Writing guide
How to write a listicle
The listicle is the format everyone mocks and everyone clicks, and both reactions are rational. The number in the headline is a contract: it tells the reader exactly how big the page is, that they can jump to any item, and that someone has already done the work of dividing the topic into pieces. That calculability is why list posts keep winning clicks against essays on the same query — the reader can price the effort before spending it.
The mockery is earned by the padded version: ten items where seven are real, entries that paraphrase vendor homepages, no stated criteria, an order that means nothing, and an ending with no verdict. That page collects the click the number earned and then transfers the research burden straight back to the reader, who leaves with more tabs open than they arrived with. The format is not the problem; the filler is.
Writing a listicle that deserves its click is mostly editorial discipline: let the honest count set the number, define criteria before choosing items, draft every entry to the same skeleton with proof and a drawback, and end with a table and a verdict — the two blocks readers and answer engines actually extract.
What it is
A listicle is an article organized as a numbered list — tools, tactics, examples, mistakes — where each entry is a self-contained item a reader can jump to. Done properly it is a ranked, criteria-driven answer to “what are my options”, not a loose collection with digits attached.
When to use it
Use a listicle when the searcher’s real question is an inventory — best tools for X, ways to do Y, mistakes that break Z — and the options genuinely enumerate. Skip it when the question has one answer (that is a how-to), when the reader is deciding between two or three named options (that is a comparison post), or when you are slicing one continuous process into fake “items” — steps in disguise annoy the reader who came to compare.
The structure that works
An intro that restates the contract
The reader clicked a number and a promise. The first 100 words confirm both, and establish why your selection can be trusted over the next result’s.
- Confirm the promise and the reader in one line: “nine CRMs for teams under 20 people”.
- State the selection criteria in one or two lines — what got items in, what kept others out.
- On buying-intent lists, name your top pick immediately; withholding it reads as a trick to force scrolling.
Deliberate order
Order is information. A ranked list makes a claim with every position; a random sequence tells the reader you accumulated instead of judged.
- Rank best-first on buying-intent lists — readers and featured snippets both treat slot one as your verdict.
- Group by use case when items serve different jobs, and say so in the intro.
- Avoid countdown order for search traffic; the reader forced to scroll for the winner scrolls back to the results instead.
Numbered H2 items on a fixed template
Parallel structure is what makes a list scannable and comparable. Structural drift between items is what makes one exhausting.
- Use the same skeleton for every item: what it is, who it is for, one piece of evidence, one drawback.
- Hold items to 100-250 words; under that nothing is proven, over it the list stops being a list.
- Name the item and its niche in the H2 itself: “4. Trello — best for visual pipelines”.
A summary table
The most-extracted block on the page — skimmers decide from it, and featured snippets and AI answers lift tables whole.
- One row per item, 3-4 columns: name, best for, the one deciding spec (price, limit, setup time).
- Place it right after the intro or right before the verdict, not mid-list.
- Check it on a phone; a table that scrolls horizontally has stopped being a summary.
The verdict
A reader who finishes a list without a recommendation does their actual deciding on someone else’s page.
- Pick one item for the majority case and give the reason in two sentences.
- Add one or two runner-ups tied to scenarios: “if budget decides, take #6”.
- Introduce nothing new here — the verdict spends judgment already shown, not fresh information.
Step by step
- 01
Pick the number the topic honestly supports
Research first, count later. List every candidate you can find, then apply one cut: would you tell a friend to seriously consider this? The survivors set the headline number — if eight survive, the title says eight. Padding to a rounder ten is the move that earned the format its reputation; readers may not name the filler items, but they feel them, and they leave at the first one.
Match the count to the query’s intent. A “best CRM” buyer can genuinely evaluate 5 to 9 options; a “content ideas” searcher can absorb 25 short ones. The deeper each item must go, the smaller the honest count — at 100 to 250 words per item, a useful list caps out around a dozen entries before it becomes a directory.
