Keyword research that finds real opportunities

Video lecture

Keyword research that finds real opportunities

14 chapters · about 8 min · full transcript

Coming soon

Chapter 1 of 14

Keyword research that finds real opportunities

A six-step process
Search Console + Google Trends
Avoiding cannibalisation

The narrated lecture is in production

Every chapter is scripted and ready. Browse the chapters and read the full transcript now — the video will appear here when it’s published.

Chapters

Transcript of the narration, chapter by chapter.

0:00 Keyword research that finds real opportunities

Most keyword research produces a spreadsheet of big numbers that nobody acts on. Good keyword research produces a prioritised map of topics your audience actually uses, matched to pages you can realistically rank and that support business goals. In this lecture you'll learn a six step process, how to use Search Console and Google Trends as hands on sources, and how to avoid the classic trap of creating five pages for one idea.

0:32 Why it matters

Why does this matter? Because volume is seductive and misleading. A high volume head term might be dominated by global brands you can't beat, or have an intent that doesn't match your business. Meanwhile, lots of specific, lower volume queries convert well and are winnable. Good research points your limited content budget at the topics that will actually pay back.

0:58 Step 1: real customer language

Step one: start with seeds from real customers. Customer questions in emails, WhatsApp chats and sales calls. Reviews, yours and competitors'. Forums and communities. Comments on YouTube and TikTok videos. Write down the words customers use, not your internal jargon. A Pakistani fintech might say digital wallet, while customers search how to send money from JazzCash to bank. That gap between your words and theirs is where opportunities hide.

1:28 Step 2: expand with tools

Step two: expand with tools. Google Search Console, for queries you already appear for. Keyword Planner, for volume ranges and ideas. Autocomplete, People also ask and related searches. Third party SEO tools like Semrush, Ahrefs or Moz for estimated volume, difficulty and competitor keywords. And YouTube and TikTok search suggestions for video opportunities. All volumes are estimates. Use them to compare, not as forecasts.

1:56 Steps 3–4

Step three: evaluate each keyword on five questions. Relevance: would this searcher be a potential customer? Intent: which content type ranks, and can we make it? Volume: is there meaningful demand for our niche? Difficulty: who ranks now, big brands or small sites and forums? And business value: how close is it to a sale or lead? Step four: favour the long tail, longer specific queries that are easier to win and convert well, especially for newer sites.

2:30 Step 5: group by topic

Here's the key idea behind step five. Group keywords by topic, not one page per keyword. Think of it like a clothes shop. You don't build a separate shop for blue jeans, denim trousers and jeans for men. It's one rail, well organised. Search engines understand synonyms. If the results pages for several keywords largely overlap, they belong on one page, with one primary keyword and secondary variants. That prevents keyword cannibalisation, where your own pages compete with each other.

3:05 Step 6: prioritise

Step six: prioritise. Score each topic from one to three on business value, ranking feasibility and effort, and start with high value, feasible, low effort topics. Now the hands on part. For an existing site, export your Search Console queries and add three formulas from the lesson. One flags striking distance queries, positions eight to twenty with decent impressions. One flags question queries starting with how, what, why and so on. And one separates brand from non brand.

3:39 Seasonality with Google Trends

Add seasonality. Search volumes in tools are averages, but demand in many markets swings with Ramadan, Eid, back to school, summer travel, Black Friday or White Friday, and tax deadlines. Check each priority topic in Google Trends for the right country over five years, note the peak months, and publish or refresh six to eight weeks before the peak, so the page is indexed and current when demand rises. Remember Trends shows relative interest, not absolute volume.

4:12 Example 1 (simple)

Worked example one, simple. A UK online plant shop exports Search Console queries. The striking distance filter surfaces pet safe plants for cats at position eleven with steady impressions, and the question filter shows how often to water a snake plant. Two quick actions: improve the pet safe plants page with a clear table and sources, and add a watering section to the snake plant page. No new pages needed. Just better answers to questions the site already nearly ranks for.

