If your robots.txt blocks GPTBot, ClaudeBot, ChatGPT-User, or Google-Extended, AI literally cannot read your site. This single change is the prerequisite for everything else. Many Indonesian business sites ship with a default block — check yours first.Jika robots.txt kamu memblokir GPTBot, ClaudeBot, ChatGPT-User, atau Google-Extended, AI benar-benar tidak bisa membaca situs kamu. Perubahan ini adalah prasyarat untuk semua hal lain. Banyak situs bisnis kecil Indonesia yang punya blokir default — cek dulu milik kamu.
Add explicit Allow rules for the 4 major AI crawlers in your robots.txt file.Tambahkan aturan Allow eksplisit untuk 4 crawler AI utama di file robots.txt kamu. · ~15 min
AI crawlers check robots.txt before fetching. Blocking any of them makes your site unreadable by that AI — the prerequisite for everything else. (mechanism, not an outcome claim: see OpenAI docs, llmstxt.org)Crawler AI cek robots.txt sebelum fetch. Blokir salah satu membuat situs kamu tidak bisa dibaca AI tersebut — prasyarat untuk semua hal lain. (mekanisme, bukan klaim hasil: lihat docs OpenAI, llmstxt.org)
User-agent: GPTBot
Allow: /
User-agent: ChatGPT-User
Allow: /
User-agent: ClaudeBot
Allow: /
User-agent: Google-Extended
Allow: /
User-agent: *
Disallow: /admin
Disallow: /api
Disallow: /private
Sitemap: https://yourdomain.com/sitemap.xml
Organization + LocalBusiness schema gives AI a clean machine-readable statement of who you are, where you are, and what you do. Without it, AI has to guess from HTML prose. JSON-LD in your <head> with name, address, phone, hours, logo, and sameAs links to social profiles.Schema Organization + LocalBusiness memberi AI pernyataan who-you-are, where-you-are, dan what-you-do yang bisa dibaca mesin. Tanpa itu, AI harus menebak dari prosa HTML. JSON-LD di <head> dengan nama, alamat, telepon, jam buka, logo, dan link sameAs ke profil sosial.
Paste the JSON-LD below into the <head> of your homepage. Replace every placeholder with your real data.Tempel JSON-LD di bawah ini ke <head> homepage kamu. Ganti setiap placeholder dengan data asli kamu. · ~20 min
Organization + LocalBusiness schema gives AI a clean machine-readable statement of who you are, where you are, and what you do — without it, AI has to guess from HTML prose. Google’s E-E-A-T guidance and the schema.org documentation both endorse this approach. (mechanism, not an outcome claim)Schema Organization + LocalBusiness memberi AI pernyataan who-you-are, where-you-are, dan what-you-do yang bisa dibaca mesin — tanpa itu, AI harus menebak dari prosa HTML. Panduan E-E-A-T Google dan dokumentasi schema.org keduanya mendukung pendekatan ini. (mekanisme, bukan klaim hasil)
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "LocalBusiness",
"@id": "https://yourdomain.com/#business",
"name": "Nama Bisnis kamu",
"url": "https://yourdomain.com",
"logo": "https://yourdomain.com/logo.png",
"image": "https://yourdomain.com/store-front.jpg",
"description": "Deskripsi 1-2 kalimat yang fokus pada kategori dan kota.",
"telephone": "+62-21-1234567",
"priceRange": "Rp 50.000 - Rp 500.000",
"address": {
"@type": "PostalAddress",
"streetAddress": "Jl. Alamat Lengkap No. 123",
"addressLocality": "Jakarta Selatan",
"addressRegion": "DKI Jakarta",
"postalCode": "12345",
"addressCountry": "ID"
},
"geo": {
"@type": "GeoCoordinates",
"latitude": -6.2615,
"longitude": 106.8107
},
"openingHoursSpecification": [
{
"@type": "OpeningHoursSpecification",
"dayOfWeek": ["Monday", "Tuesday", "Wednesday", "Thursday", "Friday", "Saturday"],
"opens": "09:00",
"closes": "21:00"
}
],
"sameAs": [
"https://www.instagram.com/yourbrand",
"https://www.facebook.com/yourbrand",
"https://www.tiktok.com/@yourbrand",
"https://tokopedia.com/yourstore",
"https://shopee.co.id/yourstore"
]
}
</script>
GBP is the #1 off-site lever for Indonesian local SEO. SGE cites GBP data verbatim — hours, address, photos, attributes, posts. Verify by postcard/phone, fill EVERY field, and post once a week.GBP adalah pengungkit off-site #1 untuk SEO lokal Indonesia. SGE mengutip data GBP secara verbatim — jam, alamat, foto, atribut, posting. Verifikasi via kartu/telepon, isi SEMUA field, posting seminggu sekali.
