Andrew Thorn

🔥 New DemandSphere feature: GA4 LLM Traffic Analytics We’re happy to announce the release of our new LLM traffic analytics feature, powered by our integration with GA4. We've had rich integration features for content performance analytics with GA4 for a long time, especially useful for a cross-channel view. This adds referrer level data to accurately capture LLM traffic, which doesn't show up well in cross-channel reports. This feature provides insight into content performance, attribution, and other traffic insights, broken out by each model, to your websites. This bridges the gap between understanding LLM visibility and human traffic and engagement on websites. Blog post with announcement in comments, check it out!

59

5 Comments

🇦🇺 Scaling an SEO Agency: Lessons from the Australian SEO Scene In addition to hearing the latest trends from SEO experts in Bangkok, I’ve been learning from Jason Mun, an experienced SEO now leads search at Overdose. Here are two takeaways I’m applying: 1. Think enterprise-level processes early: Even when your agency is small, build systems and reporting frameworks that can scale with you. 2. Find the right partner: This advice comes up often, and it applies to SEO too. If you’re strong in selling SEO strategy and pitching decks, team up with someone who excels on the technical side. If you’re hands-on with every SEO area, consider a partner who can help you grow the agency in a systematic, sustainable way. #SEO #Backlink #LinkBuilding #SEOAgency #AustraliaSEO #DigitalMarketing

44

2 Comments

The worst-kept secret in ecommerce is out in the open. OpenAI launched checkout, initially partnering with Shopify, Etsy and Stripe to enable direct purchases from ChatGPT. 🛍️ Shopify merchants will be able to sell directly through ChatGPT "very very soon". 🥣 US users will be able to buy from US Etsy sellers through ChatGPT as of yesterday. Etsy shares popped +16% on the announcement . 🌐 OpenAI also partnered with Stripe to build the Agentic Commerce Protocol (ACP) to allow others to integrate. Last night I sat down with Amy O'Brien from Vogue Business to discuss what this means for ecommerce brands: https://lnkd.in/epz9FvN9 You can also see the official announcement from OpenAI here: https://lnkd.in/eg3kp_Pv; from Shopify here: https://lnkd.in/eBf88DVs; from Etsy here: https://lnkd.in/e9huvyiS; and Stripe here: https://lnkd.in/eQm8QTEd. p.s. give Azoma a follow, very exciting announcement coming shortly 👀

110

28 Comments

The real reason your customers aren't buying ↳ Hint: Your creative isn't the problem Most DTC brands are burning ad spend ignoring what customers ACTUALLY think. I've analyzed over 50 ad accounts this year, and found these 5 customer objections are killing your sales: 1. Lack of Need     "I don't need this" → Identify your customer's EXACT pain point → Show the desired end state they desperately want → Position your product as the only bridge between pain and pleasure 2. Lack of Understanding     "I don't get it" → Make your product dead simple to understand → Show, don't tell - visuals convert better than words → Front-load benefits in the first 3 seconds of your ad 3. Lack of Trust     "This won't work for me" → Use real customer testimonial mashups → Run ads through whitelisted creator accounts → Include third-party validation (press, awards, certifications) 4. Lack of Budget     "It's too expensive" → Focus on ROI, not just price → Test different offers to find price-value sweet spot → Use bundles to increase AOV while adding perceived value 5. Lack of Urgency     "I'll buy later" → Create time-sensitive offers that drive action → Demonstrate the cost of waiting → Use scarcity when authentic (limited stock, time-bound offers) Address these 5 objections in your ad creative, and your conversion rates will double. Your competitors are ignoring these. You shouldn't.

45

42 Comments

A stupidly simple (and extremely powerful) ICP-fit AI flow Most teams overcomplicate ICP validation. You don’t need a dashboard. You need a yes/no. Here’s the flow we use to turn any website into FIT / NOT FIT in seconds. – WHY THIS MATTERS – → If you already check ICP fit now: this saves ~10 hours/week you currently burn on manual research. → If you don’t check ICP fit now: it makes your outreach way more relevant and typically 2× your reply rate. – THE FLOW – STEP 1: Feed the agent a website URL (or a column of URLs). STEP 2: Agent scrapes the site + does light web research ↳ MCP tools like Firecrawl, Exa, Jina (any equivalents work). STEP 3: It extracts the signals you care about ↳ product category, ICP geo/size, tech stack, hiring, use cases. STEP 4: Runs a single validation prompt against your ICP rules. ↳ Returns a structured output only: FIT or NOT FIT (+ why). STEP 5: Auto-sync results to Clay/CRM ↳ Only FIT accounts move to enrichment & sequence. – WHAT YOU GET – → Clarity: a binary decision you can trust. → Speed: lists go from “raw” to “ready” before your first coffee. → Lift: more relevance → more replies → more pipeline. Want the exact prompt, validation rubric, and a plug-and-play workflow? Comment FLOW and I’ll send you the setup. Follow me for more GTM engineering plays that actually move revenue.