4:47 Example 2 (illustrative)

Worked example two, realistic and illustrative. A Karachi education company researches freelancing courses. Seeds from WhatsApp enquiries include how to get paid from Upwork in Pakistan and which skills pay in dollars. Tools expand them. The team groups keywords where results overlap into eight topics, scores them, and checks Trends, which shows interest rising after exam results season. They publish the top three topics six weeks before that period, with the course page linked from each.

5:20 Group these

Let's pause and group some keywords together. Here are five: best budget laptop for students, cheap laptop for university, affordable student laptop, how to extend laptop battery life, and laptop battery replacement near me. How many pages? Search them and compare results. The first three share most of the same top results, so they're one page, a commercial comparison. Battery life tips are informational, so a separate guide. And battery replacement near me is transactional and local, so a service page. Five keywords, three pages, no cannibalisation.

5:58 Watch me do it: quick wins from Search Console

Watch me do it. I'll find quick wins for a UK plant shop from its own Search Console data. First, I open Performance, set the date range to the last three months, and export the Queries table. I paste it into a sheet. Next, I add the three formula columns from the lesson. The opportunity flag marks striking distance queries, positions eight to twenty with at least a hundred impressions, and low CTR queries. The question flag marks queries starting with how, what, why and similar words. The brand flag separates brand from non brand. Then I filter to non brand striking distance queries. Twenty three rows. Pet safe plants for cats is at position eleven with steady impressions. Then I filter question queries. How often to water a snake plant appears, with the snake plant product page ranking for it. Now I check seasonality. In Google Trends, set to the UK, I look at pet safe plants over five years. Interest rises in spring. So I plan the refresh of that page for early March. Finally, I write five actions: improve the pet safe plants page with a clear table and sources, add a watering section to the snake plant page, and three more. All from data the site already had.

7:30 Common mistakes

Common mistakes. Targeting only high volume head terms on a new site. Treating tool estimates as precise. Creating separate pages for near identical keywords. Ignoring business value in favour of traffic. And publishing seasonal content the week of the peak, when it's too late to be indexed and trusted. Plus a quick cannibalisation check: filter a priority query in Search Console and open the pages tab. If two of your URLs keep swapping, decide which one owns it.

8:04 Recap

Recap. Start with real customer language, expand with Search Console, Keyword Planner, SERP features and SEO tools. Evaluate relevance, intent, volume, difficulty and value. Favour the long tail, group keywords with overlapping results into one page, and prioritise by value, feasibility and effort. Use Search Console formulas for quick wins and Google Trends for timing. Try this now. Export your last three months of queries and find five striking distance, non brand queries.

What good keyword research produces

Keyword research is not a list of high-volume words. It is a prioritised map of topics and queries your audience uses, matched to pages you can realistically rank and that support business goals.

Step 1: Start with seeds from real customers

Brainstorm seed topics from:

Customer questions in emails, WhatsApp chats, DMs and sales calls.
Product and service categories.
Reviews (yours and competitors').
Forums and communities such as Reddit, Quora and niche Facebook groups.
Comments on relevant YouTube and TikTok videos.

Write the words customers use, not your internal jargon. A Pakistani fintech may say "digital wallet"; customers may search "how to send money from JazzCash to bank".

Step 2: Expand with tools

Useful sources include:

Google Search Console (for existing sites): queries you already appear for.
Google Keyword Planner (free with an Ads account): volume ranges and related ideas.
Autocomplete, "People also ask" and related searches.
Third-party SEO tools (for example Semrush, Ahrefs, Moz, Ubersuggest, SE Ranking): estimated volume, difficulty, competitor keywords and SERP features.
Platform search: YouTube and TikTok search suggestions for video opportunities.

All volume numbers are estimates. Use them to compare keywords, not as precise forecasts.

Step 3: Evaluate each keyword

Factor	Question
Relevance	Would someone searching this be a potential customer or influence one?
Intent	Which content type ranks, and can we produce it?
Volume	Is there meaningful demand (relative to our niche)?
Difficulty	Who ranks now? Big brands and authoritative sites, or small sites and forums?
Business value	How close is it to a sale or lead?