- 02
Write the selection criteria before selecting
Set 2 to 4 criteria before choosing a single item: “under $50 a seat, native mobile app, exports your data without lock-in”. Criteria are what turn “some tools we thought of” into a ranking a reader can trust — and your defense when a vendor or a reader asks why something is missing. If you cannot say why an item is in, it is in because every other list has it.
Then write one sentence per criterion on why it matters to this reader. Those sentences become the intro’s trust paragraph almost verbatim, and the exercise catches criteria that exist to flatter a favorite rather than to serve the person choosing.
- 03
Order the list before writing it
Pick the ordering principle first: best-first for buying-intent lists, grouped by job when items serve different use cases, severity-first for mistake lists. Then state it on the page — “ranked by value for money for teams under 20” — because unexplained order reads as no order, and no order reads as no judgment.
Sanity test: read only the H2 sequence. If swapping items 3 and 8 would change nothing about the page, you have accumulated rather than ranked. Rework the sequence until every position is defensible, starting with whether item 1 deserves the slot the snippet will quote.
- 04
Draft every item to the same skeleton
Four beats per item: what it is in one line, who it is for, one piece of evidence, one drawback — inside 100 to 250 words. The skeleton forces the two sentences most listicles omit: the proof (a price, a tested result, a concrete observed detail) and the honest negative. An item with no drawback reads as a placed ad, and one placed ad discounts the whole list.
Keep the beats parallel across items: if item 3 states pricing, item 9 states pricing. Parallel structure is what makes entries comparable at a scan; asymmetry forces re-reading, and re-reading is exactly the effort the format promised to remove.
- 05
Give every item a distinct best-for
After drafting, extract each item’s best-for phrase into a single column and hunt duplicates. Two items “best for small teams” means the list has one item and one echo — merge them, or sharpen the split: “best for small teams still on spreadsheets” versus “best for small teams already living in Slack”. The column takes five minutes and catches what item-by-item editing misses.
The same exercise is your padding detector. An item whose best-for you cannot fill is for nobody, and it is on the list to inflate the number. Cut it and update the headline count — the smaller, fully-differentiated list is the one readers finish and cite.
- 06
Build the summary table and the verdict
The table gets one row per item and 3 or 4 columns: name, best for, and the one deciding spec — price, key limit, or time to set up. Place it right after the intro or just before the verdict. Tables are the most-extracted block on list pages — featured snippets and AI answers lift them whole — and on a phone a six-column table stops being a summary.
Then write the verdict as the answer to “fine — which one?”: a single pick for the majority case with a two-sentence reason, plus one or two runner-ups tied to scenarios (“if budget decides, take #6”). No new items may appear here; a verdict that introduces entry eleven admits the ranking above it was incomplete.
- 07
Write the intro last and audit the promise
Write the intro in under 100 words, last: confirm the promise and the reader (“nine CRMs for teams under 20”), state the criteria in a line or two, and on buying-intent lists give the top pick away immediately. Handing over the answer does not lose readers — it converts skimmers into believers who keep reading for the reasoning.
Finish with a promise audit of the headline: every qualifier must be true on the page. “Free” requires actual free tiers, “2026” requires you checked in 2026, “tested” requires that you tested. The format’s bad name comes from titles that write checks the items never cash; matching the promise is the entire trick.
Before and after: what good looks like
Before
4. Trello. Trello is one of the most popular project management tools on the market today. It uses a card-based system that makes it easy to organize tasks and collaborate with your team. It offers a wide range of features, including boards, lists, automation, and numerous integrations with other tools. Trello is known for its user-friendly interface and flexibility, making it a great choice for teams of all sizes. Pricing starts with a free plan, with paid plans offering more features.
After
4. Trello — best for small teams that live in a visual pipeline. If your work moves through stages — leads, editorial, hiring — Trello’s board is the fastest tool here to learn: in our setup test, a five-person team had a working pipeline in under 20 minutes. The free plan covers unlimited cards and 10 boards, genuinely enough for one team’s workflow. The drawback is depth: no native timeline or workload view until the $10-per-user Premium tier, and reporting stays thin even there. If you need cross-project dashboards, this is the wrong pick — see #7.
