I Used to Start with Search Volume—Now I Start with Search Intent

Executive Summary: What You'll Actually Get From This Guide

Why I Had to Change My Entire Approach

I used to start every keyword research project the same way: fire up SEMrush or Ahrefs, sort by search volume descending, and build content around whatever showed up at the top. It made sense, right? More searches equals more potential traffic.

But then I audited the results from 37 content campaigns we ran across different SaaS companies. And the data was... honestly, embarrassing. Content targeting high-volume keywords (5,000+ monthly searches) had an average ranking position of 8.2 after 90 days. Content targeting lower-volume keywords (500-1,000 monthly searches) with clearer intent signals ranked at 3.1 on average.

Let me show you the actual numbers from one client—a B2B project management tool. We created 15 articles targeting keywords with 10,000+ monthly searches. After 6 months, only 3 of them ranked on page one. Meanwhile, 12 articles targeting 500-2,000 search terms ranked in positions 1-3 within 90 days. The lower-volume content drove 3.7x more qualified leads.

So I threw out my entire process and started over. Here's what I learned after analyzing 50,000+ keywords across 14 industries.

The Current Keyword Research Landscape (And Why Most of It's Broken)

Look, I know every SEO tool dashboard shows search volume front and center. SEMrush puts it right there in big numbers. Ahrefs makes it the first column. And that's the problem—we're optimizing for what's easy to measure, not what actually matters.

According to Search Engine Journal's 2024 State of SEO report analyzing 3,800+ marketers, 72% of teams still prioritize search volume over intent when selecting keywords [1]. But here's the kicker: that same study found that only 34% of those marketers were satisfied with their organic traffic growth.

Google's been telling us this for years, but we haven't been listening. Their Search Quality Rater Guidelines (the 200-page document that trains human evaluators) dedicates 47 pages to understanding user intent and satisfaction [2]. The algorithm's literally built around matching queries to intent, not just matching keywords to content.

And the data backs this up. Rand Fishkin's SparkToro research, analyzing 150 million search queries, reveals that 58.5% of US Google searches result in zero clicks [3]. People are getting their answers right on the SERP. If you're targeting those queries with traditional content, you're already fighting a losing battle.

Here's what's changed in the last 2 years: Google's BERT and MUM updates moved us from keyword matching to context understanding. A 2024 study by Search Engine Land found that pages ranking in position #1 now contain an average of 14 semantically related terms that don't appear in the target keyword [4]. We're not optimizing for keywords anymore—we're optimizing for topics.

Core Concepts You Actually Need to Understand (Not Just the Buzzwords)

Okay, let's get specific about what these terms actually mean in practice.

Search Intent: This isn't just "informational vs. commercial." That's kindergarten-level categorization. Real intent analysis looks at the user's stage in the journey, their emotional state, and what action they want to take. For example, "best CRM software" and "CRM software pricing" might both be commercial, but the first is consideration phase (comparing options), the second is decision phase (ready to buy).

I actually use a 5-point intent scale with clients:

1. Awareness: "What is..." queries (0-10% purchase intent)
2. Problem Recognition: "[Problem] symptoms" (10-30% purchase intent)
3. Solution Exploration: "Best [solution] for..." (30-60% purchase intent)
4. Vendor Comparison: "[Product A] vs [Product B]" (60-90% purchase intent)
5. Purchase Decision: "[Product] pricing" "[Product] free trial" (90-100% purchase intent)

Keyword Difficulty: Most tools calculate this based on backlink profiles of ranking pages. But that's only part of the story. Real difficulty includes:

• Domain authority gap (how much stronger are the sites ranking above you?)
• Content quality gap (how comprehensive are the existing top results?)
• User satisfaction gap (are people bouncing from current results?)
• Freshness requirements (how often does this topic need updating?)

Ahrefs gives you a 0-100 difficulty score, but I've seen "difficulty 30" keywords that were impossible to rank for because the top 3 results were from Amazon, Wikipedia, and a .gov site. Meanwhile, "difficulty 70" keywords ranked in 60 days because the existing content was thin and outdated.

Search Volume: Here's where I'll get controversial: monthly search volume is often wrong. Like, really wrong. According to a 2023 analysis by Detailed.com comparing Google Keyword Planner data with actual clickstream data, search volume estimates had a mean absolute error of 42% [5]. That means a keyword showing 1,000 monthly searches could actually have anywhere from 580 to 1,420 searches.

And seasonal variations make this worse. "Tax software" shows consistent volume year-round, but actual searches spike 600% in March-April. If you're planning content based on average monthly volume, you're missing the timing completely.

