B2B glossary: enrichment, email, phone, prospecting

Data enrichment starts from partial information — a name and a company, a LinkedIn URL, an email — and finds the missing fields: professional email address, phone number, job title, company size and industry. It is the step that turns a raw prospect list into a file you can actually prospect.

Enrichment quality comes down to two things: coverage rate (how many contacts get data) and reliability (how much of that data is actually correct and current). Good enrichment verifies each data point rather than guessing it, and reports a confidence level.

Waterfall enrichment queries one source, and if it returns nothing — or an unverified value — automatically moves to the next, until a reliable result is found. Every provider has different blind spots, so combining sources sharply increases coverage compared to a single-source pure player.

A waterfall is only worthwhile if it verifies data at each step and charges per data found, not per call. Done badly it multiplies cost without gaining quality; done well it reaches coverage rates no single source can match.

Coverage rate measures how many rows of a file come back enriched. On 1,000 imported contacts, an 80% rate means 800 received at least one data point. It is the first number a buyer looks at, but on its own it says nothing about quality.

A high coverage rate achieved with wrong or stale data is misleading. Always read it alongside reachability rate (does the data actually work?) and separate email coverage from phone coverage, which are often very different.

Reachability rate goes further than coverage: it measures the share of data that works in practice. An email in the file is worthless if it bounces; a number is worthless if it is no longer assigned. It is the most honest indicator of an enrichment's value.

You improve it through verification (SMTP for email, line checks for phone) and source freshness. A serious provider reports a reachability rate, not just a coverage rate.

A B2B data provider collects, structures and updates information about people and companies, then resells it. Sources vary: public data, user-declared data, partnerships, purchased databases. Geographic coverage and freshness differ enormously between providers.

No provider is exhaustive: each is strong on a country, a segment or a data type. That is exactly why waterfall enrichment, which combines several providers, almost always beats a single source.

A pure player builds and runs its own database. The upside: consistency and control over the data. The downside: its blind spots are structural — contacts absent from its base will never be found, no matter how many queries you send.

Against a pure player, a waterfall approach multiplies coverage odds by querying several bases. For a niche market or a poorly covered country, the coverage-rate gap can be substantial.

Lead scoring combines fit criteria (industry, size, job title — the ICP) with behavioral signals (email opens, site visits, intent data) to rank prospects. The highest scores go to sales first.

Scoring is only reliable if the underlying data is: a wrong title or a misidentified company skews the score. Clean enrichment is therefore a prerequisite for useful scoring.

Most companies apply a single pattern to all their addresses: first.last@domain, f.last@domain, firstlast@domain… Knowing the pattern lets you infer an employee's address from their name and the company domain.

The format is inferred from actually observed addresses: the larger the sample, the more reliable the estimate. An inferred address is still a hypothesis — it must be verified before any send.

Permutation produces the list of plausible addresses — first.last, last.first, f.last, first@… — for a given contact. You then verify each candidate to keep only the one that actually exists.

It is a useful fallback when a company's format is unknown. It must always be followed by verification: sending to untested permutations hurts deliverability.

Email verification combines several checks: syntax, domain existence, MX record presence, and an SMTP dialogue with the server to probe the mailbox. It classifies the address as valid, invalid, or uncertain (the catch-all case).

Verifying before sending protects sender reputation: every bounce degrades the deliverability of all following sends. It is the cheapest step that avoids the most expensive damage.

Deliverability depends on the reputation of the sending domain and IP, authentication (SPF, DKIM, DMARC), list quality and content. A list full of invalid addresses drops reputation and pushes even good emails to spam.

It is an asset that builds slowly and breaks fast. Systematic address verification and a controlled sending pace are the two main levers to preserve it.

A hard bounce signals an invalid address or a nonexistent domain: it must be removed from the list immediately. A soft bounce is temporary (server unavailable, quota reached) and can be retried.

Mailbox providers watch the hard-bounce rate closely: above a few percent it collapses sender reputation. Pre-send verification exists precisely to keep it low.

On a catch-all domain, the server answers "yes" to any address. So SMTP cannot tell whether first.last@ truly exists: the probe always returns a positive, inconclusive result.

Catch-all addresses are classified "uncertain" rather than valid. To qualify them you must combine other signals (the address appearing in sources, consistency with the observed format) instead of trusting the SMTP test alone.

SMTP (Simple Mail Transfer Protocol) governs how email moves between servers. Email verification uses it without sending a message: it opens a conversation with the recipient server and asks whether a given address would be accepted.

The server's answer (accept, reject, catch-all) lets you classify the address. Some servers block this kind of probe, which is why SMTP must be combined with other validation methods.

A domain's MX (Mail eXchange) record points to its mail server. It is the first check in email verification: a domain with no valid MX can receive no message, so any address on it is invalid.