Go to business.google.com → claim → verify → fill every field → post once a week for the next 4 weeks.Buka business.google.com → klaim → verifikasi → isi semua field → posting seminggu sekali selama 4 minggu ke depan. · ~30 min + 10 min/week
Indonesian AI search (Google SGE for google.co.id) heavily weights GBP citations. Profiles with weekly posts and 5+ photos rank in the local 3-pack 4-5x more often. (source: BrightEdge 2026 + Google Quality Rater Guidelines Nov 2024 update)AI search Indonesia (Google SGE untuk google.co.id) sangat membobotkan sitasi GBP. Profil dengan posting mingguan dan 5+ foto ranking di local 3-pack 4-5x lebih sering. (sumber: BrightEdge 2026 + Google Quality Rater Guidelines update Nov 2024)
Google Business Profile checklist:
─────────────────────────────
✓ Claim verified (postcard or phone)
✓ Business name EXACTLY as on your website
✓ Address EXACTLY as on your website (same formatting)
✓ Phone EXACTLY as on your website
✓ Website URL
✓ Hours (including special hours for holidays)
✓ Category (primary + 5-9 secondary)
✓ Attributes (women-owned, free Wi-Fi, delivery, etc.)
✓ Service area (cities you serve)
✓ Opening date
✓ Description (use your Organization schema description)
✓ 5+ photos (storefront, products, team)
✓ Logo + cover photo
✓ Weekly post (update, offer, or event)
✓ Respond to every review within 24 hours
✓ Q&A: seed 5 questions and answer them yourself
FAQPage schema is one of the strongest signal-to-noise markers you can add. It explicitly tells AI: "here are the questions this page answers." On every key product/service page, list the 5-8 most common customer questions, write direct answers, and add the JSON-LD below.Schema FAQPage adalah salah satu penanda signal-to-noise terkuat yang bisa kamu tambahkan. Secara eksplisit memberi tahu AI: "ini pertanyaan yang dijawab halaman ini." Di setiap halaman produk/layanan utama, daftar 5-8 pertanyaan yang paling sering ditanya pelanggan, tulis jawaban langsung, dan tambahkan JSON-LD di bawah.
On every key product/service page, list the 5-8 most common customer questions, write direct answers, and add the JSON-LD below.Di setiap halaman produk/layanan utama, daftar 5-8 pertanyaan yang paling sering ditanya pelanggan, tulis jawaban langsung, dan tambahkan JSON-LD di bawah. · ~45 min per page
FAQPage schema lets AI assistants skip the "guess what this page is about" step and extract a verbatim answer for short-form replies. It is one of the few signals the schema.org documentation explicitly recommends for Q&A content. (mechanism, not an outcome claim)Schema FAQPage memungkinkan asisten AI melewati langkah "tebak halaman ini tentang apa" dan mengekstrak jawaban verbatim untuk balasan singkat. Ini salah satu dari sedikit sinyal yang dokumentasi schema.org secara eksplisit rekomendasikan untuk konten Q&A. (mekanisme, bukan klaim hasil)
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "FAQPage",
"mainEntity": [
{
"@type": "Question",
"name": "Berapa harga [produk kamu]?",
"acceptedAnswer": {
"@type": "Answer",
"text": "Mulai dari Rp 150.000. Tersedia dalam 3 ukuran: S (Rp 150.000), M (Rp 175.000), L (Rp 200.000). Bisa custom size untuk pesanan 10+ pcs dengan biaya tambahan Rp 25.000 per item."
}
},
{
"@type": "Question",
"name": "Apakah [produk kamu] ada garansi?",
"acceptedAnswer": {
"@type": "Answer",
"text": "Ya, garansi 30 hari uang kembali tanpa syarat untuk cacat produksi. Hubungi WhatsApp +62-812-xxxx-xxxx dengan nomor order untuk klaim. Garansi tidak termasuk kerusakan dari pemakaian di luar normal."
}
},
{
"@type": "Question",
"name": "Dimana toko fisik [bisnis kamu]?",
"acceptedAnswer": {
"@type": "Answer",
"text": "Jl. Nama Jalan No. 123, Jakarta Selatan. Buka Senin-Sabtu 09.00-21.00 WIB, Minggu 10.00-18.00. Parkir gratis untuk 10 mobil pertama. Lokasi 5 menit dari pintu tol XYZ."
