← Review playground Ops Board ↗
Reader doc

Sales Navigator: how we find and warm up clients (RME-178)

Not indexed. Rendered from 01-library/channels/sales-navigator-playbook.md. Highlight any line and use Hypothesis on the right edge to leave a note.

How we use Sales Navigator

Your job here. Read "The daily hour" and "Sales Navigator and our domain database", about 8 minutes. Skip everything else unless a question comes up. Your choice: approve, or comment on the page.

Sales Navigator is where YT finds the right people in her own network, keeps them in a few lists, and sees each morning who is worth a comment or a message. Warm first, offer second, one offer per message. YT's own messages do the selling. Ulinc reaches people at volume, and its future is YT's call (folded section on account risk).

What YT has, as of 2026-09-28. A Sales Navigator seat, the version without CSV export, plus LinkedIn Premium. The tier is still to confirm, so TeamLink and shared lists are not assumed. Vy has neither yet. YT plans to share two free months of Sales Navigator with her (her setup is in "Setting it up together"). Learn the buttons in LinkedIn's free Essentials course first, 45 minutes (the course research, which also holds every source for the rules and limits below).

The rule that comes before everything

Sales Navigator is hands-on only. Every Sales Navigator action on this page is a click YT makes herself, at a human pace. We never add a browser extension, a scraper, a lead exporter, auto-connect, auto-message, auto-visit, or any new tool that logs in as YT or acts as her account. They break LinkedIn's terms and they put her main profile, the one real audience we have, at risk. This playbook adds no tool.

The daily hour

YT, 2026-09-29: "we need to put ONE HOUR a day on this, lots of catch-up". So Sales Navigator gets a full hour on weekdays, fitted around the RME-122 posting routine (Post 5, Reply to comments 10, the DM reply desk 15).

How the two fit together. The morning starts with the routine's first two blocks, Post and Reply to comments, 15 minutes. Then the hour. The routine's 15-minute DM desk moves to the end of the hour and grows to 15 to 20 minutes, so anyone who answers a warm-up or a comment lands in the desk the same morning. The whole weekday is about 75 minutes.

MinutesBlockWhat YT doesWhere
0 to 151. Catch-upWork the backlog, oldest valuable first: threads where someone answered and never heard back (the plays library counts 825 of them, section X4), connection invitations pending for three weeks or more (withdraw them), and the lead lists (anyone who replied moves to L4).LinkedIn messages, My Network, lead lists
15 to 252. Saved searchesOpen the saved searches with new results, starting with SN-0. Look at each person, save three to five to the right list with a one-line note (why them, which offer). For a buyer's company, run the domain check below.Saved searches, lead lists
25 to 353. Warm signalsRead lead alerts for L1 and L2, then SN-0's new matches. Act on up to three: a job change gets a short congratulation, a profile view from someone on a list gets a friendly hello, a new follower of the company page gets a thank-you comment on their next post.Sales Navigator home, Alerts
35 to 454. CommentsThree to five comments on posts from L1 and L2, using the comment rules below.Their post, in LinkedIn
45 to 605. The DM deskThe routine's desk, in its order (interested people first). Then, if time is left, up to two first messages to warmed-up people who have now seen two or three comments, one offer each.LinkedIn messages

Catch-up shrinks as the backlog clears. While the backlog is large it takes the full 15 minutes. Once the unanswered threads are down to a handful, catch-up drops to 5 minutes and the spare 10 go to block 2.

Monday: block 2 runs 10 minutes longer, and block 1 gives it the time, because the weekend's saved search results arrive together. Friday: the routine's LinkedIn export, as today.

Stop at the hour. Whatever is left waits for tomorrow. Three good comments beat ten thin ones.

Ulinc webhook: a separate ticket the Content PM is raising. Nothing in this hour changes until it lands.

Setting it up together (a live session, about 20 minutes each)

YT: "you will help me set this up, and Vy too". So setup is a live session: YT (or Vy) shares a screen, Claude reads the next step aloud from this list, and they click. Claude never clicks in her account, and never in Vy's. Nothing happens unless the person whose seat it is is there and says yes to each step.

YT's seat, in this order:

  1. Check the tier (1 minute). Open a lead search. If a "TeamLink" filter appears under Spotlights, the

seat is Advanced and SN-8 can use it. If not, SN-8 uses "Connections of".