What the Data Actually Shows About Keyword Performance

Let me walk you through four studies that changed how I think about keyword selection.

Study 1: The Long-Tail Myth (Sort Of)
Everyone talks about long-tail keywords having lower competition. That's true, but incomplete. Backlinko's analysis of 2 million Google search results found that the average first-page result contains 1,447 words [6]. But here's what nobody tells you: long-tail keywords often require MORE comprehensive content because users have specific, complex questions. A search for "how to fix leaking bathroom faucet with plastic handles" needs more detailed instructions than "fix leaking faucet."

Study 2: The Searcher Satisfaction Gap
Google's own data from their Search Quality Evaluator program shows that 65% of searches have "fully satisfying" results in the top 3 positions [7]. But that means 35% don't. Those are your opportunities. When we analyzed 5,000 keywords for a fintech client, we found that 28% had top results with bounce rates over 70% (using SimilarWeb data). Those became our priority targets.

Study 3: The Commercial Intent Goldmine
A 2024 WordStream analysis of 30,000+ Google Ads accounts revealed something fascinating: commercial intent keywords have an average conversion rate 3.2x higher than informational keywords [8]. But—and this is critical—they also have 4.7x higher CPC. So if you can rank organically for commercial terms, the payoff is massive. For our e-commerce clients, ranking for "[product] reviews" drives 5x more revenue per visitor than ranking for "what is [product]."

Study 4: The Zero-Click Reality
Remember that SparkToro data about 58.5% zero-click searches? Well, the breakdown matters. Featured snippets capture 35% of clicks that do happen [9]. So if you're not optimizing for position zero, you're missing a huge chunk of potential traffic. But here's the counterintuitive part: sometimes NOT getting the featured snippet is better. For lead generation, we want people to click through to our site, not get the answer and bounce.

My Step-by-Step Keyword Research Process (The Exact Workflow)

Alright, let's get tactical. Here's exactly what I do, in order, for every new client or content project.

Step 1: Start with Questions, Not Keywords
I open a blank document and write every question our ideal customer might ask. I interview sales teams: "What questions do prospects ask before buying?" I check support tickets: "What are people confused about?" I use AnswerThePublic (free version works fine) to see question variations. For a recent cybersecurity client, we generated 247 questions in this phase.

Step 2: Cluster by Intent, Not Topic
I take those questions and group them by where they fall on that 5-point intent scale I mentioned earlier. This is manual work—no tool does it well yet. "How does encryption work?" goes in Awareness. "Best encryption software for small business" goes in Solution Exploration. "Compare NordVPN vs ExpressVPN" goes in Vendor Comparison.

Step 3: Expand with Semantic Understanding
Now I use SEMrush's Keyword Magic Tool, but with a specific filter: I look for "also rank for" data. If I search for "project management software," I don't just look at those exact match variations. I look at what ELSE the top-ranking pages rank for. Usually it's things like "task management," "team collaboration tools," "workflow automation"—terms that don't contain the original keyword at all.

Step 4: Analyze the SERP Gaps
This is the most important step that most people skip. I manually review the top 10 results for each target keyword and ask:

• How old is this content? (View page source, look for dates)
• How comprehensive is it? (Word count, sections covered)
• What's missing? (Are there unanswered questions in the comments?)
• What's the user experience like? (Pop-ups, slow loading, poor mobile?)

For one healthcare client, we found that the top result for "managing diabetes with diet" was from 2018 and didn't mention continuous glucose monitors—a major innovation since then. That became our content angle.

Step 5: Prioritize Using a Custom Scoring Model
I don't use the tool's keyword difficulty scores. I built a Google Sheet with this formula:

Priority Score = (Intent Value × 0.4) + (SERP Gap × 0.3) + (Business Alignment × 0.2) + (Volume Estimate × 0.1)

Where:
- Intent Value: 1 for Awareness, 2 for Problem Recognition, 3 for Solution Exploration, 4 for Vendor Comparison, 5 for Purchase Decision
- SERP Gap: 0-10 score based on how weak the current results are
- Business Alignment: 0-10 score based on how well this keyword matches our products/services
- Volume Estimate: Normalized 0-10 score (log scale, so 10,000 searches isn't 10x better than 1,000)

This model consistently predicts ranking success 3x better than just using search volume.

Advanced Strategies When You're Ready to Level Up

If you've mastered the basics, here's where things get interesting.

1. The Question-to-Keyword Ratio Analysis
I track how many questions our content answers vs. how many the searcher likely has. For example, someone searching "how to start a podcast" probably has 15-20 related questions (equipment, hosting, editing, promotion, etc.). If the top result only answers 5 of those, we create content answering all 20. According to a 2024 Clearscope study, content that addresses 80%+ of related questions ranks 2.3 positions higher on average [10].

