Signal-named campaigns. Two dollars, never mixed.

What is a signal-named campaign in Nerxus One?

A signal-named campaign is a named pile of deals grouped by the detector you are selling, such as No-SSL Plumbers · Austin, not a folder of every pin you saved last Tuesday. Pulse treats that name as a claim about the offer. If the campaign says No-Website, the cards inside should still carry a no-website finding, or the board is costume.

The buyer is the marketer. The cafe, the HVAC shop, and the med spa are your customers. They never log into this board. They may later see a portal after Close. Until then the campaign is your book, isolated inside your Clerk organization, with no path into another tenant's deals.

LeadsGorilla-style tools stop at a campaign table: a name, a created date, a count, three buttons. Nerxus One keeps that table and adds a funnel, two dollars, freshness, and a follow-up surface. The campaign is still the landing page under Pipeline. The board is where you drag a card. All Leads is the master table across every campaign. Follow-ups is the clock.

A business is one row for life. A deal is that business in a campaign at a stage. The same shop can sit in No-Website and No-Reviews. Pulse will flag the double and offer a primary deal so you do not pitch the owner twice from two boards. Deleting a campaign deletes the grouping. The business remains.

Campaigns can be static lists you filled from a Scout save, or living rules that ingest new matching leads on a schedule you set. Living rules still spend Radar and Places budget through the shared cache. Pulse will not keep a rule looking alive by silently re-scanning the metro every hour. If the entitlement gate would trip, the inflow stops and the UI says so.

Groups and folders exist for vertical, city, signal family, or month. They are headers and a filter, not a second CRM. Archive hides a campaign you might reopen. Delete removes the grouping only. Export writes the enriched columns, including empty money when no diagnostic has run. Empty travels with the CSV so a later tool cannot treat a dash as zero.

How do DEV and FEE stay separate on a deal?

DEV is Deal Expected Value in dollars per month, the recovery Radar already modeled for the SMB. FEE is Deal Fee Estimate in dollars once, the project price you would charge to do the matching Forge work. Pulse stores both. It does not convert one into the other.

Radar emits monthly bands with assumptions still attached. Pipeline inherits those bands unchanged. A field that means monthly ends in Mo. A field that means one-time ends in Once. A bare number called value is treated as a defect, because TypeScript cannot tell a month from a job if you hide the unit in a nickname.

DEV is the sentence you say to the owner: you are leaving this much per month on the table. FEE is the sentence you say to yourself: this is what I invoice if they buy the site, the booking block, or the inbound voice setup. The slide-out and the Business-360 read those dollars from one picker. If two screens ever disagree, that is a bug, not a rounding story.

FEE comes from the fulfillment action ids on the findings, each carrying a typical one-time band for that fix. You can edit FEE on the deal because you know your own pricing. The public marketing site will not invent a price to decorate this paragraph. If Radar never ran, both figures show as empty, not as zero. Zero is a claim. Empty is a missing diagnostic.

Confidence rides along. Inferred pieces stay labeled est. Modeled versus observed stays labeled the way Radar labeled it. Pulse will not wash a modeled chair-hour band into a confident monthly fact so the card looks finished.

Why does the pipeline forecast sum FEE and not DEV?

The forecast sums FEE across open stages because FEE is your revenue and DEV is the customer's recovery. A column that reads like agency income and turns out to be the city's recovered rent is worse than a blank.

Each stage carries a win probability used only for weighting FEE. Defaults ship as New 0.05, Contacted 0.15, Demo 0.45, Closed 1.00, Lost 0.00. Weighted value on a deal is FEE times that probability. The board column header sums those weighted values. The campaign strip sums them across open stages. Closed is won history. Lost is a terminal column that feeds reactivation, not the forecast.

Owner-tunable weights exist in settings. Changing them recomputes the strip. Learned probabilities from your own close history are a later enhancement, not a v1 promise. Until then the defaults are defaults, printed as such.

Priority is a different number. Pulse multiplies size, opportunity score, freshness, a live reply clock, and close proximity, then subtracts staleness, so a smaller deal with a three-day-old permit and a reply can outrank a larger cold card. That rank decides what Follow-ups puts at the top. It does not replace the forecast.

When a number appears on the card and again in the slide-out, both reads go through the same money helper. Live recompute, then stored scalar, then empty. The product already shipped a defect where those two surfaces disagreed. The fix was one function, not a better tooltip.

