Customer database: how to build one for B2B sales in India

The quick answer

A customer database is one organised record of every business you sell to or want to sell to: the company, the person who decides, how to reach them, what they bought, and what you agreed with them about contact. For an Indian B2B seller it can start as a single Excel sheet with one row per business and fixed columns (company, GSTIN, city, contact person, mobile, email, source, consent, last contact, next step), and move into a CRM database when more than one or two people work it every day. What makes it useful is not the software but three habits: one row per business with no duplicates, a recorded source and consent for every contact, and a fixed routine for cleaning it.

Most pages on this search are written by foreign software companies selling customer database software. This guide is for Indian manufacturers, distributors, agencies and service firms. It covers the difference between a customer database, a CRM database and a lead list, the fields worth keeping (including Indian ones like GSTIN and DND status), how to build one step by step from your invoices, enquiries and bought lists, how reachable each trade is by phone and email according to Starleads' data, and what the Digital Personal Data Protection Rules 2025 and TRAI's rules mean for the data you keep.

Customer database, CRM database and lead list: what each one is

People use these words loosely, and the confusion costs money: a business buys CRM software expecting it to come with customers in it, or buys a contact list expecting it to behave like a customer database. They are different things that work together.

• A customer database is the record itself: the businesses you deal with, their people, contact details and history. It can live anywhere, from an Excel file to an accounting package to a CRM.
• A CRM database is a customer database kept inside customer relationship management software, which adds tasks, reminders, deal stages, call logs, email tracking and permissions for a team. The CRM is the tool; the database is what you put into it.
• A lead list or prospect list holds businesses that have not bought from you yet. It is the raw material: once a prospect replies, it moves into the customer database with a stage such as "interested" or "quoted".
• Customer data management (often searched as "customer database management") is the routine of keeping all of this correct: removing duplicates, updating numbers, recording consent and deleting what you no longer need.

The rest of this guide treats the customer database as the asset and the software as a choice you can change later. If you already know you need software, our comparison of the best CRM software in India covers prices and plans, and the guide to telecalling CRM apps covers tools built around phone calling.

The fields an Indian B2B customer database needs

A database is only as useful as its columns. Too few, and you can't sort or segment; too many, and nobody fills them. The fields below are the ones that answer the questions a B2B seller in India actually asks: who is this, how do I reach them, may I contact them, what did they buy, and what happens next. Keep one row per business, and if a business has several people you deal with (owner, purchase manager, accounts), either add a second sheet of contacts linked by a customer ID or keep the two most important people on the same row.

Two rules make these fields work. First, use drop-down lists (Data > Data Validation in Excel) for industry, stage, source and consent, so you get "Lapsed" every time and not "lapsed", "old" and "inactive". Second, keep personal notes factual and business-related: what they asked for, what you promised. The DPDP framework gives people the right to see and correct what you hold about them, so write nothing you would not want them to read.

Where the records come from, and how reachable each trade is

A B2B customer database is fed from three places. Your own records come first because they are the most valuable: these businesses already know you.

1. Your own records: sales invoices and ledgers in your accounting software, quotations sent, enquiries from your website, IndiaMART or TradeIndia leads, WhatsApp chats, visiting cards from trade fairs, and the phone contacts of your salespeople (which leave with them if they are not copied in).
2. Public registers: the MCA company master data, the GST taxpayer search and the Udyam register let you confirm a business exists, its registered name and status. They rarely give you a working phone number for the person who buys; our guide to company search in India explains each look-up.
3. Bought or compiled lists: for markets where you have no customers yet, a list of businesses in a chosen industry and city gives you the prospects to start with. Our guide to buying business leads lists what to check before you pay.

How you plan to reach prospects should depend on the trade, because the share of contacts with a mobile number or an email address varies a great deal. The table below is from Starleads' own data, published on the Insights page and checked on 28 October 2026. These are counts of contacts in Starleads' lists, not an official census of Indian businesses, but they show which trades can be reached by email and which mostly by phone.

What this means for your database design: if you sell to kirana and grocery stores, almost every row will have a mobile number and almost none an email, so the phone and WhatsApp columns (and the consent and DND columns that go with them) matter most, and an email-first CRM will sit half empty. If you sell to importers or exporters, most rows have an email, so an email sequence is realistic, and you should plan for domain authentication before sending. Lawyers are the hard case: fewer than half the rows in Starleads' data have a mobile, so expect to rely on office numbers, referrals and bar associations.

The same data also shows where business is concentrated, which helps when you decide which cities to fill first. Mumbai has 5,77,978 business listings in Starleads' data, Hyderabad 4,09,205 and Bengaluru 3,39,403; at state level Maharashtra has 11,30,630 and Delhi 9,85,771. Overall Starleads holds about 1.6 crore contacts in 6,908 lists across 201 industries and 414 cities.

How to build a customer database step by step

This is a method for a small team starting from scattered records. It assumes Excel or Google Sheets; the same steps apply if you import the result into a CRM at the end.

