Trend Hunter B2C

⬇ Download this section

Where this publication overrides General Guidelines: Audience is defined psychographically, not demographically—frame benefits around mindset and intent, never around age, gender, income, location, relationship status, or parental status. No demographic stereotyping in any form—this is the strongest anti-pattern. The editorial stance is restraint: present signals, let readers draw conclusions—report what is emerging and why, without pushing a verdict. The Trust Rule gates every trend page: a topic with no live “why now” does not surface as a trend at all. Every piece structures around four story elements—What It Is · Why It Matters · Who It’s For · How to Experience It—incorporated naturally rather than rigidly, against four reader intent tiers (Discovery / Understanding / Evaluation / Action). The voice is Smart, Approachable, Curious, Practical. Trendy slang (“slay”, “bestie”, “vibes”) and academic density are both banned. Public-facing headlines and body ban em-dashes, en-dashes, and semicolons (colons allowed) and take no leading numeral (SmartNews-bound variants excepted). Text in red throughout this page marks anything that overrides or goes beyond the General Guidelines.

Publication Overview

Trend Hunter B2C is McClatchy’s curiosity-first trend publication. It surfaces emerging trends, products, behaviors, and cultural shifts before they reach the mainstream and translates them into pragmatic value for the reader’s actual life. Format is mobile-first, short-paragraph, scannable—built for a reader who is short on time and high on curiosity.

This page is the single reference for how Trend Hunter B2C content sounds, reads, and gets structured. Read the Editorial Stance and the Trust Rule first—everything else follows from them.

Our Reader: The Curious Optimizer

The Trend Hunter B2C audience is The Curious Optimizer—a psychographic persona defined by mindset, not demographics. The persona spans every age, gender, income, and life stage, unified by how they think about new ideas and what they expect from content.

The reader:

• Seeks emerging trends before they go mainstream
• Is driven by curiosity, self-improvement, relevance, inspiration
• Wants to feel ahead, informed, and confident in their decisions
• Uses trends as tools to make better decisions for themselves and their families

See §4.2 The Curious Optimizer for the full persona profile, content framework, and CSA Target Audience Definition.

Psychographic Traits

Curiosity. Eager to explore emerging ideas, products, and shifting cultural signals. Content should feel like discovering something before everyone else knows about it.

Pragmatic Inspiration. Seeks aspirational trend content that translates into real-life action. Not just “look at this cool thing” but “here’s how you can actually use it.”

Low Time, High Value. Expects quick clarity, digestible insights, and efficient summaries that lead to immediate takeaway value. Respects their time with no fluff or padding.

Life-Optimizer. Uses trends as a tool to make better decisions. Content should help them improve some aspect of their daily life.

Taste for Novelty (with filters). Open to new ideas but relies on credibility and context to decide what’s worthy of attention. Must feel vetted and trustworthy, not gimmicky or sensational.

The Trust Rule

(REQUIRED)

If a trend is low-opportunity because it is already well covered, it should not surface as a trend at all.

Showing “here’s a trend” beside “there’s no opportunity here” is a contradiction that erodes a writer’s trust in the tool, and the writer learns to second-guess everything the platform surfaces. Protecting that trust is a first-order editorial requirement.

• Every trend page carries a clear “why now.” If a topic can’t justify its presence with a live signal, it isn’t a trend page—it’s an evergreen update, and belongs in a different workflow.
• Never frame saturation as opportunity. If the honest read is “this is everywhere already,” the content should not run under Trend Hunter unless there is something substantially new to say.
• Opportunity and surfacing move together. A trend on the page is itself a claim that there is room to say something new.

Editorial gut-check before publishing a trend page

• Can I name the signal that makes this a trend right now?
• Is there genuine room to add—a new angle, audience, or format—rather than repeating what’s already saturated?
• If a writer read only the page, would “why this, why now” be obvious?

If any answer is no, this is not a Trend Hunter trend page.

Editorial Stance & Voice

The core principle

Present signals. Let readers draw conclusions. Trend Hunter reports what is emerging and why it matters—without pushing a verdict. Writers surface the evidence (search velocity, cultural signals, cross-platform momentum) and trust the reader to reach their own judgment. This is editorial restraint as a trust mechanism, not hedging or vagueness.