  1. Build the base filters once (3 minutes). Geography (the eight countries below), Connection 1st

degree. Save it as "Base", so every search starts from it.

  1. SN-0 Warm signals first (3 minutes). It is the one the daily hour opens every morning. Save it

and turn on its alert.

  1. SN-1, SN-2, SN-3, the buyers (5 minutes). Start each from Base, add the row's filters, save, turn on

alerts.

  1. SN-4 and SN-5, partners and referrers (3 minutes).
  2. SN-6, SN-7, SN-8, the signal and second-ring searches (3 minutes).
  3. The lists L1 to L5 (2 minutes). Empty, named exactly as below.

Vy's seat, once YT shares the two free months:

  1. Her own login, always. The seat is hers, on her own profile. She never logs in as YT (User

Agreement 2.2, no account sharing).

  1. Her 1st-degree network is small, so her searches use 2nd degree instead, with the same

filters. She builds SN-0, SN-4 and SN-6 first (warm signals, partners, people posting), then SN-1.

  1. Her lists are her own. Without Advanced, lists and searches are not shared between the two seats,

so she builds her own L1 to L5.

  1. One person, one lane. Anyone already on YT's lists or in a Ulinc sequence is not messaged from Vy's

seat. What Vy's seat does with the people it finds (hand them to YT, or warm them up herself) is YT's call, and it is not decided yet.

Which posts to comment on, and how

Comments are the warm-up. The aim is that by the time YT sends one message, the person already knows her name from two or three useful comments.

Comment on:

for email or CRM, or a complaint about the spam folder or bounces. These are where YT has a real layer to add.

Skip:

How to comment (the P14 shapes, used on someone else's post). The plays library P14 is written for replies under YT's own posts. The same four shapes work on theirs:

  1. They share an insight: agree with the core point, add one layer from practice. Never pitch.
  2. They ask a question: answer it straight first, then explain. Email facts only from the Email Almanac

snapshot (01-library/email-almanac/). If the Almanac does not cover it, do not answer it as a fact.

  1. They mention list or deliverability trouble: one useful, general observation from the Almanac, and

at most "happy to compare notes" if they want more. The offer waits for a DM.

  1. They are wrong: correct gently, with what we see in practice, never with authority. Never punch down.

Always: no links, no emojis, no selling in a comment. Written as YT, in her voice, two to four sentences.

The saved searches (set up once)

Each is a Lead search. Base filters for all of them: Geography: United Kingdom, United States, Canada, Netherlands, Germany, Spain, France, Brazil (the order our network runs in), and Connection: 1st degree unless the row says otherwise. Our audience is YT's network, not strangers (ICP). Company headcount is the filter that matters most, because size is what tells us a list is worth cleaning (the ICP's open item).

Saved searchFilters on top of the baseTheir one offer
SN-0 Warm signalsSeniority and headcount as SN-1, plus Spotlights: Viewed your profile recently OR Following your company OR Posted on LinkedIn in past 30 days. Add Changed jobs in past 90 days as a second saved copy if the list gets long.None yet. Comment or say hello first
SN-1 Marketing leadersFunction: Marketing. Seniority: Director, VP, CXO, plus Head of titles (Current job title: "Head of Marketing" OR "Head of Growth" OR "Marketing Director" OR "CMO" OR "Head of CRM" OR "Head of Retention"). Company headcount: 51 to 200 and up.Clean, via the app (reviewmyemails.com/get-started)
SN-2 Ecommerce foundersCurrent job title: "Founder" OR "Co-Founder" OR "CEO". Industry: Retail, Online and Mail Order Retail, Retail Apparel and Fashion, Consumer Goods (or Keywords: ecommerce OR Shopify OR DTC). Company headcount: 11 to 50, 51 to 200.Clean, via the app
SN-3 SaaS foundersCurrent job title: "Founder" OR "Co-Founder" OR "CEO". Industry: Software Development, or Keywords: SaaS. Company headcount: 11 to 50, 51 to 200.Clean, via the app
SN-4 PartnersCurrent job title includes fractional, consultant or agency, or Industry: Marketing Services, Advertising Services. Any headcount.Partner or resell, white label (reviewmyemails.com/partners)
SN-5 ReferrersSeniority: Owner, Partner. Company headcount: 1 to 10.15% referral (reviewmyemails.com/partners)
SN-6 Posted recentlySN-1, SN-2 or SN-3 plus Spotlight: Posted on LinkedIn in past 30 days.None yet. Comment first, message later
SN-7 Moved jobsSN-1 or SN-4 plus Spotlight: Changed jobs in past 90 days.Congratulate, then their one offer for the new seat, a week later
SN-8 Second ringSN-1 filters, 2nd degree, with "Connections of" set to a trusted 1st-degree contact. TeamLink only if YT's tier has it (see below).Ask the mutual contact for an intro, never a cold pitch

