Foreign Trade Customer Development 5: Feature Analysis

Through 📚"4️⃣ Seed Customer", we've found seed customers. How do we use these customers to find more customers?

💡 Key points

• 1️⃣ Business feature extraction: including but not limited to customer website, LinkedIn, zoominfo and other channels;
• 2️⃣ Page feature extraction: what are the structural features of customer websites and social media (LinkedIn) pages?

• 🤔 Question 1️⃣: Customer's peers are also your potential customers — how do you find them quickly and accurately?
• 🤔 Question 2️⃣: What common features do customers and their peers (same customer group) share?
• 1️⃣ Similar business: peers usually have similar products or services;
• 2️⃣ Similar website structure: customer groups in the same market often have similar website features. For example, customers in English-speaking regions typically have contact us and about us pages — some manufacturers or brands also have pages like Replacement Parts and Dealer Locator.
• 3️⃣ Similar social media platforms: customers in each region have different social media preferences, so customer development can be prioritized.

👉 With these two directions of accumulation, you'll get more potential customers:

• 1️⃣ Expand more business feature words: extract as many keywords as possible from this group via customer websites and social media
• 2️⃣ Mine more page feature info: study customer websites, dig out more unique-page page features, and expand customer numbers by continuously rotating page feature words.

👉 How can page features help you find more customers beyond Google's limits?

• Google search results are typically capped at 300. To break this cap, you can restrict results to unique pages on a site (e.g. "about us"), ensuring each company appears only once — finding more independent customers.

• 1️⃣ Single Google search returns at most 300 results

  A single keyword on Google returns at most 300 results. If a company appears 10 times in the search, then 1 search only finds 30 customers. How do we break through this?

• 2️⃣ How to break through Google's result count limit?

  Restrict the search results to only show a certain page type, like the about us page — a site has one and only one. With 300 keyword results, the number of companies involved isn't 30 — it's 300.

  Take searching e-bike as an example. If a customer specializes in e-bikes, they may appear many times in results. But if I limit results to the customer's about us page — a page that exists only once per site — the company appears only once in the search. One keyword can find 300 target customers instead of 30 or fewer.

👉 If a prospect's website doesn't have an about us page, won't searching only about us miss this customer?

• Yes — that's why we need to collect as many page feature words as possible, like replacement parts, etc. By expanding coverage, we keep growing the customer pool.

Diagram: Business feature and page feature analysis methods.

1. Analysis Preparation

• Reference 👉 Customer Marketing Data Sheet to complete customer basic info
• Customer website: customer website domain (don't include http or www — just the top-level domain, for dedup)
• Customer group: includes main products, business role, language, region
• Social media info: includes linkedin/facebook/zoominfo/twitter info.
• Company HQ: usually confirmed via LinkedIn, or check on the official site.

Diagram: Customer marketing data sheet example — recording customer basic info.

2. Feature Word Library

👉 Solving two challenges

• 1️⃣ Business features: what are the features of the customer's business?
• 2️⃣ Feature classification: how should these features be classified?
• 3️⃣ Word library management: how to efficiently manage these features?

Diagram: Feature word library management — categorically recording different types of business and page feature words.

1. Feature Types

① 🌟 Product Words

• Meaning: What products do they sell? Usually physical products.
• Source: customer website or social media, in the top or top menu
• Note: collected keywords must match the original page text. If the original is in Spanish, record it in Spanish.

② ✨ Service Words

• Meaning: What services do they provide? Usually virtual value-added services
• Source: same as product words
• Note: also match the language — services provided differ significantly across groups. Take water sports products: when selling products, they usually provide winter storage, rental, and replacement parts services.

③ 🌙 Industry Words

• Meaning: What industry are they in? Usually found on social media like LinkedIn with related info, and business data also classifies by industry.
• Source: collected from social media or business data customer pages.
• Note: also match the language, and industry words differ significantly across platforms. For example, a customer may write "manufacturer" on LinkedIn, but third-party business databases may classify them as "distributor". You need to enter the specific customer page to identify accurately.

④ 🌟 Role Words

• Meaning: What role do they play in the supply chain? Usually supplier, wholesaler, brand, etc.
• Source: usually from the customer's website or social media intro pages.
• Note: also match the language.

---

2. Customer Website

• Extract feature words from the customer's website, including but not limited to: product words, service words, role words

👉 Feature sources

• 1️⃣ Website menus: the customer website's business feature words usually come from the top menu or footer menu.
• 👉 The menu shows customer product words and service words and related keywords
• 2️⃣ Intro pages: pages like about us and company introduction also show related keywords
• 👉 These pages show how customers describe their own role (role words) — like: manufacturer

• Take this website as an example: https://www.orukayak.com/

• In the top menu and footer menu, you can see customer product words and service words and related keywords

• Product keywords: Kayaks/Paddle/Carbon Paddle/Fiberglass Paddle/Bilge Pump

• Service keywords: Replacement Parts

Diagram: Oru Kayak website menu — showing product words and service words.