Open FEE on a campaign row is the sum of open-deal fee estimates. Weighted FEE is that sum after stage probabilities. The summary strip chips are filters: Hot, Stale, Due today, Closing this week. Clicking Hot does not change a dollar. It hides the cards that are not on a clock. If a campaign has never been diagnosed, Open FEE and Weighted FEE stay empty with a run-diagnostics action into Radar.

Platform admin revenue views care about FEE for the same reason you do. An operator who summed DEV would be counting the city's recovered rent as agency income. The naming rule exists so a later chart cannot 'simplify' the unit off the axis.

What are the pipeline stages New, Contacted, Demo, Closed, and Lost?

The stages are New, then Contacted, then Demo, then Closed or Lost, and v1 ships that order as the funnel every other module can name. New is identified and untouched. Contacted means outreach has gone out. Demo means they have seen the sample or the report. Closed is won. Lost is terminal and requires a reason.

The board is those five columns. Each header shows deal count, FEE in stage, and weighted FEE. You drag a card to write a stage event. Entering Contacted can queue a follow-up task and start an inactivity timer. Entering Demo can prompt you to attach the live sample or the Atlas report if none is linked. Entering Closed offers convert-to-client. Entering Lost asks why: no response, not interested, no budget, bad timing, competitor, unqualified, or other.

A reply from Relay can move New to Contacted without your drag, and it starts the speed-to-lead clock on Follow-ups. Pulse does not send mail to cause that move. It listens for the outreach event and updates the deal.

Automations may draft and queue. They do not send. There is no stage trigger that fires cold SMS or cold voice. If you want a message to leave the tenant, you are in Relay, through ComplianceGate, on a verified domain.

All Leads shows the same stages as filters on one table, across every campaign. Saved starter views ship as Hot now, Biggest open FEE, Going stale, Untouched high-opportunity, Reactivation candidates, and Closing this week. Bulk stage moves that spend or send still hit the entitlement and compliance gates.

How is the Demo stage different from a Forge DEMO?

The Demo stage is a sales column that means the prospect has seen the working sample or the branded report. A Forge DEMO is the working sample itself, a live preview born from the fulfillment action on the finding. One is a place on the board. The other is an asset with a URL.

You can attach a DEMO while the card still sits in New or Contacted. That is the sample-first motion: I noticed this listing has no booking button, so I already built a DEMO, here is the live link. The stage stays Contacted until they have actually looked. Moving the card to Demo is your claim that they have.

DEPLOY is a later, gated action on that asset. DEMO is not a typo for a meeting. If a sentence on this site ever used Demo and DEMO as if they were the same word, treat that as a defect in the copy. Pulse will not advance a stage because Forge opened a preview. Forge will not rename a stage because you deployed.

Atlas can brand the memo that travels with the DEMO. The report is proof. The DEMO is the thing they can click. The Demo column is where you park the deal after that click or that meeting has happened. Keep the three nouns in those jobs and the week stays readable.

What does Pulse do on the pipeline, and what does it refuse?

Pulse is the pipeline specialist. It values deals, ages signals, runs the reply clock, ranks next actions, names campaigns, and re-surfaces lost rows when Watchtower reports a new hook. Nova routes pipeline talk to Pulse. Pulse does not re-run detectors and does not transmit mail.

Read-only work can run unattended: scoring, freshness, reminders, a draft sitting in the approval queue. Sends, bulk diagnostics, report batches, and Forge builds spend or send, so they pass entitlements and, for mail, ComplianceGate. The platform default Agent Mode is Supervised. Pulse inherits that.

Ask Nova who is going stale, and Pulse answers from the decay profiles. A permit is hot for a short window, then warm, then cold. A hiring post ages faster. A storm decays sharply. A rating drop is an event, not a countdown. Pulse colors the chip and puts the row on Going stale. It does not invent a fresh permit to make Monday look busy.

Inactivity staleness is a second clock, independent of signal age. A Contacted deal with no reply for a week can be stale even if the permit is still warm. Stage automations queue a follow-up task. They do not mail the owner to clear the chip. You clear the chip by working the task, or by a real Relay send you approved.

Re-scan cadence belongs to Radar's Watchtower, not to Pulse. Pulse consumes the change event. It does not spend a Places call to refresh a card you glanced at. If Watchtower is paused, freshness ages in place and the chip tells you the hook is old.

