Your Agency Sells 20 Articles a Month. Is That Right?

Our own plan sells 20 articles a month for S$750. That number is on our programme page today, and we are going to argue against reading it the way most buyers read it.

Article count is a unit of supply, not a unit of value. It is easy to quote, easy to invoice and easy to check. None of that makes it the right thing to decide on. An owner with 18 genuine buyer questions and 240 articles a year has bought 222 pages of padding, and Google has a name for what that starts to look like.

Here is the question we think you should ask instead, and the arithmetic behind it.

Why every agency quotes you a number of articles

Because it is the only part of the work a buyer can count without trusting the agency.

Think about what else is on a content proposal. Research. Technical fixes. Internal links. Schema. A dashboard. Each of those is real, and each of them is unverifiable at the point of sale, by design: you cannot audit research you have not seen, and you cannot price a fix you do not understand. Article count is the one line on the page you can tick off at month end.

So agencies compete on it, because it is the number buyers compare. We do the same thing. We put "20 articles a month" on our pricing page because leaving it off made the plan harder to buy, not because 20 is the right number for your business. It is the right number for a plan that has to be one plan.

The honest version is that the right figure varies enormously, and the market knows it. On 24 October 2026 we ran "how many blog posts per month for seo" through our metasearch and read the first 15 results. Eight of them named a recommended pace. Their recommendations: one a month. Four to eight a month. Nine or more a month. Ten to fifteen to launch. Four a week. One to four a week. One drew the line for "high volume" at fifty a month. One post argued that four a month was pointless against competitors adding twenty.

That is eight answers spread across a fortyfold range, all presented as the answer. The same morning, Google Autocomplete returned four Singapore predictions for that seed, three of them asking for a number per month. The question is real. Page one does not answer it.

What the number is actually standing in for

Every sensible version of "how many articles" is really a proxy for: how many questions does a buyer ask before they pay you, and how many of them does your site answer right now?

That number is finite and it is knowable. A tuition centre has perhaps twenty. A freight forwarder has forty, because the documentation questions multiply. A travel agency has one per destination it sells, which is why we found four of eight Singapore travel agency sites listing under 20 pages in total while buyers search destination by destination.

Once you have written the answer to every one of those questions, properly, the marginal article is worth much less than the first one was. The work that is still worth paying for changes shape: updating the pages that went stale, fixing the ones that rank at position eleven, and measuring whether any of it moved. That is not a smaller job. It is a different job, and it should be priced differently.

We have also found that total page count predicts very little on its own. One rival in a Singapore search had 2,486 pages and another had 2, and the page count told us nothing about who answered the question better.

The part that should worry you

Google's spam policies define scaled content abuse as when "many pages are generated for the primary purpose of manipulating search rankings and not helping users".

Read that against a contract that commits an agency to produce a fixed number of pages every month for twelve months, regardless of whether twelve months of genuine questions exist. Nothing about volume is against the rules. What is against the rules is the purpose, and a quota invents a purpose: the page exists because the contract says twenty.

We will not pretend we are outside this. We publish heavily on our own site, 104 articles in the 30 days to 24 October 2026, and the only thing keeping that from being padding is that each one reports something we measured that day. The discipline is not the count. It is the rule that nothing goes out without a finding in it.

And the second uncomfortable fact: more articles have not been easy for us to prove. We logged 145 changes on singrank.com and tested the 34 that were old enough to judge. Not one passed a significance test. That is our own data, published, and it is the strongest argument we know against buying content by the kilo.

What we would do before signing anything

Four steps, and you can finish them this week without an agency.

1. Write down every question a buyer asks you before they pay. Not keywords. Questions, in their words, from your inbox, your WhatsApp and your phone. Most owners stop somewhere between 15 and 40.
2. For each one, open your own site and find the page that answers it in the first paragraph. Not mentions it. Answers it. Mark each question answered, half-answered or missing.
3. Count the missing ones. That number is your content project, and it is a project, not a subscription. If it is 24 and someone is selling you 20 a month for a year, you are buying 240 pages for a 24-page problem.
4. Decide what happens after the gap closes. This is the question that separates a good proposal from a bad one. Ask the agency directly: when my questions are all answered, what do I pay you for? If the answer is "more articles", you have learned something.

A fair proposal will have an answer ready: updating, consolidating, link work, measurement, and fewer new pages. Our view on what a month's work should actually show up as is in the seven things a monthly report should show.

What this costs you

Doing it this way is slower to buy and harder to sell, on both sides.

For you, it means a week of listing questions before you can judge a quote, instead of comparing two numbers in an afternoon. It also means accepting a less comfortable contract shape: a bigger first phase to close the gap, then a smaller ongoing one. Many owners prefer the flat monthly figure precisely because it is predictable, and that preference is reasonable.

For an agency, it means admitting the obvious: the work tapers. That is harder to price and harder to renew, which is why so few proposals say it out loud. Demand for the price question is climbing, too. In our own Search Console over the 28 days to 24 October 2026, impressions for "seo pricing singapore" were up 325% week on week, "seo price singapore" up 226% and "seo singapore price" up 132%. More people are comparing numbers, which makes a simple countable number more valuable to sell and less useful to buy.

There is also a case where the whole question is wrong. If nobody is searching for what you sell, no monthly figure helps, and we would rather tell you that: there are three situations where SEO is the wrong thing to buy.

One next step

Send us your list of buyer questions, even a rough one, through our contact page. We will check each one against your existing pages and tell you how many are genuinely unanswered. If the number is small, we will say so, and you will not need a twelve-month content plan. That part is free.

If the number is large and you want it closed, our programme is S$750/month with 20 articles a month, and how that price is put together is written down. We would rather you bought it knowing what the 20 is for.

FAQ

How many articles a month does a small business in Singapore need?

There is no single figure, and page one of Google gives at least eight. The useful number is the count of buyer questions your site does not currently answer in a first paragraph. For most small businesses that total sits somewhere between 15 and 40 pages once, not every month, after which the work shifts to updating and measuring.

Is publishing a lot of content risky?

Volume itself is not against Google's rules. Google's spam policies target scaled content abuse, which it defines as many pages generated primarily to manipulate rankings rather than to help users. A quota that outlives your genuine topics is how an honest content plan drifts into that definition.

Why does SingRank sell 20 articles a month then?

Because one plan has to carry one number, and 20 is the figure that fits the price. We think it is the wrong thing to decide on. If your real gap is 24 pages, say so before you sign, and we will tell you what the months after that should contain instead.

Can more articles guarantee more enquiries?

No. We logged 145 changes on our own site and tested the 34 old enough to judge; none passed a significance test. Content is how you become eligible to be found. Whether a specific page earns enquiries depends on the search, the competition and what the page actually says.

What should I ask an agency about article count?

Ask what happens when your buyer questions are all answered. A proposal that cannot describe month 13 without producing more pages is selling supply. One that describes updating, consolidating and measuring has thought about your business rather than its invoice.