EverythingThreads · BUILD for Marketing
B2B · Sector Edition
For CMOs · Heads of Brand · Comms Directors · Content Leads · In-house Agency Heads

BUILD for Marketing & Comms

Train your marketing, comms, brand, and content staff to build their own AI tools — brand-voice enforcers, ASA/CAP claim-checkers, sentiment triagers, social-listening summarisers, briefing-pack drafters — using infrastructure your brand owns end-to-end. No vendor lock-in. No customer data leaving your perimeter. No AI making claims your CAP team has to defend at 9pm. Built by your team, owned by your brand, defensible to the ASA, the CMA, the CAP code, and the EU AI Act.

📘 28 segments · 4 weeks 👥 5–25 marketers per cohort 📣 ASA / CAP / CMA / EU AI Act aligned
Email hello@everythingthreads.com →
Section 1

The problem you actually have

Your content team is already drafting blog posts, ad copy, email subject lines, social captions, and influencer briefs in ChatGPT. Your brand team is using Claude to "translate" tone of voice. Your PR team is feeding press releases through Gemini to summarise coverage. None of this is happening with formal sign-off. None of it has an audit trail. None of it has been checked against the CAP code. You know it. They know you know it. The question isn't "should we let staff use AI?" — that decision was made for you 18 months ago. The question is whether the tools they're using respect your brand voice, refuse to invent claims about your products, and produce work you can defend if the ASA opens a case.

⚠ The current state for most marketing teams
What BUILD for Marketing does about it

BUILD takes any marketer — content writer, copywriter, PR manager, brand strategist, social lead, in-house designer-turned-content-lead — from "I've never written code" to a deployed AI tool running on infrastructure your brand controls. The course is the same proven 28 segments. The difference is the Marketing Build Kit: pre-tuned system prompts that enforce your brand voice rules, refuse to invent product claims, check against the ASA/CAP code language patterns, and capstone project templates that drop straight into Segments 12 and 15 to ship voice-aware, claim-aware, code-aware tools.

Section 2

What your team will actually build

Five concrete tools your marketing staff can build during the 4-week course. Each one is real, deployable, and addresses a workflow your team already does manually — usually under campaign deadline pressure, usually with the brand team and the legal team both wanting the same draft, usually at 8pm on a Thursday.

Tool 1
Brand Voice Enforcer
Built in Segments 11–12 · Powered by the Brand Voice system prompt below + your brand rules document
Paste any draft copy — blog post, email, ad headline, social caption, landing page hero. The tool checks the draft against your brand's actual voice rules (uploaded as a structured document), flags every phrase that drifts off-voice, and suggests on-voice alternatives. Used by content teams to keep work consistent across freelancers, juniors, and AI-assisted drafts. The brand voice rules live in your GitHub, not in a vendor's database.
Example input (brand voice = warm-but-not-chummy, British English, never uses "delight"):
"We're so delighted to delight you with our delightful new feature! It's a real game-changer."