Next-best-action is a sentence with a reason: reply because the clock is four minutes old, pitch the remodel page because the permit is twelve days old, send the live sample link again because the card has sat in Demo, re-open because a lost-for-timing row just pulled a new permit. Every action that would send is a queued human step or an approval-gated Relay draft.

Campaign naming follows vertical, city, and signal. Pulse can propose a split when you saved a mixed pile, so No-Website, No-Reviews, and Hiring do not share one board. Living smart campaigns can ingest new matching leads on your schedule, capped by plan entitlements and the shared cache. Pulse will not silently burn Radar credits to keep a rule looking alive.

How does a saved lead become a deal and then a client?

Save in Scout, or a push from Radar, creates or joins a deal. The business uuid is the join. There is no second import when you open Pipeline, and there is no second import when you convert. Handoffs persist first and pass ?business=<uuid>.

If a diagnostic exists, Pulse computes DEV and FEE on enter-New. If it does not, the card shows empty money and a run-diagnostic action into Radar. Start Outreach enrolls the campaign, or a filtered subset, into Relay. Generate Report batches Atlas artifacts for proof. Build a fix opens Forge on the fulfillment action already stamped on the finding.

Business-360 is the deep record: the Radar memo, the Relay thread, the Atlas reports, the Forge assets, the activity timeline, notes, tags, and tasks. The slide-out is the same facts in a panel so you can stay on the board. Open in Pipeline from Scout, Radar, Inbox, or a report lands on this record.

Closed opens convert-to-client. Confirm writes the business as a client and hands services, diagnostics, and contacts to Steward. The deal stays visible as won history. Invoicing and the portal are offered from that handoff. You do not paste the shop into a new tool and lose the DEMO lineage.

Lost rows sit in a reactivation pool keyed by reason. A later Watchtower change can bump the same deal back to the top with a dated hook. Re-engagement is email-first and honors Relay's suppression list. A lost row does not earn a cold text.

What happens on the pipeline when a prospect replies?

A reply event from Relay starts a response SLA timer on the deal, jumps the card onto Follow-ups under Hot, and can move New to Contacted. The timer renders green under five minutes, amber from five to thirty, then red, matching the decay curve the module was built around. Thresholds are owner-configurable. The default target is five minutes.

The product cites the MIT Sloan / InsideSales 2007 Lead Response Management Study as the named anchor for that clock: odds of contacting a lead dropped hard from five minutes to thirty in that dataset. Other viral multipliers you have seen on vendor blogs are treated as inferred and are not printed here as fact.

The clock is your responsiveness, not the SMB's. Radar may have already sold a finding that the shop takes two days to answer its own form. That finding lives on the diagnostic. This timer is about you answering the person who answered you.

One-tap Reply opens the Inbox composer on that thread, with the diagnostic angle still attached. In Supervised mode Pulse can draft immediately and hold for approval. Autonomous still does not auto-send a first-touch. A reply is inbound, so a fast approved send is the compliant path.

Follow-ups also lists due tasks and the Going-stale lane. The header can jump you to the single highest-priority action across the book. If nothing is hot, the empty state can suggest high-opportunity untouched deals. It will not invent a reply to keep the lane from looking idle.

Who is this pipeline for, and who should keep a generic CRM?

This pipeline is for a solo marketer or a small team that sells websites, reputation, local SEO, inbound voice, or a retainer built from those jobs to local SMBs, and who already hunts by vertical, metro, and detector. If your week is No-Booking Dentists in one city and Permit-Lead Roofers in another, Pulse is the board for that week.

If you sell to VPs of Sales on LinkedIn, a people-graph CRM will feel more native. If you already closed the client and you need funnels, membership sites, and a staffed ops suite, GoHighLevel is the product that grew up for that book. Nerxus One starts before the client exists and refuses to drop the finding on the way in.

Agencies running many books care about isolation. Pulse runs inside a tenant. There is no path from one client's deals into another client's. v1 does not resell white-label sub-accounts. Tenant isolation is still the story you are buying.

A generic CRM will let you invent stages named Proposal and Negotiation and dump every vertical into one soup. You can do that elsewhere. Here the five stages and the signal-named campaign are how Relay, Atlas, Forge, and Steward keep pointing at the same row without a weekly export.

Common mistakes when running a local SMB pipeline

The first mistake is naming a campaign after your mood. Q3 Push is not a signal. No-SSL Plumbers · Austin is a signal. ICP and Scout already know how to hunt that. Pulse can keep the board honest only if the name still matches the cards.