}
},
{
"@type": "Question",
"name": "Bisa kirim ke luar Jabodetabek?",
"acceptedAnswer": {
"@type": "Answer",
"text": "Bisa, pakai JNE, J&T, SiCepat, atau AnterAja. Ongkos kirim dihitung berat (1 kg = Rp 15.000-25.000 tergantung kota). Order di atas Rp 500.000 gratis ongkir ke seluruh Indonesia."
}
},
{
"@type": "Question",
"name": "Berapa lama proses produksi?",
"acceptedAnswer": {
"@type": "Answer",
"text": "Ready stock: kirim hari yang sama kalau order sebelum jam 14.00. Custom: 3-5 hari kerja. Bulk order 50+ pcs: 7-14 hari kerja. Express 1 hari: tambah Rp 50.000."
}
}
]
}
</script>
Google SGE extracts passages of 134-167 words. Write 3-5 long-form pages answering the questions your customers actually ask: origin story, how-it-works, pricing, FAQ hub, "why us vs alternatives". Each page 800+ words, each section self-contained.Google SGE mengekstrak bagian 134-167 kata. Tulis 3-5 halaman panjang yang menjawab pertanyaan pelanggan kamu sebenarnya: kisah asal, cara kerja, harga, hub FAQ, "kenapa kita vs alternatif". Tiap halaman 800+ kata, tiap bagian self-contained.
Pick the 5 questions your customers ask most. Write 1 page per question, 800+ words, with a clear direct answer in the first 100 words.Pilih 5 pertanyaan yang paling sering ditanya pelanggan. Tulis 1 halaman per pertanyaan, 800+ kata, dengan jawaban langsung jelas di 100 kata pertama. · ~4-6 hours total
Passages of 100-200 words give AI a clean extraction target. Shorter passages may lack context; longer ones bury the answer. The 134-167 word range is a widely-cited target because it is large enough to be self-contained and small enough for AI to lift cleanly. (mechanism, not an outcome claim)Bagian 100-200 kata memberi AI target ekstraksi yang bersih. Bagian yang terlalu pendek mungkin kurang konteks; yang lebih panjang mengubur jawaban. Kisaran 134-167 kata adalah target yang sering dikutip karena cukup besar untuk self-contained dan cukup kecil untuk diangkat AI secara bersih. (mekanisme, bukan klaim hasil)
## Template: 1 deep page per customer question
Page title: "[Pertanyaan yang sering ditanya pelanggan kamu]?"
Structure:
1. **Direct answer** (134-167 words, top of page)
- First sentence = the answer
- Next 100-150 words = context, examples, specifics
2. **Why this matters** (134-167 words)
- Common misconception
- What most people get wrong
3. **Step-by-step** (numbered list, 3-7 steps)
4. **Comparison** (table or list)
5. **FAQ** (3-5 Q&As, also add FAQPage schema)
6. **About the author / About the business** (E-E-A-T signal)
- Photo, name, credentials, LinkedIn
7. **Last updated:** [date]
Example pages to write this week:
- "Berapa biaya [layanan kamu] di [kota]? (Update 2026)"
- "Bagaimana cara [proses utama kamu]?"
- "[Produk kamu] vs [alternatif] — mana yang lebih cocok?"
- "5 kesalahan umum saat [masalah customer kamu] + solusinya"
- "FAQ [bisnis kamu] — 20 pertanyaan terlengkap"
Indonesian AI search (Google SGE, ChatGPT) heavily weights marketplace citations. A complete Tokopedia + Shopee listing with consistent NAP (Name, Address, Phone) and original product photos is a free, durable citation source.AI search Indonesia (Google SGE, ChatGPT) sangat membobotkan sitasi marketplace. Listing Tokopedia + Shopee lengkap dengan NAP (Nama, Alamat, Telepon) konsisten dan foto produk asli adalah sumber sitasi gratis dan tahan lama.
