SQL (Sales Qualified Lead)

SQL stands for sales qualified lead. It has nothing to do with the database language of the same name, even if the two meet inside every CRM. An SQL is a lead that deserves real selling time.

What separates it from earlier stages is where the information comes from. A lead is spotted, an MQL is inferred from behavior, an SQL is confirmed by what the person says. They replied, they talked about their situation, and their words show a problem your offer can act on.

In practice an SQL almost always leads to a meeting: a discovery call, a demo, an audit. Many teams make an accepted meeting the official rule for becoming an SQL.

Frameworks differ, but the same questions keep coming back. Is the person in your target? Do they have a problem you solve? Can they decide, or at least influence the decision? Is the timing plausible, or are they describing a project two years out? Frameworks like BANT (budget, authority, need, timeline) formalize these questions.

You do not need every answer to qualify. A founder who writes 'we are actually struggling to hire salespeople, want to talk?' is an SQL even if you do not know their budget. On the other hand, a polite lead who answers 'interesting, send me a brochure' has confirmed no need at all.

Write down your criteria and your rejection reasons. A loosely defined SQL inflates your pipeline and skews every forecast.

The MQL showed interest in a topic, the SQL confirmed a need to a salesperson. In many CRMs the opportunity comes right after: it is an SQL with an amount, a likely close date and sales stages attached.

In outbound prospecting the MQL stage often disappears. You contact hand-picked people directly, and the first real qualification happens inside the conversation. The lead moves from 'contacted' to 'SQL' as soon as their reply reveals a need.

Three numbers are enough to steer: SQLs per week, the share of contacted leads who become SQLs, and the share of SQLs who sign. If the first goes up while the last goes down, your criteria have slipped.

Watch the delay between a first reply and qualification as well. According to MimikFlow's 2026 LinkedIn Prospecting Observatory (published in French), 22.0% of replies to first messages arrive within the hour and 48.6% within 24 hours. A reply handled two days later lets the momentum that could have become a meeting fade away.

Finally, compare by source. In the same study, 4.6% of strangers contacted after a search ended in an observed meeting, against 18.7% of people who had reacted to a post.

The first is counting everyone who replies as an SQL. A reply is not a need. 'Thanks, I will keep your details' is a politely closed door, and filing it as an SQL creates a pipeline that does not exist.

The second is qualifying on what you hope rather than on what the prospect said. If you cannot quote the sentence where they describe their problem, you do not have an SQL, you have a hunch. Keep that quote on the prospect's record: it will also help you prepare the call.

The third is leaving an SQL without a dated next step. A confirmed need with no meeting on the calendar often fades within days because the prospect has other fires. Offer a specific slot in the same message where you confirm you can help.

On LinkedIn, qualification happens in writing, message by message. The prospect answers your first message, you pick up on what they said, you ask one question about their situation, and a need shows up or it does not. The trap is turning the exchange into a questionnaire: five questions in a row kill the conversation.

In MimikFlow, the AI that answers prospects, the AI setter, runs this stage. You can give it up to eight qualification points. It reads the whole thread, never repeats a question the prospect already answered, asks at most one missing point per message and only offers a meeting once your blocking criteria are met.

The prospect moves to goal reached when the meeting is actually booked in your calendar, not when a link is sent. And when a question goes beyond what the AI can state with certainty, the conversation comes back to you with an alert instead of an invented answer.