Blog Workflow
From idea to published post, as a repeatable process — including the publication checklist that catches what review misses.
A blog fails on consistency, not quality. The fix is a process that works when you don't feel like writing.
Most business blogs follow the same arc: four posts in the first month, two in the second, nothing after that. The writing was fine. The process didn't exist.
Read the editorial process · Request Your Free Systems Audit
Before any of this: decide what the blog is for
Three common purposes, and they produce different content:
Search. Answering questions people type. Measured in traffic and enquiries over months.
Credibility. Demonstrating you know the work, for people already considering you. Measured in whether prospects mention having read it.
Thinking in public. Working out your own positions. Measured in the quality of conversations it produces.
Most business blogs are vaguely all three and clearly none. Pick one as primary.
The workflow
1 · Idea capture
One list. Every idea goes in, nothing gets evaluated at capture time.
For each: the question it answers, who it's for, and the source — a customer question, a conversation, something you read.
Ideas sourced from real questions outperform invented ones consistently. The best list is the questions you're tired of answering.
2 · Selection
Monthly, pick what's getting written. Against three criteria:
- Does it serve the primary purpose?
- Do you have something specific to say, or just a topic?
- Can you finish it in the time available?
The second criterion kills most ideas. That's the point.
3 · Argument
Before drafting, write one paragraph: what this post claims and why anyone should believe it.
If it doesn't come, the idea isn't ready. Return it to the list.
4 · Outline
Headings plus a line under each. Fifteen minutes, and it saves hours.
A useful structure for most business posts: the problem in the reader's terms, why the obvious answer doesn't work, what to do instead, how to start.
5 · Draft
Start to finish without editing. Mark gaps as [TK] and keep moving rather than stopping to research.
6 · Review
Three passes, in order: structural, then line, then factual. Collapsing them means line-editing paragraphs you later cut.
The factual pass is mandatory. Every number, every claim, every link.
7 · Publish
The checklist below. Mechanical, so it belongs in a procedure.
8 · Distribute
Decided when the post was selected, not after publishing. Newsletter, relevant communities, direct sends to people who asked the question.
9 · Review date
Set six months out. Posts go stale and stale posts damage credibility more than missing ones do.
The publication checklist
Most blog errors happen here, not in the writing.
- Title under 60 characters, states the benefit or the question
- URL slug: lowercase, hyphens, no stop words
- Meta description 140–155 characters
- Feature image sized correctly, alt text filled
- Every image compressed
- At least two internal links to existing posts
- Every external link opens and goes where intended
- Headings in proper order — one H1, logical H2s
- Category and tags from the existing list, no new ones
- Author and date showing
- Preview checked on a phone
- Link preview checked — title and image render correctly
- Call to action present and correct
Thirteen checks, five minutes, and it catches the errors that make a good post look careless.
Structure that works
The opening earns the rest. Two or three sentences, stating the problem in the reader's own terms. No throat-clearing about how important the topic is.
Headings that say something. "Why response time matters more than price" beats "Response times." A reader scanning only the headings should get the argument.
Short paragraphs. Two to four sentences. Most reading happens on a phone.
One idea per section. If a section needs a sub-argument, it's two sections.
End with what to do. Not a summary of what you just said. The next action.
Length
As long as the argument needs. Which in practice means:
Answering a specific question: 800 to 1,200 words. Making a case: 1,500 to 2,500. Comprehensive guides: longer, with a contents block.
Padding to reach a word count is visible and it costs you the reader. So does cutting an argument short to hit a target.
Cadence
Pick one you can sustain through a busy month. Monthly, consistently, beats weekly for six weeks.
Keep two finished posts in reserve. One bad week then doesn't break the schedule, and the buffer removes the pressure that produces thin posts.
Updating rather than adding
An existing post that ranks and gets read is worth more effort than a new one.
Quarterly, look at what's performing. For the top posts: is anything out of date, is there a question in the comments or enquiries that the post should answer, can it be expanded?
Update the content, keep the URL, change the date, note the update at the top.
Updating a strong post usually beats writing a new weak one, and it's faster.
What to measure
| Measure | Tells you |
|---|---|
| Enquiries mentioning a post | Whether it's doing the commercial job |
| Traffic at three and six months | Whether it has durable value |
| Scroll depth or read time | Whether people finish |
| Questions it removed from your inbox | The most undervalued measure |
Published count is not a measure. Four good posts beat twelve thin ones on every dimension that matters.
Consistency is the whole game. Design the process for the month you don't feel like writing.
Want your content operation to actually run?
A free systems audit covers your editorial process and where it breaks down.