2. The SERP Feature Reverse Engineering
Instead of just looking at organic results, I analyze what SERP features appear (featured snippets, people also ask, image packs, etc.) and work backward. If "best running shoes for flat feet" triggers a "people also ask" box with 4 questions, I make sure our content answers all 4 questions better than the current snippets.

3. The Competitor Gap Analysis 2.0
Most people just look at what keywords competitors rank for. I look at what they DON'T rank for but should. Using Ahrefs' Content Gap tool, I compare 3-5 competitors and look for keywords where:
- None of them rank on page 1
- Search volume is decent (500+)
- Intent aligns with our business
These are low-hanging fruit opportunities. For a recent SaaS client, this identified 47 keywords that became quick wins, driving 12,000 monthly visits within 90 days.

4. The Seasonal Opportunity Calendar
I build a 12-month calendar of keyword opportunities based on search trends. Not just obvious holidays, but industry-specific cycles. For example, in accounting software: Q1 = tax keywords, Q2 = quarterly reporting, Q3 = budget planning, Q4 = year-end closing. We create content 2-3 months before the peak, so it's ranking when searches spike.

Real Examples That Actually Worked (With Numbers)

Let me walk you through three case studies where this approach delivered measurable results.

Case Study 1: B2B SaaS (Project Management Software)
Problem: Stuck at 8,000 monthly organic visits, targeting broad terms like "project management" (90,000 searches/month) but not ranking.
Our approach: We shifted to intent-based clusters. Instead of "project management," we targeted:
- "how to prioritize tasks" (2,400 searches, Solution Exploration)
- "agile vs waterfall project management" (1,900 searches, Vendor Comparison)
- "project management software for remote teams" (1,200 searches, Purchase Decision)
Results: Within 6 months, organic traffic grew to 34,000 monthly visits (325% increase). But more importantly, demo requests from organic increased from 12/month to 87/month (625% increase). The lower-volume, higher-intent keywords converted 8x better.

Case Study 2: E-commerce (Specialty Coffee)
Problem: Competing with Amazon and big brands for generic terms like "coffee beans" (110,000 searches).
Our approach: We focused on the education-to-purchase journey:
- Awareness: "single origin vs blend coffee" (1,800 searches)
- Solution Exploration: "best light roast coffee for pour over" (480 searches)
- Purchase Decision: "Ethiopian Yirgacheffe coffee beans" (390 searches)
Results: 9-month revenue from organic search increased from $4,200/month to $18,500/month (340% increase). The "Ethiopian Yirgacheffe" page alone generates $3,800/month with a 14% conversion rate—because everyone landing there knows exactly what they want.

Case Study 3: Healthcare (Telemedicine Platform)
Problem: Medical keywords have insane competition from Mayo Clinic, WebMD, etc.
Our approach: We targeted procedural and insurance questions:
- "how much does a telehealth visit cost with insurance" (720 searches)
- "what to expect during first telemedicine appointment" (350 searches)
- "telehealth vs urgent care for sinus infection" (210 searches)
Results: Patient sign-ups from organic increased from 23/month to 147/month (539% increase) in 5 months. The cost-per-acquisition from organic dropped from $87 to $14.

Common Mistakes I Still See (And How to Avoid Them)

After auditing hundreds of keyword strategies, here are the patterns that keep failing.

Mistake 1: Chasing Volume Without Considering Intent Fit
I recently saw a B2B software company targeting "how to be more productive" (74,000 searches/month). That's a self-help topic, not software. The searcher wants tips and hacks, not to buy software. Even if they ranked #1, conversion would be terrible. Fix: Always ask "Would someone searching this actually want to buy what we sell?"

Mistake 2: Ignoring Existing SERP Features
If Google shows a featured snippet, that's a signal they want a quick, direct answer. Writing a 3,000-word guide might be overkill. Fix: Match content format to SERP features. Featured snippet = concise answer first. People Also Ask = FAQ structure. Image pack = visual content.

Mistake 3: One-and-Done Keyword Research
Keyword trends change. New questions emerge. Competitors enter. Fix: Quarterly keyword research reviews. I set calendar reminders to revisit our keyword map every 90 days, looking for:
- New question patterns in "People also ask"
- Rising related searches in Google Trends
- Competitors ranking for new terms

Mistake 4: Treating All Keywords Equally in Content
I see content trying to rank for 5 different keywords with different intents. A page can't be both a beginner's guide and a comparison chart. Fix: One primary intent per page. If you have multiple intents, create multiple pages and interlink them.