The second mistake is treating DEV as the number you will deposit. DEV is the owner's monthly leak, modeled, with assumptions. Quote it in the memo. Forecast your week on FEE. If you pitch a one-time site fee by pointing at a monthly modeled leak, say both units out loud so nobody thinks you are invoicing the leak.

The third mistake is moving a card to Demo because you built a DEMO. Building is your work. Demo is their attention. Leave the card in Contacted until they have seen the link or sat the meeting. The column header's weighted FEE assumes that difference.

The fourth mistake is working the board while a reply sits red on Follow-ups. The research the clock is built on is about minutes. A beautiful kanban will not call the owner back.

The fifth mistake is closing into a new spreadsheet so fulfillment can start clean. Clean in this product is the same uuid with the diagnostic and the DEMO still attached. Steward is the clean start. A CSV is how the finding dies.

Get started on this site is the waitlist until access opens in batches. Login will point at the app when that URL is public. The board you want is the one that already refuses to mix the two dollars.

A sixth mistake is treating Living rules as a free metro firehose. Living campaigns still spend Radar and Places budget through the shared cache. If the entitlement gate would trip, inflow stops and the UI says so. A rule that looks alive while credits are dead is a costume, same as a campaign named No-Website that is full of five-star booked shops.

What does All Leads show that the board does not?

All Leads is the master table of every deal across every campaign, with the same five stages as filters, so you can sort a mixed book without opening ten boards. The board is one campaign's columns. All Leads is the pile when you need Biggest open FEE or Untouched high-opportunity without caring which detector named the row.

Saved starter views ship as Hot now, Biggest open FEE, Going stale, Untouched high-opportunity, Reactivation candidates, and Closing this week. Those views are filters, not new stages. Bulk stage moves that spend or send still hit the entitlement and compliance gates. A bulk drag that would enroll a hundred rows in Relay still waits for Supervised approval.

Groups and folders on the campaign table are headers and a filter, not a second CRM. Archive hides a campaign you might reopen. Delete removes the grouping only. The business remains. Export writes the enriched columns, including empty money when no diagnostic has run.

The slide-out and Business-360 are the same facts in two shapes. Open in Pipeline from Scout, Radar, Inbox, or a report lands on this record. If two screens ever disagree about FEE, DEV, or the clock, that is a bug. The product already shipped that defect on money. The fix was one helper, not a better tooltip.

Use All Leads when you are deciding who to work. Use the board when you are moving a named pile. Use Follow-ups when a clock is already running. Mixing those three jobs on one kanban is how a reply sits red while you tidy column headers.

How do living campaign rules ingest new leads without burning Radar?

Living rules ingest new matching leads on a schedule you set, and they still spend Radar and Places budget through the shared cache. Pulse will not keep a rule looking alive by silently re-scanning the metro every hour. If the entitlement gate would trip, the inflow stops and the UI says so.

Static lists are the ones you filled from a Scout save or a Radar push. Living lists are the ones that keep asking the same filter. Both still join on the business uuid. A shop that already sits in the campaign does not become a second card because the rule fired again.

Campaign naming follows vertical, city, and signal. Pulse can propose a split when you saved a mixed pile, so No-Website, No-Reviews, and Hiring do not share one board. A living rule aimed at No-SSL that then swallows every saved pin is a rule you misnamed, and the board will look like costume.

Watchtower is the freshness job, not the living-campaign job. Watchtower re-scans what you already saved and emits change events. Living rules hunt for new matching rows. If you collapse those two into one 'keep Austin hot' switch, you will burn credits and still miss the permit that aged on a card you already own.

Plan entitlements cap both. This page will not invent a credit price. If a launch card is disabled because the rule would overspend, read the banner. Disabled is accurate. A preview that looks enrolled while the cache is dry is the defect this product already refused on outreach launch cards.

How does convert-to-client keep the same business id?

Convert-to-client writes the same business uuid as a client and hands services, diagnostics, and contacts to Steward, so fulfillment does not start on a blank account. Closed opens that confirm. The deal stays visible as won history. You do not paste the shop into a new tool and lose the DEMO lineage.

Handoffs persist first and pass ?business=<uuid>. That is the same rule Scout uses when it saves a lead, and the same rule Inbox uses when it opens a thread. A provider id that dies when a key rotates is not the permalink. The uuid is.