36

33 Comments

Where do 82% of AI search queries actually go? Straight into ChatGPT. That’s 700M weekly users asking questions your brand isn’t answering. Most companies still optimize for Google. But AI engines follow a different logic. The shift: SEO → AEO SEO focuses on  • keyword density  • backlinks  • domain authority AEO focuses on  • structured content  • Q&A style answers  • citations from trusted platforms Who does ChatGPT trust?  • Wikipedia (47.9%)  • Reddit (11.3%)  • Forbes (6.8%) One solid Reddit thread can outperform an entire blog section. Outdated vs effective ❌ Long walls of text → ✅ Short, FAQ-style answers ❌ Stuffed keywords → ✅ Natural, human questions ❌ Meta tags → ✅ JSON-LD schema Brands that already test AEO report:  • Tiny traffic share  • But outsized sales impact  • Because AI traffic converts How to start: 1. Restructure pages for clarity 2. Add FAQs across touchpoints 3. Post value on Reddit & Quora 4. Frame headers as questions 5. Implement schema markup 6. Monitor AI mentions Google rankings bring clicks. AI visibility brings customers. The difference is massive.

42

13 Comments

“We increased budget… but conversions dropped?” Yep. Seen this one a lot. More spend ≠ more leads - unless the account is ready for it. Here’s what usually goes wrong when scaling too early: ❌ Campaign structure is messy → Google doesn’t know what to optimise ❌ Bidding too aggressively → Blows budget on low-intent clicks ❌ No conversion tracking (or bad data) → Optimising blind ❌ Weak landing page → More traffic, same poor experience ❌ Budget spread too thin → Not enough data to exit learning phase How to scale without killing performance: ✅ Fix tracking first - know what’s actually working ✅ Consolidate campaigns for stronger signals ✅ Let smart bidding gather data before increasing spend ✅ Only scale what’s proven - don’t throw budget at guesswork ✅ Monitor lead quality - not just volume Scaling Google Ads isn’t about throwing more money in. It’s about earning the right to spend more. #GoogleAds #PaidSearch #PPC

401

66 Comments

Traditional email outreach is broken. Generic templates get deleted instantly. But AI-powered automation is rewriting the rules completely. Here's how smart companies are dominating inboxes: 1. Master Multi-Channel Outreach Platforms       [Reply.io](http://reply.io/) sequences LinkedIn + email with AI personalization   Brevo triggers campaigns across email, SMS, and CRM automatically   ActiveCampaign tracks behavior and adapts messaging in real-time     2. Personalize at Scale Without Generic Templates Jasper AI writes custom copy based on prospect data. Skrapp.io enriches contacts with firmographic intelligence. MailMaestro matches your writing style while adding context. 3. Automate Smart Sequencing and Follow-ups Mailerlite recommends optimal send times with AI. Superhuman generates contextual follow-up replies automatically. EmailTree.ai manages sequences and adjusts based on engagement. 4. Enrich Leads for Laser-Focused Targeting Skrapp.io finds verified emails and enriches contact profiles. Clearbit provides real-time firmographic data for better personalization. ZoomInfo delivers extensive B2B data with AI-enhanced accuracy. The companies still sending generic email blasts are getting left behind. Meanwhile, AI-powered outreach is generating qualified leads on autopilot. Your choice: evolve your email strategy or watch competitors steal your deals. P.S. Want to learn more about AI? 1. Scroll to the top 2. Click "Visit my website" 3. Sign-up for our free newsletter

135

132 Comments

Most people track SEO through Google. I wanted to track AI visibility — how often we appear in results from: → ChatGPT → Claude → Gemini → Perplexity Because in 12 months, these models will be the new search engines. So, I built an AI SEO Agent using n8n: 1️⃣ Our team adds key search queries into an n8n Table (like “top B2B social listening tools”). 2️⃣ The master agent hits 3 sub-agents (Gemini, Claude, Perplexity). 3️⃣ Each one scrapes the citations the model used to answer that question. 4️⃣ Results get logged automatically into an AI SEO Results Table — with the query, citations, mentioned companies, and a “Was Trigify mentioned?” flag. The outcome? Claude listed Trigify.io for “top 10 B2B social signal tools.” Not bad. But the real gold is in the patterns: → Which sources the LLMs trust (PR + backlink targets). → Which competitors keep showing up. → Which topics we’re not visible in (yet). This is what modern SEO looks like. It’s not backlinks. It’s LLM relevance. If you want the n8n workflow or the table schema I used, comment “AI SEO Agent” below and I’ll drop the setup.

224

609 Comments