Example output: 🔴 OFF-VOICE — 4 violations: (1) "delighted" / "delight" / "delightful" — banned per the brand voice rules; (2) "game-changer" — clichéd per the brand rules; (3) exclamation marks in body copy — only allowed in social captions per the brand rules. Suggested rewrite: "Our new feature is here. We've been working on it for months and we think you're going to like it." Note: the rewrite is a draft suggestion — your content lead has the final call.
Tool 2
ASA / CAP Code Claim Checker
Built in Segments 13–14 · Multi-model verification using the Claim Check prompt
Paste any draft ad copy, marketing email, landing page hero, or product page. The tool extracts every claim that could be substantiated or unsubstantiated under the CAP Code (numerical claims, comparative claims, superlatives, "up to" claims, environmental claims, health claims) and flags whether each one needs evidence on file before publication. Used by content and brand teams to triage drafts before they ever reach legal review.
Example output: 🟡 4 claims need substantiation: (1) "30% more efficient" — needs documented test methodology and reference dataset; (2) "Britain's favourite tea brand" — needs citation, dated, with methodology; "favourite" is a superlative requiring evidence under CAP 3.7; (3) "100% recyclable" — environmental claim requires substantiation under CAP 11; (4) "as recommended by 9 out of 10 dentists" — testimonial claim requires the survey methodology, sample size, and date. Recommended action: request substantiation file from product team before submitting to legal review.
Tool 3
Inbound Sentiment & Issue Triager
Built in Segment 14 · Multi-model orchestration via Promise.all()
Paste a batch of social media replies, customer reviews, NPS comments, or inbound email subject lines. The tool clusters them by sentiment, identifies emerging issues (a sudden uptick in complaints about a specific product feature, a brewing PR risk, a positive theme worth amplifying), and produces a structured triage summary suitable for the morning stand-up. Used by social and comms teams to spot a brewing crisis at 9am instead of finding out from the Times at 3pm.
Example output: 312 inbound messages over 24 hours. Sentiment distribution: 58% positive, 21% neutral, 21% negative. Theme spike: 47 complaints about the new checkout flow, all in the last 6 hours. Recommended action: ping product before the issue compounds. Positive theme worth amplifying: 23 unprompted posts praising the new packaging — could be repurposed as social proof.
Tool 4
Campaign Briefing Pack Drafter
Built in Segments 15–16 · Sector-specific system prompt + PWA on a phone
An installable web app for content and brand leads. They paste in or dictate the bullet points of a new campaign brief; the tool drafts the standard briefing pack structure (objective, audience, key message, RTBs, do's and don'ts, channel mix, KPIs) in your brand's house style, with placeholders flagged where the strategist needs to add the substantive content. Reduces the routine briefing-doc overhead so strategists spend more time on actual strategy.
Built-in safety: the tool refuses to invent campaign rationale, target audience claims, or KPIs the strategist didn't supply. Every numeric or attribution figure is flagged "[verify]" if not in the input. The "DRAFT — strategist must complete substantive sections before sharing" header is in every output.
Tool 5
Influencer/Affiliate Disclosure Checker (Browser Extension)
Built in Segments 17–19 · Chrome extension that reads the current page
A Chrome extension a brand team or social lead clicks while reviewing an influencer's draft post, an affiliate's landing page, or a partner's social caption. The extension checks for the required CAP/CMA disclosures (#ad, #gifted, paid partnership tags, affiliate disclosure, comparative-advertising honesty), flags any missing disclosures, and produces a one-line "compliant / not compliant" verdict. Replaces 20 minutes of disclosure review per post with 20 seconds of clicking.
Important: the extension is a triage tool, not a final compliance opinion. Every output ends with "Automated screening only. Final disclosure compliance is the brand's responsibility under the CAP code and the CMA's guidance on hidden advertising. This tool catches the obvious gaps; the brand team confirms."
Section 3

The Marketing Build Kit — copy these straight into Segment 15

Five ready-to-use system prompts your staff paste directly into BUILD's Segment 15 ("System Prompts — Controlling AI Behaviour") to transform the generic Text Analyser into a sector-specific marketing tool. Each one enforces brand voice, refuses to invent claims, and bakes in the CAP/CMA language patterns the ASA actually enforces.

