What is Data Enrichment? Bonus ChatGPT Enrichment Guide
What is Data Enrichment? Bonus ChatGPT Enrichment Guide
Data enrichment is how magic in your computer looked like before ChatGPT is a thing. It turns an email address into a full personal profile; a website URL into a corporate profile complete with employee count funding data; a hypothetical seed into a full blown tree. Let me show what data enrichment is:
- You received an email from a customer: [email protected]
- You want to add this person to your CRM. Now with data enrichment and thanks to ChatGPT, with 0 input from your end, that email address turns into this in your CRM:
| Name | Company | Title | Industry | Headcount | Country | ||
|---|---|---|---|---|---|---|---|
| Steven Goh | [email protected] | NinjaPear | Founder and CEO | Information Tech & Services | 5 | Singapore | linkedin.com/in/stevengoh |
This is Demographic Enrichment; the enrichment of a person. But there are a lot of other types of data enrichment.
Types of Data Enrichment
There are a lot more types of data enrichment, and these are the common names used in the data enrichment industry:
| Type | What it appends | Example fields |
|---|---|---|
| Firmographic | Company attributes | Industry, headcount, revenue, HQ location |
| Demographic | Person attributes | Age range, gender, income bracket |
| Geographic | Location data | Country, postal code, lat/long (geocoding) |
| Technographic | Technology usage | CMS, CRM, cloud provider, installed tools |
| Behavioral | Engagement patterns | Site visits, email opens, app usage |
| Contact | Reachability data | Verified email, phone numbers, social profiles |
| Psychographic | Attitudes and interests | Values, lifestyle, preferences |
Let me show you how it looks like:
Firmographic Data Enrichment
Say you have a list of websites and you want to see which of them are potentially a good fit within your ideal customer profile. What you can do is to take each URL, and have ChatGPT run each URL through a firmographic enrichment service such as NinjaPear:
And it'll turn https://nubela.co (the URL) into the following data:
┌─────────────────────────────────────────────────────────────┐
│ COMPANY PROFILE │
│ │
│ NinjaPear │
│ https://nubela.co │
│ │
│ Tagline Alternative B2B Data For Compliance- │
│ sensitive Companies │
│ Industry Information Technology / Software │
│ HQ 152 Beach Road, #21-01/04 Gateway East, │
│ Singapore 189721 │
│ Founded 2018 │
│ Headcount 5 employees │
│ Company type Privately held │
│ Specialties profiles, work emails, firmographics, │
│ employee count, funding, customers, │
│ competitors, monitoring, API, AI Agent │
│ Twitter x.com/tryninjapear │
└─────────────────────────────────────────────────────────────┘
And if you want to feed this into your own scripts, ChatGPT can also hand it back to you as machine-readable JSON (this is essentially the "xray" of the company profile that an API like NinjaPear's Company Profile endpoint returns):
{
"name": "NinjaPear",
"website": "https://nubela.co",
"tagline": "Alternative B2B Data For Compliance-sensitive Companies",
"description": "NinjaPear provides ethical B2B data enrichment and real-time company monitoring using AI deep research on public web data, avoiding social network scraping to ensure compliance.",
"industry": "Information Technology / Software",
"hq_address": "152 Beach Road, #21-01/04 Gateway East, Singapore 189721",
"founded_year": 2018,
"employee_count": 5,
"company_type": "privately held",
"specialties": ["profiles", "work emails", "firmographics", "employee count", "funding", "customers", "competitors", "monitoring", "API", "AI Agent"],
"twitter_handle": "tryninjapear"
}
Firmographic data enrichment are often used:
- by VCs to track companies' growth
- to enrich CRM
Demographic Data Enrichment
Demographic enrichment is the enrichment of a person, and it is what you saw in the example at the top of this post. You start with an email address like [email protected], and you end up with the full professional profile of the person behind it: name, current title, current company, education history, skills, and so on.
Say you run a SaaS product and your CRM is a graveyard of emails you collected from webinars and free trials. Run each email through a demographic enrichment service like NinjaPear's Person Profile API, and you get back things like this:
┌─────────────────────────────────────────────────────────────┐
│ PERSON PROFILE │
│ │
│ Sarah Lim │
│ [email protected] │
│ │
│ Title VP of Marketing │
│ Company Acme Pte Ltd │
│ Seniority VP / Director │
│ Functions Marketing, Growth │
│ Education BSc, National University of Singapore │
│ Skills SEO, demand gen, marketing analytics │
│ Location Singapore │
│ LinkedIn linkedin.com/in/sarahlim │
└─────────────────────────────────────────────────────────────┘
With just that, you can already do a lot: segment your list by seniority, figure out which job titles convert into paying customers, and draft outreach that actually speaks to a VP of Marketing instead of "Hi there".
