If you've ever spent ₹50,000 on Google Ads with nothing to show for it, the problem usually isn't the platform. It's the order in which things were set up. PPC works when every step builds on the previous one — keyword research feeds account structure, structure shapes ad copy, copy connects to landing pages, and conversions feed bidding back to the start. Skip a step and the whole thing leaks money. This guide walks through the full sequence in the exact order I follow with new clients.

The first time I ran Google Ads for a client (a small WordPress hosting company in Pune), I burned through ₹40,000 in twelve days with three conversions. The campaign was technically running. The keywords were technically relevant. The ad copy was technically fine. But every single piece was disconnected from the others.

That's the mistake nearly every beginner makes — and nearly every "PPC course" online ignores. The platform tutorials show you how to push buttons. They don't show you the order in which to push them. This guide is that order.

Step 1: Define the conversion before anything else

Don't open Google Ads. Don't research keywords. Don't write any copy. The first thing you do is decide what counts as a successful click.

For a B2B SaaS, that's usually a demo request or a free trial signup. For ecommerce, a purchase. For a local service business, a phone call or contact-form submission. For a content site, an email signup or a high-value page view.

This sounds obvious but most accounts I audit don't have a single proper conversion configured. Or they're tracking pageviews of the contact page instead of actual form submissions. Or they're counting "engaged sessions" which Google Analytics inflates by default.

Before you spend ₹1, you need: one primary conversion, tracked accurately in Google Analytics 4, imported into Google Ads. Use Google Tag Manager to fire the event. Verify in GA4 DebugView that it's actually firing on test submissions.

The link tracker on this site builds proper UTM-tagged URLs so your analytics attributes traffic correctly — without UTMs, you'll never know which ad drove the conversion.

Step 2: Keyword research. Buyer intent first

The biggest mistake new advertisers make in keyword research is targeting informational keywords with paid ads. Someone searching "how does seo work" is reading. Someone searching "freelance seo consultant in delhi" is buying. Both have the keyword "seo." Only one converts.

Use Google's Keyword Planner (inside Google Ads, free) to get volume estimates and ideas. For each keyword candidate, look at the SERP itself. If the top results are blog posts, the intent is informational — skip it. If the top results are product pages, agency landing pages, or "Ads" labels, the intent is transactional — keep it.

A typical first-time PPC keyword list should have 15-30 transactional keywords, not 300. Going broad before going profitable wastes budget on traffic that won't convert.

For research without a paid subscription, the keyword competition tool and long-tail keyword suggestion tool on this site give you a fast workable starting list.

Step 3: Account structure — one theme per ad group

Google's "Quality Score" is built around relevance. The single biggest Quality Score improvement comes from tight ad-group structure: every ad group should target keywords with the same buyer intent.

Bad structure: one ad group called "SEO Services" with 50 different keywords from "what is seo" to "hire seo consultant" to "ecommerce seo agency."

Good structure: separate ad groups for "consultant intent" (hire seo consultant, freelance seo expert), "agency intent" (best seo agency, top seo company), "service intent" (seo services for ecommerce, technical seo audit). Each gets ads written specifically for that intent.

Tight ad groups produce higher Quality Scores, which lower your cost per click. I've seen Quality Score restructures drop CPCs by 30-50% on the same keywords with no other changes.

Google Ads campaign settings being configured on a laptop with notebook beside
Tight ad-group themes are the single biggest Quality Score lever — and Quality Score directly drives cost per click.

Step 4: Ad copy, match the search intent exactly

When someone searches "freelance seo consultant in delhi," they want to see a headline that says "Freelance SEO Consultant; Delhi". Not "Boost Your Rankings With Our Award-Winning Services."

Your headline 1 should match the keyword. Your headline 2 should add a specific benefit (years of experience, guarantee, response time). Your headline 3 can include a CTA. Your description fields should expand on benefits with specifics — not abstractions.

Use Google's Responsive Search Ads and feed in 8-10 headlines and 4 descriptions. Google will mix and match to find the highest-CTR combinations. Pin your most important headline to position 1 if you need to maintain control.

Specific things to avoid in ad copy:

  • Generic adjectives ("amazing", "best", "world-class")
  • Empty CTAs ("Click here", "Learn more")
  • Mismatched promises (your ad says "free trial" but the landing page asks for credit card)

The meta tags analyzer is useful for stealing headline ideas — audit the meta titles of competitors who already rank, see what's working, and adapt.

Step 5: Landing pages, match the ad, then convert

Most accounts I audit have ads pointing to the homepage. That's the second-fastest way to waste budget after no conversion tracking.

Each ad group should have a dedicated landing page that:

  1. Repeats the keyword in the H1; instant relevance signal for both Quality Score and the visitor.
  2. States the value proposition above the fold. Specific, not generic.
  3. Has one primary CTA — not five different conversion paths.
  4. Loads fast — under 2.5 seconds Largest Contentful Paint on mobile. Test with the page speed test on this site or Google PageSpeed Insights.
  5. Includes social proof. Specific testimonials, client logos, case study numbers.
  6. Removes friction; short forms (3-5 fields max), clear privacy reassurance, no surprise next steps.

The single biggest landing-page improvement most accounts need: shorten the form. Removing 2 unnecessary fields can lift form-conversion rate 30-60% on its own.