Save each one and turn on its alerts, so new matches arrive by themselves. Filter names and industry names move a little when LinkedIn updates the product. If one is not where this table says, the Essentials course shows where it went, and the nearest match is fine.

Why founders are a buyer search, not a referral search. SN-5 owners at 1 to 10 people have no list worth cleaning, so they get the referral offer (ICP finding). A founder of an 11 to 200 person shop or SaaS usually does send to a real list, and often owns it personally, so SN-2 and SN-3 get the clean offer.

Do these filters use engagement, or just facts about the person? Mostly facts. Every base filter (geography, connection degree, function, seniority, title, industry, headcount) is firmographic: it says who the person is and where they work, and nothing about whether they have noticed YT. Only the Spotlights carry signals, and they come in two kinds:

FilterKindWhat it tells us
Geography, function, seniority, job title, industry, company headcountFacts about the personFit only: could they buy, and which offer suits them
Connection degree, "Connections of"RelationshipHow close they are to YT, not whether they are paying attention
Posted on LinkedIn in past 30 daysTheir activityThey are active, so a comment will be seen. Not interest in us
Changed jobs in past 90 daysTheir activityA new seat, a natural reason to write
Viewed your profile recentlyEngagement with YTThey looked at her. The warmest filter we have
Following your companyEngagement with RMEThey chose to follow the company page
TeamLink (Advanced only)RelationshipA colleague knows them. Little use with one seat

So SN-0 Warm signals is the search that is actually built on engagement, and the daily hour reads it first. SN-1 to SN-5 find the right people, SN-0, SN-6 and SN-7 tell us when to talk to them. LinkedIn's Spotlights help page lists them.

Where each search came from. YT asked whether these came from research or because they are best. Honest answer: our own data picked the people, and judgement picked the exact filters.

SearchWhere it came fromSource or judgement
SN-1 Marketing leadersThe Ulinc exports: Heads of and Directors sit at real companies with real lists, the actual buyers (ICP, 9,804 contacts)Source for the people. Judgement for the exact title list and the 51-plus headcount
SN-2, SN-3 FoundersThe ICP's open item says size, not seniority, predicts a list worth cleaningJudgement, built on that ICP finding. Not measured yet
SN-4 PartnersUlinc segment C (VP and Manager) holds many fractional CMOs, and replied 67.6% (Ulinc analysis)Source
SN-5 ReferrersOwners reply most (71.5%) but run one-person shops with no list worth cleaning, so they get the referral offerSource
SN-0, SN-6, SN-7 SignalsThe opener naming one specific moment (moving to a new email tool) got a 36.1% reply rate, 3 to 4 times the generic opener (plays library, X4). A signal gives us that momentSource for "a specific moment wins". Judgement that these Spotlights are the best moments to use
SN-8 Second ringNothing measuredJudgement

What the vetted marketing skills added (Bruno's marketingskills repo at commit 5b2c000, read only, nothing installed, see the vet). Method only, no facts or numbers taken from it:

into firmographic fit, technographic fit and a buying signal ("why now?"). That is why SN-0 exists next to SN-1 to SN-5, and why the domain check below adds the technographic layer.

first message until the person has a signal or has seen two or three comments.

template (line 5) apart from an engagement template (line 55). We borrow the split, not the points.

opening and the message still makes sense, it is not working. Their "Trigger, Insight, Ask" shape (line 82) matches our job-change congratulation. skills/cold-email/references/personalization.md lines 23 to 25 call a specific, timely observation the gold standard, which is what our 36.1% says too.

