LinkedIn prospecting is the work of finding the people who can actually buy what you sell, learning enough about them to say something worth reading, and getting all of it into your CRM the same day. Most teams do the first part well, the second part occasionally, and the third part never.
This is the long version of that job. It covers how to build an account list, how to find the buyers inside each account, what LinkedIn's search and invitation limits mean for your daily routine, how much research is enough, what to write, and the step that quietly wastes more time than all the others combined - moving what you found into HubSpot.
What LinkedIn prospecting actually is
LinkedIn prospecting is outbound research and outreach that uses LinkedIn as the data source. It is not lead generation in the marketing sense, where a form fills in and a lead lands in your queue. You choose the accounts, you choose the people, and you start the conversation.
There are four phases, and the order matters more than any single tactic inside them.
Teams that skip the first phase end up prospecting whoever happens to appear in a search. Teams that skip the last one rebuild the same research three weeks later, because nobody wrote it down.
Start with the account list, not the search bar
The search bar is the most expensive place to begin. It rewards whoever types the most generic filter, which is how a rep ends up staring at 4,000 results with no plan for any of them.
Write the account definition first. For most B2B teams it fits in four lines: industry, employee range, the technology or process that makes your product relevant, and a trigger that makes now the right moment. A real one reads like this: B2B software companies, 50 to 500 employees, running HubSpot, currently hiring SDRs.
Then name the accounts. Somewhere between 50 and 100 per rep per quarter works. Fewer, and a rep runs out of room to work. More, and the research gets thin, which defeats the whole exercise. If you would rather have the list scored than eyeballed, ICP Match ranks accounts against your ideal profile so the top of the list is not simply alphabetical.
One test before you move on. For every account on the list you should be able to finish this sentence out loud: we are working this account now because ____. If the blank stays empty, the account is not a prospect yet. It is a name.
Find the buyers inside each account
One contact per account is the most common and most expensive mistake in B2B prospecting. A software purchase worth a few thousand euros a year rarely rests with one person. Someone wants the change, someone signs for it, someone has to live with the tool every day, and at least one person is paid to say no to new vendors.
Job titles will not hand you these roles. Titles drift by company, by country, and by however recently the org chart was redrawn. Searching by department and seniority, then reading the profiles, works better: who reports to whom, who was promoted into a budget this year, who has been in the seat three months and needs an early win.
Aim for three to five named people per open opportunity, each with a role you can defend in a pipeline review. If you would rather have those roles classified for you than argued about on a Friday call, that is what buying committee mapping is for.
A worked example. You are selling a prospecting tool into a 200 person software company. The VP Sales is the economic buyer, because the budget line is hers. The SDR manager is the champion, because the pain is his team's. Two SDRs are the end users whose opinion decides whether the tool survives month two. RevOps is the blocker, not out of hostility, but because they own the CRM schema and nobody imports contacts into HubSpot without them. That is five people, and four of them are usually invisible to a rep who searched for "VP Sales" and stopped there.
Do this before the first message, not after the first stall. Mapping an account once the deal has gone quiet is damage control. Mapping it at the start is prospecting.
Know LinkedIn's limits before they interrupt you
Two limits shape any prospecting routine, and both tend to arrive in the middle of a good day.
The first is the commercial use limit. LinkedIn caps how much profile searching a free account can do each month, and the cap resets at midnight PST on the first of the month. Profile searches, browsing the People Also Viewed panel, and viewing members on the People tab of a company Page all count against it. Name searches from the top search box, browsing your own connections, and job searches do not. LinkedIn is blunt about the rest: it will not tell you how many searches are left, and will not lift the limit on request.
The second is the invitation limit. Connection invitations are capped, the cap applies to Basic and Premium accounts alike, and reaching it costs you roughly a week of sending. LinkedIn states that you cannot buy more invitations while restricted and cannot have the wait shortened, and withdrawing pending invitations does not clear it either.
Two habits follow from this. Search in batches and save the results, because every idle re-search spends quota you cannot see. And treat invitations as the scarce resource they are. An invitation spent on a profile you never researched is a week of sending you do not get back.
Sales Navigator raises the search ceiling and adds filters, which matters most for list building at volume. It does not fix the invitation limit, and it does not put your CRM data in front of you while you browse. Those are separate problems with separate answers.
Research that earns a reply
Ten minutes per account, not thirty. You are hunting for one specific, current, checkable reason to make contact. It usually hides in one of five places.