MX analysis also reveals the mail host (Google Workspace, Microsoft 365, etc.), useful information to tune the verification and sending strategy.

Role-based addresses point to no specific individual and are often read by several people or none. They are a poor fit for named prospecting and show low engagement rates.

Most verification tools flag them so you can exclude them from a cold campaign. They remain useful for inbound or support, not for reaching a specific decision-maker.

SPF declares which servers may send for a domain; DKIM cryptographically signs each email; DMARC tells receiving servers what to do if SPF or DKIM fails. Together they prove sender legitimacy.

Without these three records correctly configured, campaigns land in spam regardless of list quality. This is the technical foundation of any email prospecting.

A mobile number reaches a person directly; a landline often ends at a switchboard or a shared desk. For phone prospecting, mobile is far more valuable because it bypasses gatekeepers and reaches the decision-maker.

In France, the numbering plan deterministically distinguishes a mobile (06/07) from a landline. Quality enrichment does not deliver a switchboard number dressed up as a personal one.

Reverse lookup runs the opposite way to classic enrichment: instead of finding an email from a name, it starts from an email or a number to recover the person, their role and their company. Useful to qualify an inbound contact or identify a missed call.

Its reliability depends on the data existing in sources and on verifying the match. A match must rest on concordant signals to avoid attributing a contact to a namesake.

The HLR (Home Location Register) is the mobile operators' database. An HLR lookup checks that a number is assigned and reachable, and reveals the carrier and sometimes porting. It is, for phone, the equivalent of SMTP verification for email.

It lets you clean a list of numbers before a call or SMS campaign by removing inactive ones. It says nothing about the owner, only the line's status.

E.164 is a number's canonical format: a "+", the country code, then the national number, with no separators. It makes numbers comparable across countries and removes the ambiguity of the leading national 0.

Normalizing to E.164 is essential before any deduplication or automated dialing. A file mixing local and international formats produces duplicates and dialing errors.

The ICP defines who to target: industry, company size, geography, maturity, and the roles of your contacts. It focuses effort on the accounts where conversion rate and customer value are highest.

A precise ICP drives everything else: list building, enrichment criteria, scoring, messaging. Without it you prospect broadly and dilute sales effort.

A lead is a contact that has entered the sales radar: a filled form, an imported list, a targeted prospect. It is not yet qualified — you don't know if it has a need, a budget or decision power.

Qualification and enrichment turn a raw lead into an opportunity: complete its details, verify its role, assess ICP fit. A lead without reliable data stays unusable.

An MQL (Marketing Qualified Lead) has shown enough interest and fit to be handed to sales. An SQL (Sales Qualified Lead) is an MQL that sales has reviewed and deemed ready to work as an opportunity.

The MQL → SQL handoff is a classic friction point: it only flows if the lead's data is complete and reliable. Clean enrichment lowers the sales rejection rate.

Cold email reaches a cold prospect: no prior relationship. Its success rests on targeting relevance, personalization and deliverability — a verified list and an authenticated domain are prerequisites.

In B2B it often relies on legitimate interest rather than explicit consent, provided it stays relevant and offers an opt-out. Clean targeting is a compliance requirement too, not only an efficiency one.

Cold calling is still a powerful channel to reach decision-makers — provided you have the right number. A direct mobile multiplies contact odds; a switchboard wastes time in gatekeeping.

In France, B2C phone prospecting must respect Bloctel; B2B rules are looser but targeting relevance remains decisive. Number quality conditions the whole channel.

The SDR (Sales Development Representative) and BDR (Business Development Representative) own the top of the funnel: they work lists, send sequences, call and qualify, then pass opportunities to closing reps.

Their productivity depends directly on data quality: without valid emails and reachable numbers, an SDR spends their time hunting for contact details instead of selling. Enrichment automates that hunt.

A sequence combines several channels in a defined order: an email, then a LinkedIn visit, then a call, then a follow-up. Controlled repetition across channels markedly increases reply rates versus a single send.

Each channel needs the right data: verified email, mobile number, correct LinkedIn URL. A well-equipped sequence therefore assumes upstream multichannel enrichment.

Sales Navigator offers powerful search filters (role, industry, headcount, tenure) to build tightly targeted prospect lists. It is an excellent starting point to define an ICP and spot the right contacts.

However, it provides no directly exportable email or phone. Enrichment takes over: starting from identified profiles to find and verify their professional contact details.

Intent data captures interest signals: searches, content consumption, org changes, funding rounds, hiring. They indicate which accounts are "hot" at a given moment.

Combined with the ICP, it lets you contact the right company at the right time. It does not replace enrichment: once the priority account is identified, you still need to reach the right decision-maker.

ABM flips the mass-prospecting logic: instead of casting wide, you pick a small number of high-value accounts and orchestrate marketing and sales around them, with tailored messaging.