🎙 Brand Voice Enforcer · System Prompt
For Segment 15
You are a brand voice assistant for a specific brand. The brand's voice rules are provided to you as a structured document. You check draft copy against the brand voice rules and flag every drift, then suggest on-voice alternatives. INPUT YOU WILL RECEIVE: - The brand voice rules document (banned words, required register, sentence-length norms, tone descriptors, examples of on-voice and off-voice copy) - A draft to check EXPERTISE: - Reading a brand voice guide and applying it consistently - The difference between "off-voice" and "wrong" — both need flagging, but differently - The difference between "voice" (how it sounds) and "tone" (what mood it carries) and "register" (formal/informal) - Common drift patterns when AI tools generate copy: generic American register, exclamation marks, "delight" overuse, marketing clichés ("game-changer", "best-in-class", "unleash") ABSOLUTE CONSTRAINTS: - The brand voice rules document is the ONLY source of truth. You do NOT bring in your own preferences about what good copy looks like. - You do NOT rewrite the draft fully — you flag drifts and suggest specific on-voice alternatives. The content lead makes the final call. - You ALWAYS quote the specific brand rule that each flagged phrase violates. - If the brand voice document doesn't cover a specific stylistic question, you say so rather than inventing a rule. - You do NOT make claims about the brand's products, services, or factual positioning. You only check voice. OUTPUT FORMAT: 1. OVERALL VERDICT: ✓ ON VOICE / 🟡 MOSTLY ON VOICE / 🔴 OFF VOICE 2. VIOLATIONS: bulleted list, each with the offending phrase, the brand rule it violates, and a suggested on-voice alternative 3. POSITIVE NOTES: any phrases that nail the brand voice particularly well (worth keeping) 4. UNCOVERED QUESTIONS: any stylistic decisions the brand voice document doesn't cover 5. MANDATORY FOOTER: "Voice review only. Substantive copy decisions, factual accuracy, and final approval remain with the content lead. The brand voice rules document is the source of truth."
📜 ASA / CAP Code Claim Checker · System Prompt
For Segment 15
You are a marketing claims compliance checker for a UK brand. You check draft ad copy, marketing emails, landing pages, and product pages against the CAP Code and the ASA's published rulings, and flag every claim that needs substantiation before publication. EXPERTISE: - The CAP Code (Committee of Advertising Practice) — particularly sections 3 (Misleading advertising), 11 (Environmental claims), 12 (Medicines, medical devices, health-related products and beauty), and 15 (Food, food supplements and associated health/nutrition claims) - The CAP Code's required substantiation standards - Common ASA-upheld complaint patterns: unsubstantiated superlatives, missing test methodologies, environmental greenwashing, misleading "up to" claims, unverified testimonials, undisclosed influencer ads - The CMA's enforcement of consumer protection law alongside the CAP Code - The Digital Markets, Competition and Consumers Act 2024 expectations on hidden advertising and fake reviews CONSTRAINTS: - You provide flagging and triage, NOT final compliance sign-off. Every output must end with a reminder that a qualified compliance/legal professional must review before publication. - You do NOT speculate about whether the substantiation actually exists. You flag what needs to be substantiated; the brand team confirms whether the evidence is on file. - You do NOT rewrite claims to make them defensible — you flag them so the human team decides whether to substantiate, soften, or remove. - You ALWAYS cite the specific CAP rule or ASA precedent for each flag. OUTPUT FORMAT: 1. CLAIMS IDENTIFIED: bulleted list, each with the verbatim text and a category (numerical / comparative / superlative / environmental / health / testimonial / "up to") 2. SUBSTANTIATION NEEDED: for each claim, what evidence the brand needs on file 3. CAP RULE: which CAP section applies and a one-line description 4. RISK LEVEL: 🟢 LOW / 🟡 MEDIUM / 🔴 HIGH (HIGH = previously upheld ASA complaint patterns) 5. MANDATORY FOOTER: "Claims triage only. Final compliance sign-off must be done by a qualified marketing-compliance professional. Substantiation evidence must be on file before publication. CAP Code: asa.org.uk/codes-and-rulings."
📊 Inbound Sentiment & Issue Triager · System Prompt
For Segment 15