Step 4: Favour the long tail

Long-tail keywords are longer, more specific queries ("vegan protein powder without stevia UK"). Individually they have lower volume, but together they make up a large share of searches, they are often less competitive and they convert well because the intent is precise. New sites should build momentum with long-tail and question-based queries before chasing broad head terms like "protein powder".

Step 5: Group keywords by topic, not one page per keyword

Modern search engines understand synonyms and related phrases. "Best budget laptop for students", "cheap laptop for university" and "affordable student laptop" usually belong on one page. Group keywords whose results pages largely overlap into a cluster and assign one primary keyword plus secondary variants to each page. This prevents keyword cannibalisation (several of your own pages competing for the same query).

Step 6: Prioritise

Score each topic quickly, for example 1–3 on business value, ranking feasibility and effort, and start with high-value, feasible, low-effort topics. A simple spreadsheet works:

Topic	Primary keyword	Intent	Value	Feasibility	Effort	Priority
Student laptops	best budget laptop for students	Commercial	3	2	2	High
Laptop battery care	how to extend laptop battery life	Informational	1	3	1	Medium

Multilingual and regional research

Research each market separately; UK, US, UAE and Pakistani searchers use different terms ("mobile phone" vs "cell phone", "pharmacy" vs "chemist").
In the Gulf, many people search in both Arabic and English; in Pakistan, English, Urdu and Roman Urdu. Create properly localised pages rather than machine-translating everything.

Hands-on: mine Search Console for keyword opportunities

For an existing site, your best keyword source is the queries you already appear for. Export Performance → Search results → Queries (last 3–6 months) to a sheet with columns A Query | B Clicks | C Impressions | D CTR | E Position, then add:

Opportunity flag (F2): =IF(AND(C2>=100, E2>=8, E2<=20), "striking distance", IF(AND(C2>=200, D2<0.01), "low CTR", ""))
Question query (G2):  =IF(REGEXMATCH(LOWER(A2), "^(how|what|why|when|which|can|does|is|are|best)\b"), "question", "")
Brand query (H2):     =IF(REGEXMATCH(LOWER(A2), "yourbrand|your brand|yb"), "brand", "non-brand")

Filter non-brand, striking-distance and question queries: they are often the fastest wins and the raw material for FAQ sections and new cluster pages. (Thresholds are illustrative — set them for your site's scale.)

Seasonality with Google Trends

Search volume estimates are averages; demand in many markets is strongly seasonal (Ramadan, Eid, back to school, summer travel, Black Friday / White Friday, tax deadlines). Check each priority topic in Google Trends for the right country over five years, note the peak months, and publish or refresh 6–8 weeks before the peak so pages are indexed and updated when demand rises. Trends shows relative interest (0–100), not absolute volumes.

A quick cannibalisation check

In Search Console, filter by a priority query and open the Pages tab. If two or more of your URLs regularly swap impressions for the same query, you may be splitting relevance. Decide which page should own the query, strengthen it, and either differentiate or consolidate the other (see the lessons on content refresh and pruning).

Common mistakes

Targeting only high-volume head terms on a new site.
Treating tool estimates as precise.
Creating separate pages for near-identical keywords.
Ignoring business value in favour of traffic.

Start with real customer language, then expand with Search Console, Keyword Planner, SERP features and SEO tools.
Evaluate relevance, intent, volume, difficulty and business value; volumes are estimates.
Long-tail keywords are often easier to rank for and convert well, especially for new sites.
Group keywords with overlapping results into one page to avoid cannibalisation, then prioritise by value, feasibility and effort.

Build a 20-row keyword sheet for a business: seeds from customer language, expanded ideas, intent, value, feasibility, effort and a priority rating. Group related keywords into clusters.

Enrol for free to save your progress

Reading is always free. Enrol to keep your place, take the final assessment and earn a verifiable certificate.

Keyword research that finds real opportunities