Restraint is not neutrality-to-a-fault. A confident, well-sourced observation is restrained; a pushy directive is not. The line is between reporting a signal and issuing a command.

Restraint governs stance, not service. Giving the reader actionable next steps—how to experience a trend, the steps, the picks (Intent Tier 4, “How to Experience It”)—is service, not a pushed verdict. Provide the how; let the reader decide whether. “Here’s how to try it” is service; “you need to try this” is a verdict.

Voice in one breath:

• Observant, not breathless. We notice things early; we don’t hype them.
• Evidence-forward. Claims are anchored to signals a reader can see for themselves.
• Plainspoken and precise. Clear over clever; specific over sweeping.
• Confident but not prescriptive. We show the reader the road; we don’t grab the wheel.

Voice attributes

Write with a voice that is Smart, Approachable, Curious, and Practical.

• Smart—treats the reader as intelligent and informed
• Approachable—never elitist, exclusive, or insider-y
• Curious—shares genuine enthusiasm for discovery
• Practical—grounds every idea in real-world application

Do / Don’t

Tone Do’s

• Be positive but not hypey or over-the-top
• Be insightful without relying on jargon
• Be direct and digestible
• Use “you” and “your” naturally to speak to the reader
• Be curious, enthusiastic, and human
• Be trend-forward but maintain trustworthiness
• Frame benefits psychographically—”If you want…”, “If you’re looking for…”, “For anyone curious about…”

Tone Don’ts

• Never use juvenile or trendy slang—no “slay”, “bestie”, “vibes”, or attempts to sound current
• Never assume demographic attributes—age, gender, income, location, relationship status, parental status
• Never be overly technical or dense
• Never be sensational or clickbait-y
• Never be condescending or talk down to readers

Tone by context

Voice is constant; tone flexes to the moment. Restraint holds in all four.

A “deep explainer” is a depth treatment of a trend (patient, clarifying)—not a separate named format. The named formats are Trend pages (the anchor) plus the Beta/GTM Infographics, Prediction cards, and Video described under Format Standards.

Intent Tiers

(REQUIRED)

Structure content around the reader’s intent level. A single piece may address one or multiple tiers—pick what the topic warrants and signal the tier in the lead.

Tier 1: Discovery—”Entertain me”

“Show me something new or surprising in a topic I care about.”

• Hook with novelty and the unexpected
• Lead with what makes this different
• Create that “I didn’t know that” moment

Tier 2: Understanding—”Make sense of this for me”

“Explain what this is, why it matters, and who it’s for.”

• Provide clear context without over-explaining
• Connect to broader patterns or movements
• Make complex ideas accessible

Tier 3: Evaluation—”Help me compare or choose”

“Help me determine if this trend is worth my time or money.”

• Offer honest, balanced assessment
• Include practical considerations
• Respect the reader’s decision-making ability

Tier 4: Action—”Tell me what to do next”

“Give me the routines, itineraries, products, or steps.”

• Provide specific, actionable guidance
• Include clear next steps
• Make it easy to take action (the how, not a verdict—see Editorial Stance)

Article Structure

(REQUIRED)

Every piece of trend content should naturally incorporate these four elements, though not rigidly in this exact order:

What It Is

Clear definition in one sentence. Don’t bury the lead or make readers work to understand the basic concept.

Why It Matters

Relevance plus emotional value. Why should someone care about this trend? What does it mean for their life?

Who It’s For

Universal benefit framing, not demographic-specific. Frame around needs and interests, not age groups or stereotypes.

How to Experience It

Practical steps, tips, or product picks—the how, not a verdict (see Editorial Stance). Make it easy to act on without telling the reader whether they should.

Format Standards

Each named format inherits the voice and the Trust Rule; the rules below are additive. Trend pages are the anchor; the other three are Beta / GTM formats (not Alpha)—documented here as forward-state, not yet in the active CSA format roster.

Trend pages (the anchor)

The reference format and the home of the voice. Evidence-forward, measured, structured so the “why now” is unmissable. “Trend detail page” refers specifically to this anchor format.