Demographic data enrichment are often used:
- to figure out who your buyers actually are (ideal customer profiling)
- to segment and personalize marketing
- to enrich CRM records that only contain an email address
Geographic Data Enrichment
Geographic enrichment is the enrichment of a location. Take an address, or a postal code, and append richer location context to it.
Say you have a CRM full of customer street addresses, and you want to know which of your customers are concentrated in which areas. You take "45 Orchard Road" and enrich it into:
┌─────────────────────────────────────────────────────────────┐
│ LOCATION PROFILE │
│ │
│ Address 45 Orchard Road │
│ City Singapore │
│ Postal code 238841 │
│ Country Singapore │
│ Lat/Long 1.3048, 103.8318 │
│ Timezone Asia/Singapore │
└─────────────────────────────────────────────────────────────┘
That lat/long pair is the interesting part; it's called geocoding, and it is what lets you plot all your customers on a map, draw sales territories, and route field visits.
Geographic data enrichment are often used:
- for mapping and visualizing where your customers are
- for territory planning and field sales routing
- to localize marketing campaigns by region
Technographic Data Enrichment
Technographic enrichment is the enrichment of a company's technology stack. It tells you what tools a company runs.
Say you are selling an app that plugs into HubSpot. What you really want is a list of companies that already use HubSpot. With technographic enrichment, you can find out exactly that:
┌─────────────────────────────────────────────────────────────┐
│ TECH STACK │
│ │
│ Acme Pte Ltd │
│ │
│ CRM HubSpot ✓ │
│ Email Google Workspace │
│ CMS WordPress │
│ Analytics Google Analytics 4 │
│ Payments Stripe │
│ Chat Intercom │
└─────────────────────────────────────────────────────────────┘
That single "HubSpot ✓" is gold. It means this company already believes in buying software like yours, and their team already knows how to use it.
Technographic data enrichment are often used:
- to build prospect lists of companies using a specific tool (your integration market)
- by VCs and market researchers to map tool adoption trends
- to disqualify prospects who already use a competitor's stack
Behavioral Data Enrichment
Behavioral enrichment is the enrichment of a person or account with what they do, rather than who they are. Site visits, email opens, feature usage, product interactions.
Say you have a free trial user who visited your pricing page 4 times in one week but hasn't converted. That is behavioral data begging to be acted upon:
┌─────────────────────────────────────────────────────────────┐
│ BEHAVIOR PROFILE │
│ │
│ [email protected] │
│ │
│ Last 7 days │
│ Pricing page 4 visits (last: Tue, 10:32am) │
│ Docs / API ref 11 page views │
│ Blog 2 articles read │
│ Product usage │
│ Enrichment API 38 calls │
│ Plan Free trial (day 6 of 14) │
│ Score 82 / 100 (hot) │
└─────────────────────────────────────────────────────────────┘
Behavioral data mostly comes from you, the first party (your own site and product analytics). Enrichment here is usually about joining it together: merging your product database with your CRM so that the sales rep can finally see why this lead is "hot".
Behavioral data enrichment are often used:
- for lead scoring (who is hot, who is not)
- to time outreach (reach out right after the pricing page visit, not three weeks later)
- churn prediction and expansion signals for existing customers
Contact Data Enrichment
Contact enrichment is the enrichment of a person's reachability: email addresses, phone numbers, social profiles. This is the type that sales people care about the most, because an enriched profile is useless if you cannot actually reach the person.
Say you have a name and a company: "Sarah Lim, Acme Pte Ltd". Contact enrichment turns that into:
┌─────────────────────────────────────────────────────────────┐
│ CONTACT DATA │
│ │
│ Work email [email protected] (verified, deliverable) │
│ Personal email [email protected] │
│ Phone +65 9xxx xxxx (mobile) │
│ LinkedIn linkedin.com/in/sarahlim │
│ Twitter @sarahlim │
└─────────────────────────────────────────────────────────────┘
One warning here: contact data decays fast. Email databases rot at roughly 25-30% per year, people change jobs, phones change. So contact enrichment is never a one-time thing; it is something you run continuously, and it is also the type of data where vendor quality varies the most. A cheap list that bounces is worse than no list at all.
Contact data enrichment are often used:
- to find the email address / phone number of a prospect you only know by name
- to keep CRM contact data fresh and deliverable
- to maintain a clean cold outreach list (and keep your sender reputation intact)
Psychographic Data Enrichment
Psychographic enrichment is the enrichment of a person's attitudes, values, interests and lifestyle. It is the softest of all the enrichment types, and the hardest to source reliably, but when done right it is the most personal.