Claim or create your Tokopedia + Shopee store today. Use the EXACT same business name, address, and phone as on your website. Add 10+ products with full descriptions + original photos.Klaim atau buat toko Tokopedia + Shopee kamu hari ini. Pakai nama bisnis, alamat, dan telepon YANG SAMA persis seperti di website. Tambahkan 10+ produk dengan deskripsi lengkap + foto asli. · ~2-3 hours
Tokopedia and Shopee rank as high-authority domains for Indonesian queries. Having a consistent business listing there gives AI a third-party source that confirms your name, address, and category — independent of your own website. (mechanism, not an outcome claim)Tokopedia dan Shopee berperingkat sebagai domain otoritas tinggi untuk query Indonesia. Punya listing bisnis konsisten di sana memberi AI sumber pihak ketiga yang mengonfirmasi nama, alamat, dan kategori kamu — independen dari website kamu sendiri. (mekanisme, bukan klaim hasil)
Marketplace NAP checklist:
─────────────────────────
Your business name (display): "Kopi Kenangan"
Your business name (legal): "PT Kopi Kenangan Indonesia"
Use the EXACT same name, address, phone on:
- your website
- your GBP
- your Tokopedia store
- your Shopee store
- your Instagram bio
- your Facebook page
- your TikTok bio
- your LinkedIn company page
Even small differences ("Kopi Kenangan" vs "Kopi Kenangan Indonesia")
can break Google's entity matching.
Storefront essentials:
- Logo (same as website)
- Cover photo (same brand colors)
- Description (200+ chars, include city + category)
- 10+ products with REAL photos (not stock)
- 1-2 line product descriptions with specifics
- Same product categories as your website
- Active customer service (reply within 1 hour)
E-E-A-T (Experience, Expertise, Authority, Trust) is the binary gate in Google SGE Stage 3. Add a real author byline + bio to every blog post. If your business is notable, create a Wikidata entry — that becomes the seed for Wikipedia, Google Knowledge Panel, and SGE answers.E-E-A-T (Experience, Expertise, Authority, Trust) adalah gerbang binary di Google SGE Stage 3. Tambahkan byline penulis asli + bio di setiap posting blog. Kalau bisnis kamu notable, buat entri Wikidata — itu jadi benih untuk Wikipedia, Google Knowledge Panel, dan jawaban SGE.
1) Add author byline + bio to every blog post. 2) Check if a Wikidata entry exists for your brand; if no, create one with verifiable sources.1) Tambahkan byline penulis + bio di setiap posting blog. 2) Cek apakah entri Wikidata ada untuk brand kamu; jika tidak, buat satu dengan sumber yang dapat diverifikasi. · ~1 hour for byline, ~3 hours for Wikidata
E-E-A-T (Experience, Expertise, Authority, Trust) is the framework Google publishes for evaluating content quality. Author bylines and verifiable credentials map directly to it. Wikidata entries feed the Google Knowledge Graph, which AI surfaces use to disambiguate entities. (mechanism, not an outcome claim)E-E-A-T (Experience, Expertise, Authority, Trust) adalah framework yang Google publikasikan untuk mengevaluasi kualitas konten. Byline penulis dan kredensial yang dapat diverifikasi memetakan langsung ke sana. Entri Wikidata memberi makan Google Knowledge Graph, yang digunakan permukaan AI untuk membedakan entitas. (mekanisme, bukan klaim hasil)
<!-- 1. Add author byline to every blog post -->
<article>
<h1>Judul artikel kamu</h1>
<!-- Author byline at the top, in HTML not just text -->
<div class="byline">
<img src="/team/author-name.jpg" alt="Nama Penulis" width="48" height="48" />
<div>
<a href="/about/author-name" rel="author">
<strong>Nama Penulis</strong>
</a>
<span>Founder & CEO, [Nama Bisnis kamu]</span>
<time datetime="2026-07-25">25 Juli 2026</time>
</div>
</div>
<p>Konten artikel...</p>
<!-- 2. Author schema markup in <head> or at bottom of article -->
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "Article",
"headline": "Judul artikel kamu",
"author": {
"@type": "Person",
"name": "Nama Penulis",
"jobTitle": "Founder & CEO",
"worksFor": {
"@type": "Organization",
"name": "Nama Bisnis kamu"
},
"url": "https://yourdomain.com/about/author-name",
"sameAs": [
"https://www.linkedin.com/in/author-name",
"https://twitter.com/author-name"
],
"knowsAbout": ["kopi spesialti", "warung kopi", "F&B Indonesia"]
},
"datePublished": "2026-07-25",
"dateModified": "2026-07-25",
"publisher": {
"@type": "Organization",
"name": "Nama Bisnis kamu",
"logo": {
"@type": "ImageObject",
"url": "https://yourdomain.com/logo.png"
}
}
}
</script>
</article>
<!-- 3. Wikidata: if your business is notable (years operating, media coverage,
awards), create an entry at https://www.wikidata.org with:
- instance of: business, restaurant, etc.
- country: Indonesia
- inception: founding year
- headquarters: city
- same as your website URL
- sourced with at least 2 verifiable references (news articles, gov records) -->