• Open with the signal, not the conclusion. Lead on what’s moving and how we know.
• Establish “why now” within the first screen—no burying it below the fold.
• Layer supporting signals (search, cultural, cross-platform) as distinct, attributable points.
• Close without a verdict. Leave the reader with the picture, not a directive.

Infographics / images (Beta/GTM, not Alpha)

Visual-first: the image carries the load, text is minimal and supporting.

• Keep on-image text short and scannable—labels and anchors, not paragraphs.
• Caption adds context the image can’t; it doesn’t restate it.
• One idea per graphic. If it needs two, it’s two graphics.

The per-unit asset set

Every trend unit—a deep-explainer treatment of the trend plus its two bridge pieces—carries a consistent visual set. Ranked by production value and reuse across syndication:

#2 is a Trajectory / Momentum card, not a prediction. It is backward-looking by construction—it reports where observed search demand has gone, sourced from SEMrush / Search Console (or published-engagement history), never from ATLAS’s internal prediction or heat scores, which are unvalidated. This is a different artifact from the Prediction cards format below, which is forward-looking by definition and carries an explicit confidence signal.

Deliberately excluded: Trend-of-the-Week sparkline and directional-arrow widgets. Those are cross-node weekly rollup formats, not per-unit assets, so they don’t compete for a node-level slot.

Infographic brand ruleset

The minimum every infographic follows regardless of type, to limit surprises and keep facts and presentation consistent. Each rule carries how it is enforced:

Palette and typeface superseded, 2026-07-28. §1 previously read “Oswald” and “coral / teal / purple.” The color NAMES came from the content lead’s guidance; the HEXES behind them were placeholders, inherited from the internal console design system, and the generator’s own header recorded them as provisional pending TH design sign-off. That palette has since been obtained — the live TH design system (the Hunter sandbox, staging.b2c.trendhunter.com/sandbox, documented in ops-hub/TRENDHUNTER_DESIGN_SYSTEM.md) — so the provisional status is resolved against the real thing rather than re-guessed. These are public TH brand assets, so the TH system governs them.

The three accents map onto TH semantic tokens, which fit these cards better than the colors they replace: the system defines Trend Up and Trend Down roles for exactly the rising/declining meaning a momentum or comparison card carries. Note that Primary #A3FC03 and Trend Up #BEFC4F are different roles, not shades of one another — do not substitute either for the other.

• Gate—checked automatically on the finished asset. An asset that breaks the rule cannot be approved, so you don’t have to check it yourself.
• Gate, partial—checked automatically on the figures the tooling formats. A figure typed in by hand is not re-checked; check that one yourself.
• Gate + Manual—part of the rule is machine-enforced and the remainder is yours; the row says which part is which.
• HITL gate (human-in-the-loop)—the asset cannot be approved until a person confirms the rule was met, or (where the rule is conditional) declares it doesn’t apply. The check trusts the confirmation; it does not re-derive it.
• Manual—not automated. You are the only check.

1. Visual consistency

2. Data accuracy & sourcing

3. Copy & claims

4. Presentation & legibility

5. Pre-publish check

Two further rules apply to every asset:

• No forecast framing. Gate—forward-looking language anywhere on the asset blocks it. None of the five asset types is exempt.
• No hidden content. Gate—hidden or zero-opacity content anywhere in an asset blocks it.

No workflow state is drawn on an asset (2026-07-30). Three badges used to be: a SAMPLE DATA · NOT REAL FIGURES / UNVERIFIED data-state mark, an AWAITING REVIEW mark on a --draft render, and a DRAFT · NOT FOR PUBLISH palette mark. All three are withdrawn, along with the gate that enforced the first and the matching prefixes on the alt text. The states themselves are unchanged and still fail closed — backing is still classified from evidence rather than from what a spec says about itself, an unsigned asset is still refused approval — and all of it is recorded in .meta.json and reported by the CLI. The reasoning: these marks answer questions the person running the pipeline has, and that person already knows what they fed it; drawing them put an operational annotation on a reader-facing artifact. Anything a reader needs must be true of the asset itself.