You are an inbound communications triage assistant for a brand's social and customer-care team. You receive a batch of inbound messages (social replies, reviews, NPS comments, support tickets, inbound emails) and produce a structured triage summary for the morning stand-up. EXPERTISE: - Sentiment classification (positive, neutral, negative) at scale - Issue clustering and theme detection - Spotting sudden anomalies (unusual spike in complaints about a specific feature, surge in mentions of a competitor, brewing PR risk) - The difference between "noise" and "signal" in inbound data ABSOLUTE CONSTRAINTS: - You do NOT identify individual customers. Names, handles, and email addresses are stripped. - You do NOT make accusations about specific products being defective — you flag patterns for the human team to investigate. - You preserve outliers and unusual messages individually rather than averaging them out. - You do NOT respond to any customer message — that's the customer-care team's job. OUTPUT FORMAT: 1. METADATA: total messages, time window, channels covered 2. SENTIMENT DISTRIBUTION: positive %, neutral %, negative % 3. PRIMARY THEMES (clustered): each with count and a representative anonymised example 4. SPIKES & ANOMALIES: any sudden upticks in a specific topic vs the previous comparable window 5. POSITIVE THEMES WORTH AMPLIFYING: positive clusters that could become social proof 6. INDIVIDUAL OUTLIERS: messages that don't fit any cluster but are worth a human read 7. RECOMMENDED ACTIONS: bulleted list for the morning stand-up 8. MANDATORY FOOTER: "Triage only. Customer responses, brand decisions, and PR judgements remain with the human team. Sentiment classification is statistical, not editorial."
📋 Campaign Briefing Pack Drafter · System Prompt
For Segment 15 + PWA in 17–19
You are a campaign briefing pack drafting assistant. You receive bullet points dictated by a brand strategist or content lead and produce a draft briefing pack in the standard structure used by the brand. EXPERTISE: - Standard campaign briefing structure: Objective, Audience, Key Message, Reasons-to-Believe (RTBs), Tone & Voice, Do's and Don'ts, Channel Mix, KPIs, Timing, Budget envelope (if stated) - The difference between a brief and a creative concept (this drafts briefs, not concepts) - How to flag unanswered questions for the strategist to fill in ABSOLUTE CONSTRAINTS: - You NEVER invent target audience claims, KPIs, market sizes, or competitor positioning the strategist didn't supply. - Every numeric figure or audience claim is either taken verbatim from the input OR marked "[strategist to confirm]". - You NEVER invent the campaign rationale or strategic positioning — the strategist supplies it, you structure it. - You ALWAYS produce a draft suitable for the strategist to refine and approve, not a final brief. - You ALWAYS include the "DRAFT — strategist must complete substantive sections before sharing" header. OUTPUT FORMAT: 1. CAMPAIGN NAME (as stated) 2. OBJECTIVE: one sentence — what does this campaign need to achieve? 3. AUDIENCE: as stated, structured as primary / secondary 4. KEY MESSAGE: one sentence — what's the single thing the audience should remember? 5. REASONS-TO-BELIEVE: bulleted list of 3-5 RTBs (drawn from input) 6. TONE & VOICE: as stated by the strategist 7. DO'S AND DON'TS: bulleted lists drawn from input 8. CHANNEL MIX: as stated 9. KPIs: as stated, marked [strategist to confirm baseline] if not specified 10. TIMING: as stated 11. UNANSWERED QUESTIONS: bulleted list of things the strategist needs to fill in before this brief is usable 12. MANDATORY HEADER: "DRAFT — strategist must complete substantive sections, verify all figures, and approve before sharing with creative team."
📢 Influencer / Affiliate Disclosure Checker · System Prompt
For Segment 15 + Browser Extension in 17–19
You are an influencer and affiliate disclosure compliance assistant for a UK brand. You receive text extracted from an influencer post, affiliate landing page, or partner social caption, and check whether the required disclosures are present and prominent enough under the CAP Code and the CMA's guidance on hidden advertising. EXPERTISE: - The CAP Code's rules on identification of marketing communications (Section 2) - The ASA's published guidance on influencer marketing - The CMA's "Influencer marketing: a guide to making clear when posts are ads" (and the recent enforcement priorities) - Common drift patterns: disclosure buried in hashtags at the end, "gifted" without "ad", #spon used incorrectly, affiliate links without disclosure - The Digital Markets, Competition and Consumers Act 2024's strengthened consumer protection provisions CONSTRAINTS: - You triage; you do NOT make a final compliance verdict. The brand team confirms. - You do NOT speculate about whether the influencer was paid — you check what the post itself shows. - You ALWAYS check for disclosure prominence (top of post vs buried in hashtag tail), not just presence. - You do NOT make legal threats or accusations against the influencer or affiliate — you flag what's missing for the brand team to address. OUTPUT FORMAT: 1. POST TYPE: as identified (Instagram caption, TikTok description, blog post, landing page, etc.) 2. DISCLOSURES PRESENT: bulleted list of disclosures found, with their position (top / middle / hashtag tail) 3. DISCLOSURES MISSING: bulleted list of disclosures expected but not present (or not prominent enough) 4. PROMINENCE: ✓ PROMINENT / 🟡 BURIED / 🔴 ABSENT 5. RISK LEVEL: 🟢 LOW / 🟡 MEDIUM / 🔴 HIGH 6. SUGGESTED REMEDIATION: specific suggestions for the brand to send back to the influencer/affiliate 7. MANDATORY FOOTER: "Disclosure screening only. Final compliance is the brand's responsibility under the CAP Code and CMA guidance. Hidden advertising remains a CMA enforcement priority. asa.org.uk + gov.uk/cma."
Section 4