Invoicing and the portal are offered from that handoff. The dentist sees a portal, not this board. Pipeline notes, health internals, and agency costs stay on your side. Isolation is the sentence the portal already lives by.

If you close into a spreadsheet so 'fulfillment can start clean,' you killed the finding. Clean in this product is the same uuid with the diagnostic and the DEMO still attached. Steward is the clean start. A CSV is how the working sample becomes FINAL-v7 in Drive.

A lost row does not convert. Lost sits in a reactivation pool keyed by reason. A later Watchtower change can bump the same deal back to the top with a dated hook. Re-engagement is email-first and honors Relay's suppression list. A lost row does not earn a cold text, and it does not earn a new uuid.

How does Pulse rank Follow-ups without replacing the forecast?

Follow-ups ranks next actions by priority, which multiplies size, opportunity score, freshness, a live reply clock, and close proximity, then subtracts staleness. That rank decides what sits at the top of the clock. It does not replace the forecast, which still sums weighted FEE across open stages.

A smaller deal with a three-day-old permit and a reply can outrank a larger cold card. The column header's weighted FEE can still prefer the larger card. Those two numbers answering two different questions is the product working. Collapsing them into one 'value' chip is how Monday's list starts lying.

The header can jump you to the single highest-priority action across the book. If nothing is hot, the empty state can suggest high-opportunity untouched deals. It will not invent a reply to keep the lane from looking idle. Empty Follow-ups plus a quiet Inbox is a Scout day, not a reason to mail a stale pile.

Inactivity staleness is a second clock, independent of signal age. A Contacted deal with no reply for a week can be stale even if the permit is still warm. Stage automations queue a follow-up task. They do not mail the owner to clear the chip. You clear the chip by working the task, or by a real Relay send you approved.

Reply clocks render green under five minutes, amber from five to thirty, then red, matching the decay the module was built around. Thresholds are owner-configurable. The default target is five minutes. The named research anchor is the MIT Sloan / InsideSales 2007 Lead Response Management Study. Other viral multipliers on vendor blogs stay unlabeled as fact here.

When should a lost deal come back onto the board?

A lost deal should come back when Watchtower reports a new dated hook that matches a reactivation reason you already stored, not when you feel optimistic on a Monday. Lost-for-timing plus a new permit is a sentence. Lost-for-not-interested plus the same old listing is a nuisance.

Entering Lost asks why: no response, not interested, no budget, bad timing, competitor, unqualified, or other. That reason keys the pool. Pulse can propose a re-open when the change event fits. You still drag, or you confirm. The classifier in Inbox will not do that drag because a sentence contained the word sure.

Re-engagement is email-first. Suppression still wins. A new permit does not wash an unsubscribe. If they said stop, the handle stays blocked on that channel. Scout's finding can still exist. You cannot mail that address.

Do not re-import the shop to 'start fresh.' The uuid already has the diagnostic, the DEMO, and the lost reason. A second import is how you pitch them twice and teach Pulse nothing. Primary-deal flags exist for doubles across campaigns. Use them.

Reactivation is not a storm playbook that already ran overnight. Governance is live. Autonomous dispatch is a separate half of Automation. This page will not tell you a lost roofer became a meeting while you slept. Supervised is how a careful marketer spends the first month of re-opens too.

How do Pulse and Steward split the same business row?

Pulse owns the deal while it is a hunt. Steward owns the client after Close. The business row is shared. Pipeline is not a CRM for the dentist. After they pay, the record does not start over, and the dentist still does not log into this board.

Services, diagnostics, and contacts carry over. The DEMO you showed becomes a lineage parent for the live asset when a human hits DEPLOY. Leftover findings become upsells on the same uuid: missing service pages open SEO, a collect-page gap opens Echo, a slow form opens conversion. You do not open a new hunt as if the shop were a stranger.

Health on the client side stays Healthy, Watch, or At-risk. Pulse priority is a different number for open deals. Do not paste a health band onto a New card and call it a forecast. Do not paste weighted FEE onto a QBR and call it recovered rent. Units stay on the field names.

v1 does not resell white-label sub-accounts. Tenant isolation still holds. There is no path from one client's deals into another client's. Agencies running many books are buying that boundary, not a franchise login they can hand to the shop.

If you only need a generic CRM with invented stages named Proposal and Negotiation, use that elsewhere. Here the five stages and the signal-named campaign are how Relay, Atlas, Forge, and Steward keep pointing at the same row without a weekly export.