear dashboards in Looker Studio forced me to confront what the data was actually saying, not what I hoped it said. The act of explaining performance to stakeholders sharpened my own decisions.

The lesson

If you can't explain why a campaign is winning or losing in one sentence, you don't understand it yet — and you definitely shouldn't scale it.

4. Know when to stop testing.

A/B testing feels productive. But endless testing is often just an excuse to avoid commitment. Once a variant proved itself with statistical confidence, the right move was to commit budget and move on — not to keep tweaking for marginal gains that never materialized.

The discipline of declaring a winner and scaling it is underrated. Most accounts don't fail from bad tests. They stall from never acting on good ones.

Dashboard screenshot placeholder
Arva Paramania account — the most efficient in the portfolio at IDR 10.7K CPC.

5. The account is never "done."

Seventeen months in, the accounts still needed weekly attention. Search terms shift, competitors enter, seasonality bites. The marketers who win paid search aren't the ones with the cleverest initial setup — they're the ones who show up consistently, week after week, making small corrections before they become big problems.

If I had to compress 17 months into one line: paid search rewards discipline over cleverness. The budget control, the consistent reporting, the willingness to stop what isn't working — these unglamorous habits are what turned IDR 793M into measurable, repeatable growth.