The 70/30 model — what's generic, what's marketing-specific

BUILD for Marketing isn't a separate course. It's the existing 28-segment BUILD course (the same one any other professional takes), plus the Marketing Build Kit your staff drop in at three specific points. This is intentional and matters for procurement and brand-governance reasons.

70% — the BUILD course core (unchanged)

The technical pipeline your staff learn is identical regardless of sector: HTML / CSS / JavaScript frontends, Cloudflare Workers as a secure proxy, the Anthropic API for AI calls, GitHub for version control, Netlify for hosting. This is the standardised, defensible infrastructure layer that the brand controls end-to-end. Same code. Same architecture. Same security posture. Same audit trail. Reviewable by your IT team once.

30% — the marketing customisation

The marketing-specific layer is the system prompts (Segment 15), the brand voice rules document (uploaded as a sector data file), the use case examples (Segments 12 and 14), and the capstone project briefs (Segment 28). The Build Kit also includes marketing-tuned versions of: the Multi-Model Compare tool (Segment 13) for cross-checking claims, the System Prompt framework (Segment 15) for brand-voice-aware language, and the Final Project rubric for marketing-relevant capstone projects.

Why this matters for brand governance

Because the underlying technical architecture is identical to every other BUILD cohort, your IT and brand-governance team can review and approve it once, and that approval covers every staff member who ever takes the course. The marketing customisation is purely at the prompt and brand-rules layer — which is where your existing creative judgement lives. IT reviews the architecture once. Brand reviews the voice rules. Compliance reviews the claim-check prompts. Everyone stays in their lane.

Section 5

Compliance & regulatory alignment

BUILD for Marketing is positioned to help your brand meet (and document) compliance with multiple converging requirements.

CAP Code & ASA
The CAP Code applies to all marketing communications regardless of whether a human or an AI wrote the words. Tools your team builds with BUILD bake the CAP rules into the prompts at draft time, so claims that need substantiation get flagged before they reach the legal review queue. Materially fewer ASA escalations.
CMA & Consumer Protection
The Digital Markets, Competition and Consumers Act 2024 strengthens the CMA's powers on hidden advertising, fake reviews, and subscription traps. The Influencer Disclosure Checker and the Inbound Sentiment Triager produce auditable artefacts that demonstrate the brand is screening for these issues actively, not reactively.
EU AI Act (Article 4)
From August 2026, organisations using AI systems must demonstrate "a sufficient level of AI literacy" among staff who operate them. BUILD produces a per-staff-member artefact (live tool, GitHub portfolio, README documentation) that evidences exactly this literacy.
UK GDPR & customer data
Tools built with BUILD use the Cloudflare Worker proxy pattern: API keys never leave the server, requests are routed through infrastructure your brand owns, and you can deploy with regional pinning to keep customer data inside the UK or EEA. Materially better than staff pasting NPS surveys into ChatGPT.
Equality Act 2010
Brands are bound by the Equality Act 2010 in their advertising and customer communications — the CAP Code's harm-and-offence rules and the ASA's enforcement against discriminatory ads sit alongside the Equality Act provisions. Tools your team build with BUILD are auditable end to end, which gives the brand a stronger answer if a campaign is ever scrutinised under the protected-characteristic provisions.
⚖ A specific note on ASA upheld AI complaints

The ASA has already upheld complaints against ads that contained AI-hallucinated claims. The brand was held responsible regardless of who or what produced the words. "Our copywriter used ChatGPT" is not a defence. The defensible position is "we have prompt-level controls that flag claims requiring substantiation, and our brand team reviewed before publication." BUILD-graduated brands have those artefacts.

Section 6

Pricing — for marketing teams

Three tiers based on cohort size. All prices are the brand-wide commercial rate, not per-seat consumer pricing. Includes the full BUILD course, the Marketing Build Kit, the Manager Pack, brand voice rules template, and email support across the rollout.

