A $69 carbon steel wok earns Serious Eats about $3.10 at Amazon's 4.5% kitchen rate, according to Ahrefs' affiliate marketing guide. Commissions that small only add up across thousands of readers and hundreds of recommendations. So your niche has to offer enough buyers, enough topics and a realistic chance of ranking.
Most beginners pick a niche by gut feel. This guide replaces that with a scorecard you can run in an afternoon on two or three candidates. You will score each one, check the result against real search results, and test whether the niche could grow an audience beyond Google.
Step 1: List two or three candidates and set the scoring rules
Write down two or three niches you could see yourself covering for a year. Stop at three. Scoring ten ideas shallowly tells you less than scoring three carefully.
Ahrefs' guide names four criteria for a good niche: competition, commercial value, breadth and personal interest. Score each from 1 to 5 using this rubric. Treat 2 and 4 as in-between judgments.
| Criterion | 1 (weak) | 3 (acceptable) | 5 (strong) | |---|---|---|---| | Competition | Page one is almost all large publishers with strong brands | Large publishers share page one with some independent sites | Independent sites, forums or thin pages rank prominently | | Commercial value | Few "best X", "X vs Y" or "X review" queries, or no merchant program | Some buying-intent queries and at least one program with modest terms | Many buying-intent queries and several programs with clear, workable terms | | Breadth | You can list fewer than 10 article or video ideas | You can list around 30 | You can list 50 or more, with room to expand into adjacent topics | | Personal interest | You would dread month three | You are mildly curious | You would read and watch this for fun |
Pass rule: a niche passes if it totals 14 or more out of 20 and scores at least 3 on every criterion. The minimum-of-3 rule matters. A niche with huge commercial value and no interest from you still fails, because affiliate marketing is a long game and slow early results test your motivation.
These thresholds are a starting convention, not a validated formula. Adjust them if you like, but set them before you score so you can't bend them to fit your favorite idea.
Step 2: Score competition and commercial value from real search results
Open an incognito window and search the queries a buyer would type: "best [product]", "[product] vs [product]" and "[product] review". Do this for five to ten products in each niche, and note who ranks in the top ten for each query.
For competition, ask one question: could a solo site plausibly sit on this page? Ahrefs' guide says a solo creator will not outrank Wirecutter or Good Housekeeping on "best vacuum cleaner". If every result comes from a major publisher, score competition 1 or 2. If independent sites, niche blogs, forum threads or YouTube videos appear, score higher.
For commercial value, count the distinct buying-intent queries you can find, then confirm that merchants pay for them. A "best [product]" query signals a reader comparing options, but it has value only if a merchant program exists.
Commission structures vary widely. According to the guide, Amazon pays 1 to 10% depending on category, Shopify pays up to $150 per new paying merchant, and some programs pay recurring commissions. A niche built entirely on low-rate, low-price products needs far more traffic to earn the same amount. See related: how to label free articles on a paywalled site: test plan for additional background.
If you pull keyword numbers from a tool, record the tool name and the date next to every figure. Volumes shift and tools disagree, so an undated number is not evidence.
Step 3: Score breadth and personal interest
Test breadth by brainstorming. Set a 20-minute timer and list every article or video you could make: comparisons, buyer guides, how-tos, maintenance, accessories, beginner mistakes.
The guide's own example helps here. A site about pulled pork sandwiches would run out of topics within a month, and a site about food is too broad to tackle. Smoking meats is about right, with plenty to cover and room to expand later.
If you stall at 10 ideas, the niche is too narrow. If your list splits into unrelated clusters with no shared reader, it is too broad. A good niche has one reader whose next question is always another topic you can cover.
For personal interest, be honest rather than ambitious. Ask whether you would still research this topic in a month when the site earns nothing. The guide cites a 2023 Authority Hacker survey of 2,270 affiliate marketers: those with less than a year of experience averaged $636 per month. It adds that a handful of very successful affiliates pull the overall averages up, so the typical affiliate earns less than the headline figure. Expect a slow start, and score interest on whether you can sustain it.