LinkedIn scraping, the same as our rule. Its lead sources (Apollo, ZoomInfo, Clay) stay out, per the vet.

The lead lists (where people live once saved)

Save a person only after YT has looked at them. Five lists, named so the list says the offer:

Saving a lead is what makes Sales Navigator show their posts, job changes and news in YT's alerts. That is the whole trick: the lists decide what the morning feed shows her.

One offer per message (the offer ladder)

Every message carries one offer, matched to who they are, exactly as in the reply playbook. Never stack two.

WhoThe one offerThe play to start from
L1 buyers, once warmed upClean your list, first 1,000 free in the appP2 the trade
L3 owners and connectors15% referralP3 the referral brief, P6 the 15%
Anyone who writes about email, or says "collab"Collaborate, one concrete next stepP10 the collab yes
L2 agencies and fractional CMOsResell, white labelthe partner branch of the reply playbook

The first message after a warm-up names what they posted, not what we sell. The offer is the second sentence at the earliest.

Should the lists go to Instantly, or should we find leads in Instantly?

YT asked whether to export and connect to Instantly, or find leads in Instantly. Research only, no account was touched.

Exporting is off the table on this seat. It has no CSV export (folded section below), and an exporter extension would break LinkedIn's terms. So the choice is really between three routes:

RouteHow it worksGood forThe catch
A. Instantly's own lead finderInstantly calls it SuperSearch: a B2B database it says holds 610M+ leads, filtered by title, headcount, industry, location and technology, paid in Instantly credits (1 to 2 or more per verified work email) (help page)Cold email lists at volumeThese people have never seen YT on LinkedIn, so it is a cold channel, not a warm one
B. Copy company names by handYT copies a handful of company names from a Sales Navigator list, a few a dayChecking the best companies against our own dataSlow by design. A handful, never a sweep
C. Our own domain databaseThe workflow below. Start from the company, check its domain in our database, then decidePicking who is worth YT's time, with evidenceNeeds the DB check ticket first (proposal below)

The recommendation: keep LinkedIn and Instantly apart. LinkedIn stays the warm, hand-worked channel. If Instantly is used at all, it is for cold email, found with route A or C, never fed from Sales Navigator by hand-copying people.

Cold sending from Instantly is RED ZONE. Every campaign needs YT's explicit yes, one campaign at a time. And before any domain is written to, its DNS is re-checked live (MX, SPF, DMARC) and its website is loaded, because our cached data drifts. A "you have no DMARC" opener especially must be verified live first. Which Instantly plan YT has, and how many credits, is not in the repo.

Sales Navigator and our domain database

YT: "with the TLD/Hetzner DB and everything online we can be great". Sales Navigator tells us who the person is. Our own domain database (about 342 million domains) tells us whether their company really sends email and whether its setup has a visible problem. Put together, YT spends her warm-up time on the companies where a clean is most likely to matter.

This is a design, not a tool yet. Nobody queries the database from this playbook. Building the check is a separate ticket, proposed in 2026-09-29-sales-navigator-domain-db-check.md. Until it exists, step 2 below waits.

The workflow, one company at a time:

  1. Find the company in Sales Navigator. From a saved search, open the person's company and note its

website domain. No copying out in bulk, a handful a day inside block 2 of the daily hour.

  1. Check the domain in our database. It returns what we already hold: has MX (can it receive mail),

which ESP it sends with, SPF and DMARC present or missing, blocklisted or not, and its lead tier.

  1. Sort it into one of four buckets:
BucketWhat the database showsWhat YT does
TopHas MX, a real ESP in SPF, and a visible problem (no DMARC, a broken SPF, or a blocklist hit)Save to L1 with the problem in the note. Warm up first
GoodHas MX and a real ESP, no visible problemSave to L1. Warm up, the offer is a clean before their next big send
LaterHas MX but no ESP we can seeSave to L5 with a date. They may send through a subdomain we never scanned
SkipNo MX, or the domain is deadDo not save. Nothing to clean
  1. Warm them up on LinkedIn, exactly as the daily hour says: comments first, then one message with one

offer. The database finding never goes into a comment, and never into a first message as a cold "we scanned your domain". It tells YT who to spend time on and gives her the one specific moment to name once they are talking.