1. Decide the columns first. Copy the field list above into row 1 of a new sheet, delete the fields you will never use, and add drop-down lists for industry, stage, source and consent. Freeze the header row.
2. Export your customers from accounting. Your accounting software can export the customer or party master and a sales register. Paste the party names, GSTINs, addresses and phone numbers into the sheet and mark the source "Existing customer". Add the last invoice date and the total billed in the last 12 months from the sales register.
3. Collect enquiries and cards. Export enquiries from your website form and any marketplace you use, and type in trade fair cards and the numbers salespeople keep on their phones. Mark each one's source. Anyone who sent you an enquiry about your product is a warmer prospect than any bought list.
4. Clean the phone numbers. Numbers arrive as "+91 98xxx xxxxx", "098xxxxxxxx" and "98xxx-xxxxx". In a helper column, strip spaces, hyphens and the plus sign and keep the last ten digits, for example =RIGHT(SUBSTITUTE(SUBSTITUTE(SUBSTITUTE(F2," ",""),"-",""),"+",""),10). Then flag anything that is not ten digits with =LEN(G2)<>10 and look at those rows by hand.
5. Standardise names and places. Use =TRIM() to remove stray spaces and =PROPER() for consistent capitals in names and cities. Decide one spelling per city ("Bengaluru" or "Bangalore", "Gurugram" or "Gurgaon") and use Find and Replace so every row agrees.
6. Remove duplicates. Sort by GSTIN, then by cleaned mobile number, then by email. Flag repeats with =COUNTIF($G$2:$G$10000,G2)>1 on each of those columns. Merge duplicate rows by hand, keeping the most recent contact details and the earliest source. Excel's Remove Duplicates button deletes rows without merging them, so use it only on a copy.
7. Check the email addresses. A simple test catches obvious errors: =AND(ISNUMBER(SEARCH("@",H2)),ISNUMBER(SEARCH(".",H2,SEARCH("@",H2)))). Before any bulk email, run the addresses through an email verification service; our comparison of email verification tools covers the main ones.
8. Add prospects from a list. Import a list for the industry and cities you want next, mark the source as the list and date, set the stage to "Prospect", and run the same cleaning and duplicate checks against your existing customers so you don't cold-call someone who already buys from you.
9. Record consent and DND status. Fill in what each contact has agreed to and check numbers against the NCPR before promotional calls or SMS. Our DND check guide explains how.
10. Set the routine. Agree who updates the last contact and next step after every call (the salesperson, the same day), who owns the master file, and where it lives: one shared file or one CRM, never copies on several laptops.

Once the sheet has more than a few thousand rows, or more than two people updating it, import it into a CRM. Most CRMs accept a CSV upload and ask you to map each column to a field; the clean columns and drop-down values from steps 1 to 6 are what make that import go smoothly.

Keeping the database clean: a maintenance routine

Business contact data goes out of date on its own: people change jobs, firms change numbers, shops close and new ones open. You can't stop that, but a fixed routine keeps the damage small. The schedule below suits a small B2B team; adjust the timing to how often you contact your market.

When a number bounces or a call reaches the wrong business, don't just delete the row. Mark it "wrong number" with the date, so nobody re-imports the same bad record from an old file next month. And when you buy a fresh list for a market you already work, compare it with your database first: the new list may hold updated numbers for businesses you already have.

The rules: DPDP Act and Rules 2025, TRAI, and email

A B2B customer database still holds personal data. A proprietor's name and mobile number, or a purchase manager's email, belong to individuals, even when you use them only for business. Three sets of rules apply to what you collect and how you use it. This is a summary, not legal advice; for a large database or anything unusual, ask a lawyer.

Digital Personal Data Protection Act 2023 and Rules 2025

According to the Press Information Bureau's release of 14 November 2025, the government has notified the DPDP Rules, 2025, which give full effect to the DPDP Act passed in August 2023. The points that matter for a customer database:

• Phased start: the Rules give organisations an 18-month phased compliance timeline to adjust their systems.
• Consent notices: businesses (called "Data Fiduciaries") must give a separate, clear consent notice explaining the specific purpose for which personal data is collected and used.
• Seven principles: consent and transparency, purpose limitation, data minimisation, accuracy, storage limitation, security safeguards and accountability. For a database, that means collecting only the fields you use, keeping them correct, deleting what you no longer need and protecting the file.
• People's rights: individuals can ask to access, correct, update or erase their personal data, or nominate someone to do so, and the business must respond within 90 days.
• Contact point: businesses must display contact details, such as a designated officer or Data Protection Officer, for questions about personal data.
• Breaches: if personal data leaks, affected people must be told promptly in plain language what happened and what is being done.