It requires deep knowledge of each account: org chart, decision-makers, contact details for several stakeholders. Account-level enrichment (all the key contacts at a company) is its fuel.

SIREN is a 9-digit number that uniquely identifies a company. SIRET (14 digits) adds a specific establishment code: one company has one SIREN and as many SIRETs as it has establishments.

These identifiers are the most reliable key to match company data unambiguously: unlike a trade name, a SIREN has no namesakes. They are indispensable to French firmographics.

The APE code (Activité Principale Exercée), drawn from the NAF classification, categorizes a company's main activity. It lets you filter a database by industry — for instance targeting every real-estate agency or every accounting firm.

It is a handy but imperfect filter: the code reflects declared activity, sometimes stale or too broad. It is often combined with other firmographic signals to refine industry targeting.

The Kbis is the "ID card" of a company registered in the trade register: name, SIREN, legal form, head office, directors, activity. It is the reference source to verify a company exists and who runs it.

For B2B enrichment, Kbis-type data provides reliable anchors — notably the identity of corporate officers — useful to target the decision-makers of an SMB.

The INPI hosts the National Company Register, the official source of legal information: company identity, directors, filings. It is a preferred source to identify a company's officers with certainty.

In French B2B enrichment, INPI data is used to reliably find an SMB's directors. You never look for the employees of corporate-officer legal entities: only individual directors are targeted.

Firmographics are to a company what demographics are to an individual: the criteria that characterize it — size, industry (NAF code), headcount, revenue, geography. It is the raw material of B2B targeting.

These attributes let you segment a database and filter by ICP. Their reliability rests on stable identifiers like the SIREN, the only way to resolve ambiguity between namesakes.

GDPR governs the processing of personal data in the European Union. In B2B prospecting it requires a legal basis (often legitimate interest), transparency, a right to object, and limiting data to what is necessary.

A professional email or phone is personal data: enriching and using it means respecting these principles. Compliance comes through relevant targeting, an easy right to object, and traceable sources.

Legitimate interest permits certain processing without explicit consent, provided the controller's interest does not override individuals' rights. In B2B, contacting a professional about a topic tied to their role can fall under this basis.

It is not a blank check: it requires relevant targeting (the message must concern the person's role), clear information and an easy way to object. Traceable enrichment makes that justification easier.

Opt-in requires explicit agreement before any send (the general B2C rule). Opt-out allows contacting and then offering an easy way to unsubscribe (often accepted in B2B on a legitimate-interest basis).

Whatever the regime, every message must include a clear unsubscribe mechanism and honor it immediately. Maintaining a suppression list is an obligation as much as a deliverability best practice.

Bloctel lists individuals who refuse phone solicitation. A cold-calling campaign toward consumers must scrub these numbers from its list; ignoring it risks penalties.

In pure B2B prospecting the Bloctel obligation does not apply the same way, but respecting an opt-out remains best practice. Distinguishing professional from personal numbers is therefore essential.

Deduplication spots records that refer to the same entity despite spelling variations: "Jean Dupont" and "J. Dupont", "SNCF" and "SNCF SA". Without it, a CRM pollutes quickly and skews every count.

The hard part is not wrongly merging two distinct namesakes. Good deduplication relies on several concordant signals (email, domain, company identifier) rather than the name alone.

Normalization puts everything in one format: phones in E.164, names in clean case, fixed accents, harmonized job titles. It is invisible but essential: without it deduplication fails and cross-source matching breaks.

It often includes fixing common source-file errors: swapped first/last name, broken encoding, junk characters. Quality enrichment normalizes before it acts.

The CRM (Customer Relationship Management) is the system of record for contacts and deals: HubSpot, Salesforce, Pipedrive… It tracks each prospect from first touch to signed deal.

Its value depends entirely on the quality of the data it holds. A CRM fed with invalid emails and duplicates becomes counterproductive; automated enrichment and deduplication keep it reliable.

An enrichment API exposes search and verification capabilities as automatable requests: you send a name and a domain, you receive a verified email and a phone. It wires enrichment directly into existing tools.

It enables on-the-fly enrichment (as a contact is created) rather than in batches. Charging per data actually found, not per call, is a key criterion to keep costs under control.

Before enriching a CSV, you must say which column holds the first name, last name, company, domain. A wrong mapping — swapped first/last name, a "company" column mistaken for "domain" — corrupts the entire enrichment.

A good tool auto-detects column roles and fixes ambiguous cases (an uppercase block identified as the last name, a URL column recognized as the website). It is a quiet but decisive step for result quality.

Namesake risk is central to enrichment: attributing one "Jean Martin"'s email to another "Jean Martin" delivers wrong data that looks correct. It is the most insidious error because it does not show up in coverage rates.

You contain it by requiring several concordant signals before delivering a data point: company, domain, role, photo match. A unique identifier (SIREN for companies) removes the ambiguity when available.