Prediction cards (Beta/GTM, not Alpha)

A distinct, named format—not a trend-page variant. Forward-looking by definition; every prediction carries a signaled confidence level so the reader knows how firm the forecast is.

• State the prediction plainly, then the confidence signal, then the basis for it.
• Confidence is explicit and consistent (e.g., emerging / building / strong)—never implied by tone alone.
• Frame as informed forecast: “signals point to” and “we expect,” never “this will.”

Video (Beta/GTM, not Alpha)

Spoken, human, same restraint. Follows a repeatable hook / content / close / credential architecture.

Format Guidelines

• Modern, clean, insightful, actionable, lightly aspirational
• Short paragraphs—2–3 sentences maximum
• Bulleted lists for scannability
• Pull-out “Quick Takeaways” for key insights
• Scannable subheadings that convey value—not generic labels

Headlines

• Focus on benefit or discovery angle
• Be specific rather than vague—specific nouns beat vague hooks (“X” over “this one thing”)
• Deliver on the promise in the body—no bait-and-switch
• The signal goes up front: readers should grasp “what’s moving” from the headline alone
• A great headline truncated is still a good headline
• Curiosity gaps or withheld payoffs must be earned and must serve the reader—restraint applies to headlines as much as body
• Sentence case—capitalize only the first word and proper nouns (unless a destination house style specifies otherwise)

Length—target the 90–104 character sweet spot. TH Headline Analysis (2026-06-02, 368 articles): headlines in the 90–104 band averaged the highest page views (614) and the strongest engagement; the <75 band drops to 352 PVs and 105+ falls off (340 PVs). The corpus median (86) runs slightly short—if a headline is under ~80 chars, it is usually missing an angle, a credential, or specificity. Add the missing element rather than padding. (Overrides the universal 80–100 default. A destination distribution platform with its own data-validated range—Apple News 90–120, SmartNews 70–90—takes precedence when the piece is built for that platform.)

Punctuation—avoid em-dashes (—), en-dashes (–), and semicolons (;) in public-facing headlines and body text. Em/en-dashes read as an AI-generation tell to the audience. Hyphens (-) in compound words are fine, and colons (:) are allowed (but don’t overuse them—vary sentence structure). Restructure the phrasing rather than swapping one banned mark for another. Avoid these marks unless a destination platform’s standard explicitly requires them. No terminal punctuation on headlines except a question mark.

Exception—proper nouns and titles. A banned mark that is part of a proper noun, publication name, or work or study title (e.g., the em-dash in Sympathy for Lady Vengeance—The Director’s Cut) is intrinsic to the name, not authorial punctuation, and is preserved verbatim. Where a destination platform enforces a hard character filter that cannot keep it, substitute a comma as the defined fallback, never stripping the mark with no replacement.

Numerals—keep numerals as numerals, even under 10 (space is tight in headlines), but never lead with a numeral—the signal opens the headline, not a count. A number-led headline is permitted only on a SmartNews-bound variant, where number-led heds are data-validated to perform.

Intent—match the headline’s question-type to the article’s job:

Choose deliberately: a How headline will not drive discovery traffic, and a What/Why headline on a deep explainer leaves engagement on the table.

Name the expert. When a piece draws on a credentialed source, study, or named expert, put that credential in the headline—not buried in the deck or body. Naming the expert is the single strongest performance lever observed (roughly double the page views, with higher engagement), and the only pattern that wins on both clicks and depth at once.

“Everything to Know.” Do not reach for “Everything to Know” / “Everything You Need to Know” as a fallback framing—it shows no measurable lift and tends to overpromise, so readers click and then bounce. Use it only when comprehensive coverage is genuinely the angle and a specific subject leads the headline. The topic does the work, not the phrase.

Openings

• Lead with the most interesting insight
• Don’t bury the hook in the third paragraph
• Create immediate value or intrigue
• First scroll must deliver what the headline promised. Especially for Experiences and travel round-ups—these reliably land the click but lose readers when the open does not pay off the headline. Front-load the promised payoff.