  1. Before any message names their setup, re-check it live. The cached record can be weeks old. A live

DNS look and a website load come first, every time. Any email fact in the message comes from the Almanac snapshot only.

  1. Prospeo or other free credits come last, and only when needed. On LinkedIn we do not need an email

address. A contact lookup only matters if the company later goes into a cold email campaign, which is RED ZONE (YT's yes per campaign). So a lookup is spent only on Top companies that already passed the live check, never on the whole list.

What stays out of git. Company names, domains and people found this way stay in LinkedIn and the local CRM, never in this repo, Slack or a public page.

TeamLink, InMail and alerts: what YT's seat can and cannot do

TeamLink shows which colleagues know a prospect. LinkedIn's help page says it is on Advanced and Advanced Plus only, while the compare-plans page lists it under Core. The help page is more specific, so it is not assumed until YT confirms her tier (one look: a "TeamLink" spotlight in a lead search means she has it). With one seat, TeamLink adds little anyway, because it only helps when there is a team. Until then, SN-8 uses the "Connections of" filter and Relationship Explorer, which do the same job for YT's own network.

InMail. Core gives 50 a month. Use them only for SN-1 to SN-3 buyers who are not connected and who YT has a real reason to write to (a post of theirs, a mutual contact). One offer, the reply playbook's rules, 40 to 110 words.

Alerts. Two kinds: saved-search alerts (new people who match) and lead alerts (posts, job changes and news from people on lists). The daily hour reads lead alerts first, because warm-up needs posts.

What the no-CSV version cannot do, and the workaround

LinkedIn's help page says it "currently doesn't offer the option to export account and lead information from Sales Navigator into a CSV or XLS file". Only Advanced Plus has CRM sync. So YT's seat cannot:

  • download a lead list or a saved search as a spreadsheet
  • push leads into a CRM in bulk
  • merge Sales Navigator lists with our own data automatically

The workaround, all by hand and all allowed:

  1. The lead list is the record for prospects. People stay in Sales Navigator, with a one-line note

each. Nothing is copied out in bulk.

  1. The light CRM gets one row per touch, not per person saved. When YT comments or messages, the

Content PM records it (the table in "What gets recorded", below).

  1. LinkedIn's own data export covers 1st-degree connections. Settings, Data privacy, Get a copy of

your data, Connections. It is the same weekly export the routine already asks for on Fridays, and it is LinkedIn's own tool, so it is safe. It does not include Sales Navigator lists or notes.

  1. Lists from our own database come in, not out. RME-174 (monthly account lists) is the future feed. A

CSV upload of accounts needs Advanced or Advanced Plus, so on a lower tier those accounts are typed in or searched by name.

Never fill the gap with an exporter extension or a scraper. Every "export Sales Navigator to CSV" guide of 2026 recommends one, and every one breaks LinkedIn's User Agreement section 8.2.

Setup, in order (done once)
  1. Account IQ, done 2026-09-28. YT entered the company description so Sales Navigator knows what we

sell and to whom. The approved text is in the Account IQ description. If the offer changes, YT approves a new version first, then pastes it in by hand.

  1. Take the Essentials course, 45 minutes.
  2. Personas. One per offer, using the saved search filters: a buyer persona (SN-1 to SN-3), a partner

persona (SN-4) and a referrer persona (SN-5). Personas shape which leads Sales Navigator recommends, so they should match our ICP, not widen it.

  1. Saved searches. Build SN-0 to SN-8 in the live session (above), save each, turn on alerts.
  2. Lead lists. Create L1 to L5, empty, ready for the first Monday review.

Coming later: RME-174. Monthly account lists from our own database will feed the saved searches. Nothing changes here until it lands.