• Take this page as an example: About Us – Bending Branches
• On this page, we can see the customer describes their own role as manufacturer

Diagram: Bending Branches' "About Us" page showing customer role as manufacturer.

---

① 📝 Recording Info

• Sample sheet reference: 👉 Customer Marketing Data Sheet

💡 Why record info through a spreadsheet?

• To dedupe: once a keyword is collected, avoid collecting it again.
• To trace: when we find searched customers aren't accurate, we can use the spreadsheet to trace and analyze.

• Fill-in instructions
• 1️⃣ Keyword: keyword collected from customer website
• 2️⃣ Description: description of this keyword
• 3️⃣ Feature type: choose type from dropdown
• :::note 👉 Click to see a screenshot of feature type dropdown Diagram: Feature type dropdown example. :::
• 4️⃣ Source customer: fill in customer's domain, matching the customer website in 2️⃣ Customer List
• :::note 👉 Click to see source customer matching customer website Diagram: The customer website field in the customer list. :::
• 5️⃣ Language: this is a formula — uses source customer to look up the matching language in 2️⃣ Customer List.
• :::note 👉 Click to see language matching lookup formula Diagram: VLOOKUP formula for querying language in the spreadsheet.
• You can copy this formula to study. =VLOOKUP(D2,'2️⃣ Customer List'!A:E,5,0)
• If you don't understand, ask the AI Diagram: Asking AI to explain the VLOOKUP formula. :::
• 6️⃣ Source type: choose the keyword source — website/linkedin/zoominfo/facebook
• :::note 👉 Click to see source type dropdown screenshot Diagram: Source type dropdown example. :::
• 7️⃣ Source link: which link did it come from — just paste the link.
• 8️⃣ Screenshot: take one if you want...
• 9️⃣ Unique identifier: very important — used for dedup.
• :::note 👉 Click to see unique identifier formula Diagram: CONCATENATE formula for generating unique identifiers in the spreadsheet.
• You can also copy this formula: =CONCATENATE(VLOOKUP(D2,'2️⃣ Customer List'!A:B,2,0),"-",F2,"-",A2)
• If you don't understand, you can ask AI to explain Diagram: Asking AI to explain the CONCATENATE formula. :::

Diagram: How feature word info is recorded in the customer marketing data sheet.

3. Social Media — LinkedIn

• Extract keyword info from customer's social media, using LinkedIn as the example.

tip

• Difference: vs. website 👉 on LinkedIn you can directly extract customer industry words

• Take this social media page as an example: https://www.linkedin.com/company/confluence-outdoor

• Basic info: from customer's intro, main business, you can extract product words, service words, role words, etc.

• Industry info: LinkedIn has industry info — usually written by the customer themselves, fairly accurate.

Diagram: Confluence Outdoor's LinkedIn homepage showing basic info and industry info.

---

① 📝 Recording Info

• For filling instructions, reference the customer website section above.
• Sample sheet reference: 👉 Customer Marketing Data Sheet

Diagram: Customer marketing data sheet recording feature word info extracted from LinkedIn.

4. Business Data — ZoomInfo

• Extract customer keywords from business data, using zoominfo as the example.

tip

👉 Business data is more about extracting customer industry words, role words, etc., though a small portion also has product words and service words.

• Take Confluence Outdoor as an example
• From this page, we can easily extract the customer's industry/product/role info.

Diagram: Confluence Outdoor's info page on ZoomInfo — extractable industry, product, role info.

---

① 📝 Recording Info

• Following the above, we can extract customer product words, service words, role words, etc.
• Sample sheet reference: 👉 Customer Marketing Data Sheet

Diagram: Customer marketing data sheet recording feature word info extracted from ZoomInfo.

3. Page Word Library

• 🔔 Before starting, please reference 📚Search Syntax

1. Problem at Hand

---

❓ Customers appearing multiple times in search results

• Multiple appearances: when searching on Google, a single website often appears multiple times in the results.
• Google search limit: for our customer search purposes, if a customer appears multiple times, then under Google's 300-result cap, we effectively find fewer customers.
• 👉 So the question: how do you make a customer appear only once in your search?

---

• Maybe that text isn't clear enough — let's use seabirddesigns.com as an example to show this issue.

👉 Case study: seabirddesigns.com appearing multiple times in search and optimization

Using seabirddesigns.com as an example, this directly shows how to optimize search results by restricting to unique pages — avoiding the same customer being associated with multiple pages and appearing repeatedly, thus improving search efficiency. See actual operations below.