If your opening line would still make sense sent to a hundred other people, it is not research. It is a mail merge with extra steps.
Write the finding down in one sentence before you write the message. If you cannot compress it into a sentence, you have collected facts about a company rather than a reason to talk to a person.
Outreach: the request, the message, the follow-up
Keep the connection request short and specific. Name the trigger, say why you are reaching out, ask for nothing. Two or three sentences is plenty, and a pitch in the first message is the fastest way to be ignored politely.
A sequence that respects both the buyer and the invitation limit looks like this.
- Engage with something they published, so your name is not completely cold.
- Send the connection request with a two line note that references the trigger.
- Once they accept, wait a day, then ask one question they can answer in a sentence.
- Follow up twice, four or five days apart, adding something new each time.
- Stop. Move them to a nurture list and come back next quarter.
What that looks like in practice, for an SDR manager who posted three job openings last week:
Saw you are hiring three SDRs this quarter. Most teams that scale the desk that fast hit the same wall about six weeks in, which is CRM data nobody trusts. Curious how you are planning to handle it, because the answer is usually either a process or a tool and both have a cost. No pitch, genuinely interested.
It names the trigger, it says something specific about their situation, and the ask is an opinion rather than a meeting. It also gives them an easy way to reply in one line, which is the only bar a first message has to clear.
Run the same play on two or three people at the account rather than one. If your champion goes quiet, changes role, or leaves, a single thread means the opportunity goes quiet with them.
And keep the ask small. A first message that asks for 30 minutes is asking a stranger to spend more time on you than you have spent on them. A question they can answer while walking between meetings gets answered.
The step everyone skips: get it into the CRM the same day
This is where prospecting quietly falls apart. A rep spends a good morning building a list, has three useful conversations, and records none of it properly. The notes sit in a spreadsheet, the contact never reaches HubSpot, and on Thursday a colleague messages the same person with the same offer.
The obvious fix, HubSpot's native Sales Navigator integration, is out of reach for a lot of teams. It requires a Sales Navigator Advanced Plus account plus an assigned Sales Hub Professional or Enterprise seat. That is two paid upgrades before a single contact syncs.
The cheaper answer is a CRM overlay: your HubSpot record shown on the profile you are already looking at. Lifecycle stage, owner, open deals, last activity, with no second tab. From there the workflow is one click instead of ten. You can import the contact straight into HubSpot, add a verified email and phone number, and pull an entire search or saved list in bulk on the days you are list building rather than researching.
Two problems disappear in the first week. Duplicates stop, because the overlay shows a contact already exists before anyone creates a second record. Double outreach stops, because ownership is visible on the profile instead of buried in a report. Teams using LeadLx report saving 8 or more hours per rep per week, and most of that saving comes from exactly this part of the job, the part nobody puts on a slide.
Five mistakes that quietly kill a prospecting motion
A weekly rhythm that survives a busy quarter
Prospecting dies whenever it depends on motivation. Put it in blocks on the calendar and it survives the quarter.
- Monday, 45 minutes. Add and score accounts, then map three people at each new account.
- Tuesday to Thursday, 45 minutes a day. Research ten accounts, send the messages, log every one the same day.
- Friday, 20 minutes. Clean up. Close dead sequences, merge duplicates, check that every conversation this week has an owner and a next step in HubSpot.
That is under four hours a week. It is also more real prospecting than most teams manage in a month, because the work is bounded and it gets recorded while it is still fresh.
The four numbers worth tracking
Four numbers tell you whether any of this is working, and they diagnose different problems.
- Invitation acceptance rate. Low means your targeting or your note is wrong, not that LinkedIn is saturated.
- Reply rate on first messages. Low with high acceptance means the research is thin.
- Meetings per 100 contacts touched. The only number that connects the activity to pipeline.
- Named stakeholders per open opportunity. If the average is one, your forecast is a hope, not a plan.
Do not import someone else's benchmarks for these. Measure your own baseline for 30 days, then work on one number at a time. Fixing the acceptance rate and the reply rate at once tells you nothing about which change did the work.
Where to start this week
Pick 25 accounts that match the definition you wrote. Map three people at each. Research ten of those accounts properly and write to the people you found, with the trigger in the first line. Log every one in HubSpot the same day.
That is 75 mapped contacts and 10 researched accounts inside one week, which beats a month of unfocused searching. For the CRM half of this, which is where most of the lost time hides, keep an eye on the LeadLx blog as the rest of this series goes up.