Ulinc and Sales Navigator: which step lives where
StepToolWho does itWhat happens
1. FindSaved searches SN-1 to SN-8YT, by handNew matches arrive as alerts
2. ChooseLead lists L1 to L3YT, by handShe looks at each person and saves the right ones, one offer each
3. Reach at volumeUlinc, one automation with a branch per offer (clean for Heads and Directors, partner for VP and Manager, referral for Owners, as the ICP says)Ulinc, on YT's accountConnection request and one opener. Ulinc campaigns are usually started from a Sales Navigator search URL, and LinkedIn caps any one search at 2,500 people. How YT's campaigns are fed today is not in the repo
4. Warm upLead alerts on L1 and L2, and SN-0YT, by handThe daily hour
5. ReplyThe DM reply deskClaude drafts, YT sends by handReplies from Ulinc people and everyone else land in the same inbox
6. RecordThe light CRM (RME-172)The Content PM runs crm.pyUlinc's status maps to a stage (crm.py import-ulinc), Sales Navigator touches are recorded as below

One person, one lane. Someone YT is working by hand (L1 and L4) should not also be in a Ulinc sequence, or they get two voices from one person. Where the lead-list filter allows it, exclude L1 and L4 from any search that feeds Ulinc. Otherwise take them out of the campaign in Ulinc.

Who goes where. L1 buyers reply slowly and matter most, so they are worked by hand. L2 and L3 are the volume that Ulinc suits.

What gets recorded in the light CRM

The RME-172 CRM is one local row per person, names kept in a separate local file. When YT comments, congratulates or messages someone from Sales Navigator, the Content PM records it like any other touch:

CRM fieldWhat to put
sourcesalesnav plus the saved search, for example salesnav:SN-7, or ulinc plus the branch, for example ulinc:clean
channellinkedin-comment or linkedin-message
stageto-answer after a comment, sent after a message
playthe play used (for example P14 for a comment)
next step, next datefor example "check for a reply", seven days on
outcomefilled later, from the weekly export
The account risk, in plain words

LinkedIn's User Agreement (effective 3 Nov 2025, section 8.2) says members must not "use bots or other unauthorized automated methods to access the Services, add or download contacts, send or redirect messages". Ulinc does exactly that on YT's account, and RME-168 (PR #964) names it the biggest single risk on her main profile. LinkedIn says members who use such tools "risk having their accounts restricted or shut down". Caps and care lower the chance of a restriction. They do not make the use allowed.

LinkedIn gives no invitation number. It restricts invitations when many go out in a short time, when many are ignored, left pending or marked as spam, or when volume comes with suspected automation. A restriction usually lasts a week, up to a month when too many invitations sit pending. Automation vendors repeat about 100 invitations per rolling seven days and 15 to 25 a day. Treat that as folklore with a sales motive.

Options for YT, not a recommendation to drop Ulinc. They can be combined.

OptionWhat changesWhat it costs us
A. Cap the volumeInvitations from Ulinc plus by hand stay well under 100 a week, in business hours. Withdraw invitations pending for weeks. Re-push the ~5,253 people from the stopped campaigns in small weekly batches.Slower reach
B. Narrow what Ulinc doesUlinc sends the connection request and one opener only. Every follow-up is hand-sent from the reply desk.More of YT's minutes on replies
C. Split by valueL1 buyers are worked by hand from Sales Navigator. Ulinc keeps the partner and referral branches.Fewer automated buyer touches
D. Move Ulinc to another real profileUlinc runs on a teammate's own real profile, with their informed yes. Never a made-up persona and never YT's login on someone else's machine.Loses YT's reach, and moves the risk to that person
Vy before her seat arrives

Until YT shares the two free months, Vy has her own free LinkedIn profile, with no Sales Navigator and no Premium. What she can do now:

LinkedIn's free public guide, then sit in while YT takes the Essentials course.

  • From her own profile, by hand: follow the company page, comment on YT's and the company's posts, and

connect with people she genuinely knows.

  • Help draft comments when YT pastes a post in. She cannot see YT's lists.
  • Never log in as YT, in Sales Navigator, Ulinc or LinkedIn (User Agreement 2.2, no account sharing).
Phase 2: upgrade options (after launch, nothing bought now)

Prices checked 2026-09-28, USD. Decide once the app is officially live and both have done the learning.

SetupMonthlyWhat it gives the two of them
Vy on Core, YT as today$119.99 for Vy ($1,079.88 a year)Vy searches and keeps her own lists. No sharing between them
Both on Advanced, same contract$159.99 each, $319.98 together ($1,799.88 a year each)Shared lead lists and saved searches, TeamLink, team billing, account CSV upload
Advanced Plusprice not checkedCRM sync, the only built-in way to get lists out

Never, in plain words

and in the local CRM.