Step 6: Bidding strategy. Manual first, then automated

Google's automated bidding (Maximise Conversions, Target CPA, Maximise Conversion Value) works well after it has data. Before it has data — usually defined as 30 conversions in 30 days — automated bidding flounders.

Start with Manual CPC and a max CPC bid based on your target CPA divided by your expected conversion rate. For example: target ₹500 CPA, 5% expected conversion rate, max CPC = ₹500 × 5% = ₹25.

Run on Manual CPC for 2-4 weeks until you have enough conversion data. Then switch to Maximise Conversions with a target CPA matching what you've actually achieved. Don't aim for a CPA dramatically lower than your current, Google will reduce volume but not magically improve conversion rate.

Step 7: Negative keywords; the biggest leak fix

Most beginners ignore negative keywords. This single step has saved me more client money than any other optimization.

Open the Search Terms report in Google Ads (Keywords → Search Terms). Look at the actual queries that triggered your ads. You'll find:

  • Irrelevant queries ("free seo tutorial youtube" if you sell paid services)
  • Wrong-intent queries ("seo jobs salary" if you're not a recruiter)
  • Competitor queries (sometimes useful, often expensive — depends on your strategy)
  • Branded queries (might want to bid on these separately)

Add the irrelevant ones as negative keywords. Use match types: exact for specific terms, phrase for query patterns, broad for entire categories. Review the Search Terms report weekly.

A typical first 30-day account adds 50-200 negative keywords. The CPA improvements compound over time as Google's algorithm learns your true qualified traffic patterns.

Step 8: Daily monitoring — the first 14 days are critical

The first two weeks of any new campaign need daily attention. After that, weekly check-ins are usually enough. Things to monitor:

  • Impressions — too few means your keywords have low search volume or your bids are too low
  • CTR, under 2% on Search means your ads or keyword targeting are off
  • CPC, sudden spikes mean increased competition or Quality Score drop
  • Conversion rate; under 2-3% means landing-page issues
  • CPA — track against target weekly
  • Search Terms report — daily for the first 14 days, then weekly

The check server status tool is useful when conversion rate drops without explanation; sometimes the issue is your landing page being slow or returning errors.

Step 9: A/B test ad copy and landing pages systematically

After 14 days you should have 3-5 ad variants per ad group running. Pause the lowest-performing 2 and replace with new variants. Always keep at least 3 variants running so Google's machine learning has options.

For landing pages, Google Optimize is gone. Replace with VWO, Optimizely or Unbounce's built-in testing.

Step 10: Scale only what's profitable

Once a campaign has been profitable for 30 days at small budget, scale it gradually, increase daily budget by 20-30% per week, not 200% overnight. Sudden budget increases trigger Google's auction adjustments and often crash conversion rates.

Scaling rules I follow:

  • Profitable at ₹500/day for 30 days → increase to ₹650/day for 14 days, then ₹850/day, then ₹1,100/day
  • If CPA increases by more than 20% during scaling, hold budget steady for 14 days
  • Add new geographies, ad formats (Performance Max, Display) only after Search is stable

Common mistakes that destroy CPA

A short list of patterns I see destroying client budgets:

  1. Broad match keywords without close watch. Broad match has improved but still triggers irrelevant queries. Use phrase or exact match for the first 30-60 days, then expand cautiously.
  2. Ignoring mobile vs desktop performance. Conversion rates differ wildly. Adjust bid modifiers per device.
  3. Setting daily budget too low. Google's pacing algorithm starves your campaign of traffic when budget is restrictive. Either set realistic budget or accept slow learning.
  4. Not using ad extensions. Sitelink, callout, structured snippet, image extensions improve CTR free. Most accounts I audit have 0-2 extensions; well-optimised accounts have 6-8.
  5. Rotating ads on "Optimise" instead of "Show evenly". When testing new ad copy, "Show evenly" gives every variant equal chance — important for clean test data.
The single highest-use thing in PPC is the gap between the ad and the landing page. Every percentage point you can lift in landing-page conversion rate roughly equals lowering your CPC by the same percentage. Spend more time on landing pages than on bidding settings.

Tools I keep open during PPC work

A realistic timeline for a new account

Here's what to expect with proper setup and a ₹500-1500 daily budget:

  • Week 1: Setup, conversion tracking validation, initial keyword research and ad-group structure.
  • Week 2: Campaign launch, daily monitoring, first round of negative keywords.
  • Week 3-4: Optimisation — pausing weak ads, adjusting bids, refining negatives.
  • Week 5-8: Switch to Maximise Conversions bidding once 30+ conversions accumulated. First tests of new ad variants.
  • Week 9-12: Profitable consistency. Begin scaling cautiously.
  • Month 4+: Add new ad groups, expand into Display/PMax, test new geographies.

Most accounts that follow this sequence are profitable by month two and scaling by month three. Most accounts that skip steps are still unprofitable at month six.

Final thoughts

PPC is a discipline that punishes shortcuts. Every step in this guide exists because skipping it costs money — sometimes a lot of it. Run through the sequence on a small test budget for two weeks before scaling, and the same campaigns that lost money in your first month will start making it back in your second. Pay-per-click rewards patience and process more than cleverness.

Need help applying this to your own site? I'm Shani Maurya — a freelance web developer and digital marketer based in Delhi. If you'd like a hands-on audit or full implementation, get in touch — I usually reply within a few hours.