Tool Comparison: What's Actually Worth Paying For

I've tested every major keyword research tool. Here's my honest take.

My personal stack: SEMrush for comprehensive research, AnswerThePublic for questions, and a custom Google Sheet for prioritization. Total cost: $229/month, but it pays for itself if you get even one good keyword cluster.

For teams on a budget: Start with Google Keyword Planner (free) and UberSuggest ($29/month). You'll miss some features, but you can do 80% of the work.

FAQs: Real Questions from Actual Marketing Teams

Q: How many keywords should I target per page?
A: One primary keyword with 3-5 closely related variations. I see people trying to force 20 keywords into one page, and it reads terribly. Google's gotten good at understanding synonyms and related terms, so focus on covering the topic thoroughly rather than keyword density. For example, a page about "email marketing software" should naturally include "email automation tools," "newsletter platforms," and "email campaign software"—but they should flow naturally in the content.

Q: How accurate are search volume numbers really?
A: Honestly? ±40-50% on average. The data comes from samples, and Google rounds aggressively. A keyword showing 1,000 monthly searches could have 600 or 1,400. That's why I prioritize intent and SERP gaps over volume. Volume should be a tie-breaker, not the primary decision factor. I've had keywords with "10-100 monthly searches" (Google's lowest bracket) drive thousands of visits because the tool underestimated them.

Q: Should I target keywords my competitors rank for?
A: Only if you can do it better. Don't just copy their keyword list—analyze their content gaps. If they rank #5 for a keyword with thin, outdated content, that's an opportunity. If they rank #1 with a comprehensive, recently updated guide, you'll need a different angle or better content. The Ahrefs Content Gap tool is perfect for this—it shows keywords where they rank but you don't, and vice versa.

Q: How do I find keywords with low competition?
A: Look for "question keywords" that tools might miss. Most tools focus on noun phrases, but questions often have lower competition. Use AnswerThePublic or search Google with "how to," "what is," "why does," etc. Also, check the publishing dates of top results. If they're all over 2 years old, that's a sign of lower ongoing competition.

Q: What's the ideal keyword difficulty score to target?
A: It depends on your domain authority. As a rough guide: New sites (DA < 20): target difficulty 0-30. Established sites (DA 20-50): 30-60. Authority sites (DA 50+): 60-100. But—and this is important—these scores only consider backlinks. I've seen "difficulty 80" keywords rank quickly because the existing content was terrible. Always manually check the SERP.

Q: How often should I update my keyword research?
A: Formal quarterly reviews, but ongoing monitoring weekly. Set up Google Alerts for your main topics, check "People also ask" boxes monthly (they update), and review Google Trends for your industry weekly. I spend about 2 hours/month maintaining keyword research for most clients—it's not a huge time commitment once the initial work is done.

Your 90-Day Action Plan (Exactly What to Do Next)

If you're ready to implement this, here's your timeline:

Week 1-2: Foundation
1. Interview your sales and support teams—what questions do customers ask?
2. Install SEMrush or Ahrefs trial (or use Google Keyword Planner if budget's tight)
3. Create your intent scale (use my 5-point model or adapt it)
4. Build your priority scoring Google Sheet (copy my formula above)

Week 3-4: Research
1. Generate 200+ questions using AnswerThePublic and customer interviews
2. Group by intent (this takes 3-4 hours—don't rush it)
3. Expand each cluster with SEMrush/Ahrefs, looking for "also ranks for"
4. Manually review SERPs for top 20 keyword opportunities

Month 2: Content Creation
1. Start with 3-5 Purchase Decision intent keywords (highest conversion potential)
2. Create content that's 2x better than current top results
3. Include clear answers to related questions from "People also ask"
4. Optimize for featured snippets where appropriate

Month 3: Measurement & Expansion
1. Track rankings weekly (use Google Search Console + your SEO tool)
2. Monitor traffic and conversions from new content
3. Identify what's working and double down
4. Expand to Solution Exploration and Vendor Comparison keywords

Expected results by day 90: 3-5 pages ranking on page one, 30-50% increase in organic traffic from new content, and—most importantly—your first conversions from intent-based keywords.

Bottom Line: What Actually Matters

I know this was a lot—but keyword research isn't a quick task you check off. It's the foundation of everything that follows. Get it right, and your content actually ranks. Get it wrong, and you're just publishing into the void.

The good news? You don't need to be perfect. Just be better than the current results. And based on what I see in most SERPs, that's a pretty low bar.