Pilot Cohort
£4,500 / cohort
Up to 10 marketers
  • Full 28-segment BUILD course
  • Marketing Build Kit (5 system prompts)
  • Brand voice rules template
  • Manager Pack + Capstone rubric
  • Email support across the 4 weeks
  • One IT whitelist consultation
Department Rollout
£9,500 / cohort
Up to 25 marketers
  • Everything in Pilot
  • Buddy pairing + cohort kickoff call
  • Mid-point manager check-in (60 min)
  • Capstone showcase facilitated by ET
  • Brand-specific voice rules workshop
  • One brand-specific prompt customisation
Brand-Wide
From £18,000
25–100+ marketers across multiple brands or markets
  • Everything in Department Rollout
  • Multiple parallel cohorts
  • Train-the-trainer for in-house champion
  • Custom Marketing Build Kit additions
  • White-label option for internal LMS
  • Quarterly check-ins for 12 months

All prices ex-VAT. Procurement-friendly invoicing available. Multi-brand groups (holding companies with multiple brand teams) get a portfolio rate — email for the figure. hello@everythingthreads.com

Section 7

FAQ — for marketing leadership

Can our content team actually do this? They're writers, not developers.
That's exactly who BUILD is designed for. The course starts at "what is a terminal" and finishes with a deployed, working AI tool. Across hundreds of non-developer students — including content writers, brand strategists, and PR managers — completion rates for cohorts with manager air cover sit in the 80%+ range. Marketers who finish BUILD become the in-house AI champions for everyone else.
How is this different from the MarTech-AI vendors we already use?
Three differences. First, ownership: tools built with BUILD belong to your brand, run on infrastructure you control, and your brand voice rules live in your own GitHub instead of a vendor's database. Second, cost: vendor seat licences compound forever; BUILD is a one-off cohort cost plus ~£20/month in compute. Third, oversight: your IT and brand teams can review the actual code and the actual prompts — something the major MarTech-AI vendors don't offer at any price.
Will the brand voice tool actually capture our voice?
It captures the voice that your brand voice document specifies. The tool is only as good as the brand rules you give it — vague voice docs produce vague enforcement, specific voice docs produce specific enforcement. The Build Kit includes a brand voice rules template that asks the right questions: banned words, required register, sentence-length norms, do's and don'ts, on-voice and off-voice examples. Filling in the template is part of the workshop in the Department Rollout tier.
What about customer data and GDPR?
The Cloudflare Worker proxy pattern (taught in Segment 11) keeps API keys server-side and routes requests through infrastructure your brand controls. With regional pinning, you can keep customer data inside the UK or EEA. Critically, BUILD teaches marketers to think about data flow as a first-class concern — most brands find their staff understand customer-data risks materially better AFTER BUILD than before.
Who owns the tools the marketers build?
Your brand. The code lives in your brand's GitHub. The infrastructure is provisioned in your brand's accounts. The brand voice rules live in your repo. Standard work-for-hire applies as it would for any other internal asset.
What if a marketer builds something that produces an ASA-non-compliant claim?
Every system prompt in the Marketing Build Kit explicitly flags claims that need substantiation. The Claim Checker prompt refuses to write substantiated claims that aren't in the input — it only checks them. Segment 24 (Testing Your AI Tools) walks through edge case handling. The standard footer on every output is "compliance must be reviewed by a qualified professional before publication." The tools are designed to support brand judgement, not replace it.
How long does the rollout take from kick-off to first cohort?
Typically 2–3 weeks from contract signature to Day 1 of the cohort. Most of that time is brand voice rules document preparation and IT whitelisting. Once the course starts, it runs 4 weeks. Total elapsed time from "we want this" to "we have marketers with deployed tools" is around 7 weeks.
Can we run this across multiple brands in our holding company?
Yes — we offer portfolio rates for multi-brand groups. Each brand brings its own voice rules document, runs the same course content, and ends up with brand-specific tools that all share the same auditable infrastructure. Email us with the number of brands and total marketer count for a portfolio quote.
Ready to talk?

If you're a CMO, head of brand, comms director, or in-house agency head and you want to bring BUILD for Marketing to your team, the next step is a 30-minute discovery call. We'll walk through your current AI use, your brand voice constraints, and which cohort tier makes sense.

Email hello@everythingthreads.com →

EverythingThreads is contact-by-email only. We reply within 2 working days. For urgent matters during a paid rollout, mark the email subject "URGENT" and we'll prioritise.

Sister sector editions: Legal · Finance · Education · Healthcare · Governance · HR & People Ops · all sectors hub