Example 1: Regular search Kayaks (no page restriction)

When searching Kayaks on Google, seabirddesigns.com has multiple associated pages. (You can use 📚site syntax)

<u><span style={{color: '#e67c00'}}>site:seabirddesigns.com</span></u> <u><span style={{color: '#e67c00'}}>Kayaks</span></u>

Diagram: Google search result for site:seabirddesigns.com Kayaks — shows multiple associated pages.

Example 2: Restrict to about us page when searching Kayaks

If we restrict to only the about us page, we write a new syntax (using 📚intitle and 📚"" syntax)

<u><span style={{color: '#e67c00'}}>site:seabirddesigns.com</span></u> <u><span style={{color: '#e67c00'}}>Kayaks</span></u> <u><span style={{color: '#e67c00'}}>AND</span></u> <u><span style={{color: '#e67c00'}}>intitle:"about us"</span></u>

Diagram: Google search result for site:seabirddesigns.com Kayaks AND intitle:"about us" — restricted to only "about us" pages.

• Summary: without restricting results to a specific page, searching a product word can cause a single website to appear multiple times — severely disrupting both the count and quality of customers found.

2. How to Solve

👉 We solve the above problem in 3️⃣ ways

• 1️⃣ Unique pages: a site has one and only one of these pages — like "about us" or "contact us". This lets us find as many results as possible.
• 2️⃣ Matching language: Chinese "关于我们" corresponds to English "about us", German "Über uns", Russian "О нас" — this makes it easier to find minor-language customers.
• 3️⃣ Never translate: all keywords must come from the customer's website — never translate directly, since translations may not match customer usage habits.

👉 Language precision: record page features matching the customer website's language

When searching minor-language customers, using target-language page feature words (like German websites' "Über uns" instead of Chinese "关于我们") is crucial. This ensures search result precision and avoids missing potential customers due to language mismatch.

Case: Searching German "data cable (Datenkabel)" customers

Background: Directly searching "Datenkabel" on Google produces lots of results, but the same customer (especially high-authority sites like Amazon) may appear multiple times — lowering search efficiency. The goal is to make each customer appear only once.

Diagram: Google direct search for "Datenkabel" — shows high-authority sites like Amazon appearing multiple times.

Wrong attempt: using Chinese page feature words Restrict to only Chinese "关于我们" pages for the search (using 📚intitle, 📚AND, 📚"" syntax):

<u><span style={{color: '#e67c00'}}>Datenkabel</span></u> <u><span style={{color: '#e67c00'}}>AND</span></u> <u><span style={{color: '#e67c00'}}>intitle:"关于我们"</span></u>

Result: 0. Because German customer websites typically use German page features like "Über uns".

Diagram: Google search for Datenkabel AND intitle:"关于我们" — no results due to language mismatch.

Correct example: using German page feature words Switch the page feature word to German "Über uns" (using 📚intitle, 📚AND, 📚"" syntax):

<u><span style={{color: '#e67c00'}}>Datenkabel</span></u> <u><span style={{color: '#e67c00'}}>AND</span></u> <u><span style={{color: '#e67c00'}}>intitle:"Über uns"</span></u>

Result: the search results become very precise.

Diagram: Google search for Datenkabel AND intitle:"Über uns" — using German page feature words for precise results.

3. Analyze Pages

• The following are typical unique pages — we should organize and catalog them.

Page Title	Source Link	Screenshot
About Us	about-us	Diagram: Example of Kayak Chicago website's About Us page.
Contact Us	contact-us	Diagram: Example of Jackson Kayak website's Contact Us page.
Refer a friend	refer-a-friend	Diagram: Example of Manta Racks website's Refer a Friend page.
Become a host new	become-a-host-new	Diagram: Example of Oru Kayak website's Become a host new page.

🤔 Why not collect products-type pages as unique page features?

Many e-commerce sites (B2B or B2C) have product links containing products (like .../products/item-a, .../products/item-b). These pages don't meet the "unique page" standard, because a site can have many different product pages containing products. So we don't recommend collecting them as page feature words.

Using https://www.orukayak.com/ as an example, let's look at a few links:

https://www.orukayak.com/products/lake
https://www.orukayak.com/products/beach-lt
https://www.orukayak.com/products/the-coast-xt

• 👉 These links all have a clear feature — each link contains products and doesn't meet the unique page standard.
• 🔔 For many online stores, B2B or B2C, customers like to add products to product links, so we don't recommend collecting product pages as features.

4. Recording Features

• Sample sheet reference: 👉 Customer Marketing Data Sheet

](https://cos.files.maozhishi.com/data/web/web-files/img/20240816002947.png)

Diagram: How page features are recorded in the customer marketing data sheet.