The before is a spec-sheet paraphrase: “most popular”, “wide range of features”, “teams of all sizes” — 80 words that fit any tool and help no one choose. The after opens with a distinct best-for, proves it with a tested number (a working pipeline in under 20 minutes), states the free plan’s real limits, and spends a quarter of its length on the drawback — including a pointer to a better pick for the mismatched reader. “The wrong pick — see #7” is what a trustworthy list sounds like.
Mistakes that sink a listicle
| Mistake | The fix |
|---|---|
| Padding to a round number | Let the survivors of your cut set the count. “8 CRMs” with eight real entries outperforms “10 CRMs” where readers bounce at the first filler item. |
| Items that paraphrase the vendor’s homepage | Add one verified detail and one drawback per item. If you have not used it, say what you did instead — ran the free tier, compared pricing at a real scenario — and claim only that. |
| No stated criteria | Two lines in the intro: what got items in, what kept others out. Criteria turn “some tools we thought of” into a defensible ranking. |
| Interchangeable best-fors | Assign each item a distinct reader before publishing. Three items “great for small teams” is one item and two duplicates. |
| Random order | Rank best-first for buying lists or group by job — and say which in the intro. Order is a claim; no order is an admission. |
| Ending without a verdict | Name one pick for most readers and one or two scenario alternatives. A list that just stops at item ten hands the decision — and the reader — to the next result. |
Pre-publish checklist
- The headline number equals the number of items a reader should actually consider.
- Selection criteria appear in the intro in one or two lines.
- The intro is under 100 words and, on buying-intent lists, gives the top pick away.
- The ordering principle is stated, and item 1 deserves its slot.
- Every item follows the same skeleton: what, who for, proof, drawback.
- Every item carries a distinct best-for; no duplicates survived the column check.
- Each item includes at least one number — a price, a limit, a tested result.
- Every item names a drawback; zero-drawback entries were rewritten or cut.
- The summary table has one row per item, 3-4 columns, and fits a phone screen.
- The verdict picks one item for the majority case plus scenario runner-ups.
- Every qualifier in the title — “free”, “2026”, “tested” — is true on the page.
Free tools for this work
Check your listicle before it ships.
Frequently asked questions
- How many items should a listicle have?
- As many as survive an honest cut — for most topics that is 5 to 12. Below five, write a comparison or a guide instead; past fifteen, readers stop comparing and start scrolling, so group entries under subheadings. Never pad to reach a rounder number: the first filler item is where readers leave.
- How long should a listicle be?
- Budget 100 to 250 words per item, plus a sub-100-word intro and a short verdict — so a nine-item list lands around 1,500 to 2,500 words. Item depth beats total length: eight entries with proof and a drawback outrank twelve one-liners on the same query, because thin items satisfy nobody’s search.
- Do listicles still work for SEO?
- Yes — the format matches inventory-shaped queries (“best”, “ways to”, “examples of”), and its summary tables and parallel H2s are exactly the structures featured snippets and AI answers extract. What stopped working is the padded, criteria-free version: thin interchangeable items get skipped by readers and answer engines alike.
- Should each list item be an H2?
- Yes — put the number in the H2 itself: “4. Trello — best for visual pipelines”. Numbered H2s let readers jump, generate a usable table of contents, and hand search engines the list structure explicitly. Keep one item per H2, and do not nest sub-items more than one level down; deeper nesting breaks the scan the format sells.
Free download
The Pre-Publish Checklist
This guide covers one format. The checklist covers every article. 47 checks WriteZu runs before an article ships — as a checklist you can run by hand.
No spam. The checklist, then only what's worth knowing.
Or let WriteZu handle the writing for you
It follows this exact structure — researched, drafted, scored against quality gates, and published to your CMS — while you approve from your inbox.
Start free