Word Choice

• Use contemporary but timeless language
• Avoid slang that will date quickly
• Prefer concrete over abstract

Style Mechanics

AP style

AP style is the baseline for all Trend Hunter written content. When a house rule on this page conflicts with AP, the house rule wins; otherwise defer to AP. (See the AP-Compatible tiers, §10.6.)

Punctuation

Ban em-dashes (—), en-dashes (–) and semicolons (;) in public-facing headlines and body text. This is a ban, not a preference—”avoid overusing” is not this rule. Em/en-dashes read as an AI-generation tell to the audience, which is why the mark is removed rather than rationed.

• Colons (:) are allowed, though don’t lean on them—vary sentence structure. Hyphens (-) in compound words are fine.
• Restructure the phrasing rather than swapping one banned mark for another.
• Proper nouns and titles are exempt: a banned mark inside a name, publication or work title is intrinsic to the name, not authorial punctuation, and is preserved verbatim. Where a destination platform enforces a hard character filter that cannot keep it, substitute a comma—never strip the mark with no replacement.
• Only a destination platform’s own standard explicitly requiring the mark overrides this.

Stated here as well as under Headlines deliberately: the ban governs body text, and a writer looking for a punctuation rule looks in Style Mechanics. It is one rule in two places, not two rules—punctuation.banned_marks in the conflict register is the machine-readable source, and write-assist enforces it programmatically.

Numbers, data & claims

• Attribute every data point: source and timeframe. “Up since spring per Google Trends,” not “way up.”
• If a claim can’t be sourced, it doesn’t run as a signal.
• Don’t lead a trend headline with a numeral (see Headlines).

Bylines & disclosure

The rule is about WHERE the disclosure goes, never WHETHER it is required. Both halves below are one rule and may not be quoted apart. A restatement that carries only the first half—”no AI disclaimer is required”—is a misstatement of this standard, not a shorthand for it.

• Byline: a standard human byline—no AI-assistant tag on the byline line itself. Editors remain accountable for accuracy and voice regardless of how a draft was generated.
• Therefore the disclosure moves; it does not disappear. When a piece is published via CUE (the platform’s standard surface for TH B2C), the standard AI disclosure applies—at the top of the article body / the CUE “Created With AI” checkbox, not the byline (compliance.ai_disclosure, distribution: cue). TH B2C gets no exemption from it.

If you are copying this rule into a writer-facing guide, copy both bullets. The byline bullet on its own reads as a blanket exemption and has already been reproduced that way downstream. Where only one sentence fits, use: “No AI tag on the byline; the standard AI disclosure still applies in the article body and the CUE checkbox.”

Auto-population

ATLAS-generated content is pre-filled as a starting draft for human editing—never published as-is.

• Treat auto-populated text as a first draft, not a final one.
• Verify every factual claim and every signal before publishing.
• Apply the voice and the Trust Rule during editing—auto-population does not.

Internal language

House terminology

• Do not use the word “influencer” internally. Use the team’s preferred term for creators / video talent.
• “Prediction card” is a proper, distinct format name—not a synonym for a trend page.
• “Trend detail page” refers specifically to the anchor format, not any trend-adjacent content.

Distribution & Channel Fit

The same trend often ships to multiple channels. Voice never changes across channels; framing and length adapt to each surface’s reader and constraints. A channel variant is a genuine reframe for that audience—not a lightly reworded duplicate—and follows the standard varianting workflow: a semantic-difference requirement, peer review, and escalation.

Channels, categories & collections (taxonomy)

Three independent structures organize Trend Hunter content. Keep the terms distinct:

• Channel—the reader-facing editorial section. Three channels (Mind-Body, Everyday Living, Experiences), each an editorial lens / reader-intent frame, not a topic bucket. (Channel definitions are set by editorial; scope pending final lock.)
• Category—the ATLAS topic tag on a trend (a fixed set of 14: wellness, food & beverage, beauty, fashion, travel, and so on). ATLAS-assigned, carried through.
• Channel and Category are independent axes. Any of the 14 categories can appear in any channel—categories are not partitioned across channels. A piece carries both a channel and a category; any combination is valid. Channel is assigned at the (curated) collection level and inherited by the piece, not derived from the category.
• Collection—a curated editorial grouping of trends. A collection belongs to a channel.
• Cluster—an algorithmic grouping of related trends (distinct from a curated Collection).
• Connection—the ATLAS similarity graph (edges between trends).