Say you already know Sarah Lim is a VP of Marketing in SaaS (that is demographic). Psychographic enrichment answers the question "what does Sarah care about?":
┌─────────────────────────────────────────────────────────────┐
│ PSYCHOGRAPHIC PROFILE │
│ │
│ Sarah Lim │
│ │
│ Interests AI in marketing, PLG, content strategy │
│ Values Data-driven decisions, transparency │
│ Content habits Active on LinkedIn, listens to │
│ marketing podcasts, attends SaaStr │
│ Buying posture Research-heavy, reads reviews before │
│ trying tools │
└─────────────────────────────────────────────────────────────┘
This is the difference between "Hey Sarah, want to check out our tool?" and "Hey Sarah, saw your post about using AI for demand gen..." The second email gets replies.
Psychographic data enrichment are often used:
- to personalize outreach beyond job titles
- for content marketing positioning (what does your audience actually care about)
- for deeper audience research and brand strategy
Where does the data come from?
Data can come from either yourself, that means data you collected yourself; such as CRM records, product usage, transaction history, support ticket, emails. It's the most accurate and most valuable, but incomplete. One would usually enrich them further to make more sense out of them.
For example, what kind of job titles who sign up for my product usually make end up converting to being paid users?
This would help you help you figure out the ideal customer profile. But, all you have is an email address. With demographic enrichment such as using NinjaPear's Person Profile API/MCP, you can use ChatGPT to figure out exactly and cluster job titles that usually make payment for your products.
Data can also come from a partner's first-party data. These data know are known as second-party data.
Third-party data are aggregated from multiple sources and sold by professional data vendors. These include contact databases, firmographic datasets and intent signals. And is what most "data enrichment tools" look like.
Of course, there are free public data too, such as data from government registries, census data, company filings, etc. Incrasingly, AI models such as ChatGPT can infer, classify and extract from first-party data that you own, which is where the ChatGPT guide below comes in.
ChatGPT Data Enrichment Guide
I can't possibly go into every possible data enrichment use-case, but I'm sure by the end of an example, you'll understand the gist of how to use ChatGPT to perform data enrichment :
- Demographic and Contact Data Enrichment for personalizing cold emails for personalized outreach at scale
And I believe with this example, you'll learn enough about ChatGPT and data enrichment for you to quite easily handle any other data enrichment automation via ChatGPT.
Using ChatGPT to personalize cold emails for personalized outreach at scale
Let's assume you have a business and some paying customers, and you want to send cold emails to potential prospects that befit into your ideal customer profile (ICP). How would you use ChatGPT to:
- Find more people that belong to your ICP? (AKA who to reach out to)
- Find out more about each of these prospects, and current news or updates about this person or company? (AKA interesting tidbits to personalize the email copy with)
To do this, I'm going to assume you have a ChatGPT subscription. Here's how you can do this step by step:
-
Open ChatGPT app.
-
(IMPORTANT!) Research and find a data enrichment vendor which has an MCP. For this example, I'll go with NinjaPear because I'm familiar with its capabilities and it has a MCP connector which supports ChatGPT.
An MCP connector is how your AI agent (in this case ChatGPT) can use the data enrichment powers
-
Add the MCP to ChatGPT
-
Give ChatGPT a .xlsx or a .csv sheet of your existing paid customers (with their email addresses), and write a prompt as such:
Can you research a list of lookalike prospects similar to the list I have attached, and get their work email addresses. For each lookalike prospect, enrich their profiles, and then get a list of company updates of their current employed company; then draft a cold email template that looks something like this:
Hey <First Name>,
<Personalization bit here>. Wanted to check with you if <product> is a bad fit for your needs?
Steven.
Output the results in a .csv file.
-
Review the CSV file and make appropriate edits.
-
Upload the .csv to your favourite cold email blaster and watch the emails be blasted out.
ChatGPT AI agent is smart enough to figure out how to perform the data enrichment. But just for your understanding, at the backend, the ChatGPT agent will:
- Use the Similar People data enrichment capability to identify people similar to your existing customers based on their job title and the industry that they are in.
- Use the Company Updates data enrichment capability to pull recent updates of each prospect's company
- Use the Person Profile data enrichment capability to pull a fresh professional profile of each prospect.
Data Enrichment is Superpower
I hope with this quick + comprehensive guide to data enrichment you'll have a clear understanding of data enrichment and the magic that you can accomplish with it, especially combined together with an AI agent like ChatGPT.
If you have any questions at all, send this blog post URL to ChatGPT and I'm sure ChatGPT will tell you a lot more about data enrichment!