• Fill-in instructions

  • 1️⃣ Feature source: fill in the source link of the collected info
  • 2️⃣ Page type: what type of page? About us, contact us, replacement parts — sites usually have one of each
  • 3️⃣ Source title: title of the source page
  • 4️⃣ Source link: not the full link — just part of the URL. :::tip 👉 Matching feature source, source title, source link Diagram: Page feature info source, title, and partial link example. :::
  • 5️⃣ Screenshot: screenshot of the page feature, for fast review later.
  • 6️⃣ Page formula: involves 5 syntaxes: 📚1️⃣ Title contains (intitle), 📚2️⃣ URL contains (inurl), 📚3️⃣ AND, 📚4️⃣ OR (OR / |), 📚5️⃣ Exact match ("")) 👉 Decide whether to use AND or OR based on the situation

    • 🌟 inurl:about-us | intitle:"About Us" Why use OR?

      • When the URL contains "about-us" or the page title is "about us", either match confirms it's an "About Us" page.

    • 🌟 inurl:experience AND intitle:"Experience" Why use AND?

      • Either alone can't confirm it's the "try before buy" page — even with "AND" doesn't guarantee the result is this page.

  • 7️⃣ Notes: notes about the page info — for your own reference
  • 8️⃣ Source customer: fill in the customer's domain :::note 👉 Click to see source customer matching customer website Diagram: The customer website field in the customer list. :::
  • 9️⃣ Language: linked via the source customer column :::note 👉 Click to see language formula info

Diagram: The formula in the spreadsheet for looking up language by "source customer".

:::

4. Exclusion Word Library

• To be updated

5. Common Questions

• 👉 If I restrict the search to certain pages and the customer's site doesn't have that page, won't I miss them?
• Yes — so we need to collect as many page feature words as possible, like replacement parts, privacy policy, company information — these common unique pages. By expanding coverage, we keep growing the customer pool.

6. Further Reading and Resources

1. FAQ

• ❓ During feature analysis, what if I have too few seed customers?

  • A: If you have too few seed customers, first review the methods in 📚"4️⃣ Seed Customer" and try to expand more high-quality seed customers. You can also relax the initial filtering standards and look for more potential seed customer sources in similar fields or upstream/downstream industries. Analysis accuracy heavily depends on seed customer quality and quantity.

• ❓ Extracted business feature words aren't finding enough customers — how to optimize?

  • A: First check keyword accuracy and language matching (English sites use English keywords). Then try synonyms, near-synonyms, or broader industry words. At the same time, deep-analyze existing customers' websites and social media to dig out more unique and segmented business feature words. Reference the Feature Word Library section above to systematically manage and iterate your keywords.

• ❓ What types of page feature words are there? How to collect efficiently?

  • A: Page feature words mainly refer to unique page identifiers on a website, like "About Us" (About Us), "Contact Us" (Contact Us), "Become a Dealer" (Become a Dealer), "Privacy Policy" (Privacy Policy), etc. The efficient collection method: analyze multiple seed customer websites, record commonly used unique pages in different languages, and systematically organize using the table in the Page Word Library section above.

• ❓ How do I tell if a page qualifies as a "unique page"?

  • A: Generally, "unique pages" have content and functions that are distinctive within a site — they usually appear only once. For example, pages related to company introduction, contact info, or legal statements. Product list pages and blog article list pages usually don't qualify as unique pages — they paginate or generate multiple similar links as content grows.

2. Learning Tips

• Continuously iterate the word library: feature word libraries and page word libraries aren't static — they need continuous additions and optimizations based on actual search results and new customer cases. Regularly reviewing and updating the word library is key to improving customer development efficiency.
• Multi-dimensional cross-validation: combining business features and page features in your searches improves potential customer precision. Don't rely on a single feature in isolation.
• Deep-analyze competitors: study competitors' customer development strategies — see what types of customer features they focus on. This can inspire your feature analysis.
• Master search syntax: skillfully using the advanced commands in 📚Search Syntax — like site:, intitle:, inurl:, AND, OR, "" — lets you combine feature words flexibly for searches.

3. Related Reading

• 📚"1️⃣ Theoretical Foundation": understand the basic theory and overall framework of foreign trade customer development, building a solid foundation for feature analysis.
• 📚"4️⃣ Seed Customer": learn how to find and screen high-quality seed customers — the prerequisite for feature analysis.
• 📚"6️⃣ Search Practice": combine the feature analysis methods in this article — learn how to use various search techniques in real-world operations to find potential customers efficiently.

---

Permanent update address: https://docs.lafa.xin/zhinan/feature-analysis-section/