Inclusivity Standards

(REQUIRED)

Frame content by psychographic need, never by demographic assumption.

Good Examples

• “If you want to stay ahead of what’s next…”
• “If you’re looking for smarter ways to…”
• “For anyone curious about…”
• “If you care about saving time…”
• “If you love discovering new approaches…”

Bad Examples—Never Use

• “Millennials are obsessed with…”
• “Women over 40 should try…”
• “High-income shoppers prefer…”
• “Parents will love…”
• “Gen Z is all about…”

Image alt text

(REQUIRED)

Every editorial image carries alt text describing what is in the image. The article headline is not alt text.

Reusing the headline is the current practice on the sister titles, and it was adopted for a real reason: it makes a file findable later. It fails the purpose alt text serves. A screen-reader user hears the headline announced, then hears it again where the picture is, and learns nothing about the picture. The headline describes the article; alt text describes the image.

What alt text is not. Not the credit, which is its own field. Not the caption, which a sighted reader also sees. Not a keyword slot. And not a filing label: the find-my-file need that produced the headline practice belongs in the filename or the image library’s own metadata, where it does not sit between a reader and the page.

Scope. This extends the alt-text requirement already gated on generated assets to every image placed in the CMS, including stock and licensed photography. Accessibility is why the rule exists; discoverability is a side effect, not the goal.

Review & Escalation

Standards only hold if the workflow enforces them. Review confirms three things in order:

1. Trust Rule first. Does this justify surfacing as a trend? If not, stop—no amount of polish fixes a page that shouldn’t exist.
2. Voice second. Signals presented, no verdict pushed, evidence attributed.
3. Mechanics third. AP style, headline within the platform target, claims sourced, house terminology.

Escalate when: a trend’s “why now” is genuinely contested; a prediction’s confidence level is hard to call; or tool friction is blocking the standard (escalate as a platform issue, not an individual failure).

Anti-Patterns

Beyond the system quality standards, these patterns specifically undermine credibility with the Curious Optimizer audience:

Pre-Publish Checklist

Trust Rule + audience + voice:

• Signal named and attributed (source + timeframe); “why now” obvious within the first screen
• Room to add something new—not saturated coverage
• No verdict pushed—reader left to conclude
• Reader is framed psychographically (mindset / interests / intent)—no demographic assumptions about age, gender, income, location, relationship status, or parental status
• Voice is Smart, Approachable, Curious, and Practical; no juvenile or trendy slang (“slay”, “bestie”, “vibes”, etc.); no condescension

Structure:

• “What It Is”—clear one-sentence definition surfaced early
• “Why It Matters”—relevance and emotional value present
• “Who It’s For”—universal benefit framing, not demographic-specific
• “How to Experience It”—actionable steps, tips, or picks included
• Intent tier(s) clear (Discovery / Understanding / Evaluation / Action)

Format:

• Short paragraphs (2–3 sentences max)
• Bulleted lists used for scannability
• Pull-out “Quick Takeaways” for key insights where appropriate
• Subheadings convey value, not generic labels
• Hook in the first paragraph, not buried

Headline:

• Focuses on benefit or discovery angle; specific, not vague; body delivers on the promise
• Front-loads the signal and fits the platform target band (90–104 default)
• Sentence case; no leading numeral (SmartNews-bound variants may lead with a number); no terminal punctuation except a question mark; no em-dash, en-dash, or semicolon

Mechanics:

• AP style and house terminology applied (no “influencer”)
• Auto-populated text verified, not published raw
• Byline is a standard human byline; when published via CUE, the standard AI disclosure is applied (top of body / CUE checkbox)—not exempt

Anti-patterns avoided:

• No saturated topic surfaced as a trend; no pushed verdict; no unsourced signal
• No generic “trend roundup” without insight; no dense academic explanation
• No demographic stereotyping; no clickbait or sensational framing; no buried takeaways