The penalties are large. The PIB's explainer of 17 November 2025 says failing to keep reasonable security safeguards can attract a penalty of up to ₹250 crore, failing to notify a breach up to ₹200 crore, and other violations up to ₹50 crore. It also says a four-member Data Protection Board of India will take complaints online, with appeals to TDSAT. In practice, for a small seller: password-protect the file, limit who can export it, don't email the full database around, keep a record of where each contact came from, and act on every request to stop or delete.

TRAI rules for calls and SMS

Promotional calls and SMS are governed by TRAI's Telecom Commercial Communications Customer Preference Regulations. TRAI's preference registration page explains that people can register their preferences (the National Customer Preference Register, or DND) by SMS or call to 1909, through an app or their operator's website. Its advice to senders sets out what a business must do before sending commercial communication: register as a principal entity, register headers and content templates on the DLT platform, and register the consent templates it uses to collect customers' consent. A promotional message or call, in TRAI's words, is one for which the sender has not taken explicit consent from the recipient. That is why the consent and DND columns belong in your database. Our guides to DLT registration for SMS and DND checks cover the steps.

Email

Google's email sender guidelines require senders of more than 5,000 messages a day to Gmail accounts to keep spam rates reported in Postmaster Tools below 0.3% and to support one-click unsubscribe in marketing messages. Even if you send far fewer, recording unsubscribes in the database and never mailing those addresses again keeps your domain out of spam folders. Our guide to SPF, DKIM and DMARC explains domain authentication.

Starleads sells business and professional contacts only, with no consumer lists and no lists of people picked out by their wealth or investments, and anyone whose number or email appears in its lists can ask to be removed through its Remove my data page. If you add Starleads contacts to your database, the DND check, consent records and opt-outs are your responsibility as the sender.

Putting the database to work

A clean database pays for itself only when you use it to decide whom to contact next. Four simple uses come straight from the columns above.

• Rank your customers. Sort by billing in the last 12 months. Mark the top fifth "A", the next third "B" and the rest "C". A accounts get a visit or call every month; C accounts get a quarterly message. Sorted this way, the database shows you how much of your sales comes from a few accounts, and which ones they are.
• Win back lapsed customers. Filter "last order more than six months ago" and call them before you call any cold prospect. They already know your product, so a short call asking what changed often brings back an order or tells you what a competitor is offering.
• Find more businesses like your best customers. Look at the industry and city of your A accounts. If most are, say, chemists in Surat or transporters in Nagpur, buy or build a prospect list for that exact industry and nearby cities rather than a broad all-trades list. Starleads' list pages show the count and the share with mobile and email for each industry and city before you buy, and a city or state list costs ₹499 (All-India bulk data for an industry costs ₹1,999), with 60 sample contacts free to download.
• Measure your channels. Count customers by source. If trade fair contacts turn into customers twice as often as marketplace enquiries, that tells you where next year's budget should go. Our guide to the B2B sales funnel shows how to work back from a sales target to the number of calls and meetings you need.

When you start calling the new prospects, our 7-day calling plan for a B2B contact list gives a day-by-day routine that writes its results back into the same database.

Questions people ask about customer databases

What is a customer database?

It is an organised record of the businesses or people you sell to and want to sell to, with their contact details, purchase history and the history of your contact with them. In B2B it has one row per business, with the decision-maker named.

What is a CRM database, and is it the same as a customer database?

A CRM database is a customer database stored inside CRM software, which adds reminders, deal stages, call logs and team access. The data is the same; the software makes it easier for several people to work it. A CRM starts empty: you fill it from your own records and from prospect lists.

How do I create a customer database in Excel?

Put fixed columns in row 1 (customer ID, business name, GSTIN, industry, city, contact person, mobile, email, source, consent, stage, last order, last contact, next step), add drop-down lists for the repeating values, paste in your customers from your accounting software and enquiries, clean the phone numbers, and remove duplicates by GSTIN and mobile. The step-by-step section above gives the formulas.

What is customer database management?

It is the routine that keeps the database correct and lawful: entering every contact the same day, merging duplicates, refreshing order data from accounts, acting on opt-outs and data requests, and deleting data you no longer need.

Can I buy a customer database?

You can't buy your customers, but you can buy a list of prospects: businesses in the industries and cities you sell to, which you then contact and turn into customers. Check a sample, the share of rows with a mobile number and email, and the seller's removal process before you pay. Never buy lists of private individuals; Starleads doesn't sell them.

Is it legal to keep and use a B2B customer database in India?

Yes, if you follow the rules: the DPDP Act and Rules for the personal data in it (purpose, consent notices, security, and responding to requests within 90 days), TRAI's rules for promotional calls and SMS (DND, DLT and consent), and email senders' requirements such as unsubscribe links. Take legal advice for anything beyond routine B2B selling.

How often should I clean a customer database?

Update rows after every contact, merge duplicates weekly, act on opt-outs monthly, refresh purchase data quarterly and delete stale prospects yearly. If more than a few rows bounce or reach the wrong business in a single calling session, clean that segment before you continue.