Step 4: Run a worked example
The three niches below are hypothetical. They show how the rubric separates a too-narrow, a too-broad and a workable candidate, following the guide's pulled pork, food and smoking meats examples. The query counts and scores are illustrative, not tool data.
| Hypothetical niche | Buying-intent queries found (illustrative) | Competition | Commercial | Breadth | Interest | Total | Result | |---|---|---|---|---|---|---|---| | Pulled pork sandwiches (too narrow) | About 5 | 4 | 2 | 1 | 4 | 11 | Fail | | Food (too broad) | Hundreds | 1 | 4 | 5 | 3 | 13 | Fail | | Smoking meats at home (workable) | About 60 | 3 | 4 | 4 | 4 | 15 | Pass |
In this example, pulled pork sandwiches score well on competition because few big sites would target the exact phrase. But the topic runs dry and offers little to sell, so breadth 1 and commercial 2 sink it.
Food scores a 5 on breadth. Page one for nearly every query would belong to publishers with massive brands, though, so competition scores 1. That breaks the minimum-of-3 rule and the total falls short anyway. See a closer look at affiliate site audit: measure your google traffic risk for additional background.
Smoking meats passes without a perfect score, which is typical. A workable niche is rarely exceptional on every criterion. It is adequate on all of them, with enough product categories (smokers, thermometers, pellets, accessories) to compare and review.
Step 5: Verify, then check for traffic beyond Google
A score is only as good as the evidence behind it. Before you commit, run this checklist on your top candidate.
- Spot-check the top ten results for five buying-intent queries and record every site that ranks.
- Count independent sites versus large publishers. If independents hold at least a few spots, a solo site has precedent.
- Confirm that merchant programs exist for the products you would recommend, and read the commission rates, cookie durations and payout terms.
- Search YouTube for the same products. Active videos with real comments signal demand beyond text search.
- Look for communities, such as forums, subreddits or Facebook groups, where people in the niche ask for recommendations.
Then run the check most beginners skip: could this site earn traffic that does not come from Google? Ahrefs' guide stresses that search-only affiliate sites are fragile. It reports that Future plc's affiliate revenue fell 24% in its most recent half-year, which the company blamed on lower search audiences. The growing publishers, by contrast, have strong brands that people visit directly.
The guide also notes that HouseFresh lost about 95% of its Google traffic after a 2023 update. By October 2025 it was back to pre-update levels, helped in part by a YouTube channel.
Answer yes or no to four questions about your niche:
- Do YouTube videos on these topics attract real viewing and comments?
- Would readers plausibly subscribe to a newsletter about it, such as a regular buying update?
- Could the site become a name people search for directly?
- Does the niche have communities where you could contribute and be found?
Also read: see how to choose b2b marketing channels: a scoring model
As a rule of thumb, three or more yeses mean the niche could support a brand. Fewer than three does not disqualify it, but it should lower your confidence and push you to plan a video or newsletter channel from day one.
This check echoes a question from Google's helpful content guidance, quoted in the guide: whether you have "an existing or intended audience" that would find your content useful if they came directly to you.
Finally, record your scores, the pass or fail result, and two sentences of reasoning. That note is your documented decision, and you can revisit it if the niche underperforms.
Limitations
A scorecard is a judgment aid. Two people can score the same niche differently, and that is fine as long as the reasoning is written down. Keyword volumes are estimates that vary by tool and date, so treat them as relative signals, not promises.
Nothing here predicts rankings or income. Ahrefs' guide warns that industry growth is not a personal income forecast and that affiliate income can swing from month to month. AI answers in search may also reduce clicks: an Ahrefs study cited in the guide found that an AI Overview correlates with a 58% lower average clickthrough rate for the top-ranking page. A passing score means the niche is worth testing, not that it will work.
What to do next
After you pick a niche, build content around your highest-scoring buying-intent queries and start one direct-traffic channel, such as YouTube or a newsletter. Revisit the scorecard after your first batch of content. If competition or interest scores drop in practice, adjust the niche before you invest another month.
