Choosing a niche for ad revenue: the rate is the smaller half
Every niche list ranks verticals by what advertisers pay, and that is the less important of the two numbers. Expected revenue is the rate multiplied by traffic you can realistically win — and the highest-paying niches are the hardest to rank in precisely because the rates attracted everyone else first.
- Write down both numbers, not one. For each candidate niche, estimate what it pays and — separately — how much traffic you could plausibly hold in two years. Multiplying them produces a very different ranking from either alone.
- Look at who currently ranks. Search four or five terms you would want. If page one is banks, insurers, national publishers and comparison platforms, you are not competing on content quality; you are competing with acquisition budgets.
- Find the sub-question the incumbents ignore. Large sites cover broad terms and leave specifics alone. A narrow, well answered question in an expensive vertical beats a broad one you will never rank for.
- Check what your audience does to ads. Some audiences block them at rates that quietly halve the effective rate. That belongs in the estimate, not in a surprise six months later.
- Pick something you will still be writing about in two years. Authority compounds and takes that long to matter. A niche chosen purely on rate, that you find tedious, loses to a moderate one you keep feeding.
Two numbers, and everyone quotes the wrong one
Search for the best niches for ad revenue and you will find the same table
everywhere: verticals ranked by what advertisers pay. Legal at the top,
insurance and finance behind it, gaming and lifestyle at the bottom.
The table is not wrong. It is half the calculation.
Expected revenue = the rate × traffic you can realistically win
Every published ranking gives you the first term and silently assumes the
second. And the second is almost always the binding one — because the two terms
are negatively correlated. Rates attract competition, competition raises the
cost of ranking, and the niches that pay most are the ones where everyone with
an acquisition budget arrived years ago.
What that looks like in practice
Legal contains the most expensive keywords sold anywhere, almost
entirely from personal injury. It also has traffic that is small, episodic and
intensely local, contested by firms buying their own clients. High rate,
minimal winnable traffic.
Food has unremarkable per-click rates and produces some
of the best publisher revenue online, because a meal-planning visitor opens
several pages, stays on each while cooking, and comes back next week. Modest
rate, enormous winnable traffic, and session behaviour that multiplies both.
Gaming has vast audiences and among the worst economics
available: young readers, the highest ad-block rates online, and advertisers who
are mostly other games on user-acquisition budgets. Large traffic, low rate,
and a chunk of the audience that never produces an impression at all.
Rank those three by CPC and you get one order. Rank them by what a new
independent publisher would actually earn and you get close to the reverse.
The check that takes ten minutes
Open a search for four or five terms you would want to rank for, and look at who
is already there.
If page one is banks, insurers, national publications and comparison platforms,
you are not competing on content quality. You are competing with organisations
that capture the whole customer rather than an ad impression, and can therefore
outspend you on content indefinitely without it being irrational.
That is not a reason to avoid the vertical. It is a reason to enter it somewhere
else — at the specific question the incumbents have no incentive to answer well.
Insurance rewards this more than most: nobody wins "car
insurance", and plenty of independents do well on a particular exclusion, a
particular situation, or a particular region.
The third number nobody puts in the table
What your audience does to your ads.
A technically capable audience blocks at rates that quietly remove a large share
of your impressions before any of this arithmetic applies.
Technology and gaming lose most;
food, parenting and
home and garden lose far less.
That belongs in the estimate at the start rather than arriving as a
surprise once the site is built.
The honest recommendation
Pick the intersection of three things: a subject you will still be writing about
in two years, an audience that recurs rather than visits once, and a vertical
where you can name the questions the incumbents are ignoring.
If that lands on personal finance or
insurance, excellent — those rates are real. If it lands on
home and garden or automotive, also
excellent, and considerably more winnable.
If it lands on gaming, go in with your eyes open and plan to earn from
something other than display.
Whatever you choose, what each niche actually
pays is published from measured data rather than estimated — which is
the number worth planning against once there is enough of it.
Frequently asked
- Which niche has the highest RPM?
- On per-click rates, legal and insurance top essentially every published ranking, driven by personal injury and by policies that renew for years. On actual publisher revenue the answer is different, because RPM depends on pages per session and page length as much as on click value — which is why food sites earn well on unremarkable CPCs and gaming sites earn poorly on enormous audiences.
- Should I just pick the highest-paying niche?
- Only if you can rank in it. Finance, insurance and legal are the hardest verticals in search precisely because the rates attracted everyone with a budget. A high rate on traffic you never win is zero.
- How small a niche is too small?
- The question is not size but whether the audience recurs. A narrow topic people return to monthly beats a broad one they visit once, because repeat visitors cost nothing to acquire and compound your session numbers.
- Can I change niche later?
- You can, and it costs more than people expect. Topical authority is accumulated per subject; pivoting resets much of it. Better to narrow within a niche than to jump between them.