Signal sources
Continuous monitors on the public signals that reveal your buyers.
Apify job-board scrapersG2 / AppExchange reviewstech-adoption + hiring feeds
August to October 2026 | Controllable Pipeline Generation
Not another hire. A running machine.
Prepared by ControlFlow for Girish Jashnani, Founder & CEO, Flosum · July 23, 2026
Terms and start dates held through July 31, 2026"The problem at Flosum is we have too many people. We don't have enough pipeline." Girish, first call. This program exists to retire that sentence.
Flosum is starting an outbound motion from zero while carrying live campaign deadlines. That combination requires a built system, not a hire: clean DNC and outreach-event logging, CRM synced to omnichannel sending, warm sending infrastructure, and an agentic layer that researches accounts, builds campaigns, monitors them, and kills what is not working.
ControlFlow runs that machine today. This proposal puts Flosum on it, with Jawad Haider personally owning the account and your team plugged in where it is genuinely faster: list inputs, compliance sign-off, brand assets, and the integration points into your Salesforce.
Everything here reflects our calls: your campaign sheet, your July timelines, your churned accounts, your competitive landscape, and a 20-rep sales team waiting on signals.
Before any contract existed, the machine did a morning of work on your market. You watched it happen:
| What you saw | What it proved |
|---|---|
| A live signal pull on your TAM: evaluation-phase job posts naming your competitors, a regulated health-IT firm hiring a Copado engineer, named operators publicly on record about painful Copado deployments | The signal engines find your buyers, with evidence links, on demand |
| The warmup-clock call: domains need 2 to 3 weeks before real volume, so start before the hire | You adopted it the same night |
| The churned-accounts-first sequencing, because they skip cold infrastructure entirely | Now the fastest line in your own plan |
That was one operator working by hand for a morning. The program runs it continuously, at volume, with your name on the results.
| Component | What it does | Status |
|---|---|---|
| Sending infrastructure | Dedicated domains and inboxes, automated warmup, rotation, auto-pause on placement dips, hard-bounce control under 2% | Live; Flosum capacity stands up during setup |
| Signal engines | Monitors competitor install-base signals: job posts naming your competitors or manual change-set processes, review sentiment, evaluation-phase accounts, Agentforce adoption | Live; demonstrated on your TAM before signing |
| Agentic campaign layer | Per-account research, campaign assembly, continuous monitoring, kill-switches on underperformers | Live |
| CRM discipline | DNC and suppression synced to your Salesforce so no active deal or open conversation ever gets a cold touch; every outreach event logged | Built during setup |
| Reply handoff | Positive replies routed to your team in real time with context, plus a weekly A-prospect phone list for your callers | Built during setup |
Your engineers own exactly one integration point: the sync into your Salesforce. Nobody at Flosum spends a quarter learning deliverability.
You asked to see the architecture. It is infrastructure, not a product with a UI, so here is how the machine is actually wired: signal in at the top, a booked meeting out at the bottom, and a feedback loop that sharpens it every week.
Continuous monitors on the public signals that reveal your buyers.
Every account enriched, researched, graded, and filtered before it is eligible to contact.
Your CRM is the source of truth. Open deals and active conversations are suppressed; every outreach event is logged back to it.
Warm infrastructure on email, plus LinkedIn from your reps' accounts. Rotation and auto-pause protect deliverability.
Every campaign watched continuously. Underperformers killed, winners scaled, no dead sends burning reputation.
Positive replies hit Slack with context in minutes, and a weekly A-prospect phone list goes to your callers. Your team books the meeting.
Stage 5 feeds stage 2. What dies tells the machine what to build next, so targeting and copy sharpen every week. That compounding is why an established engine beats a fresh build, and why month three beats month one.
| Week | What ships | What you see |
|---|---|---|
| Week 1 | Kickoff. Domains and inboxes purchased, warmup started. Churned-account list pulled from your Salesforce with suppression rules agreed. Churned-accounts campaign live by end of week; it does not wait on cold infrastructure. | Slack channel live. Launch notification. First replies from churned accounts. |
| Week 2 | Signal engines pointed at your TAM: competitor job-post monitors, review mining, evaluation-phase detection. Lists built and graded. Copy drafted for your review. | Written lead report #1. Copy for sign-off. |
| Week 3 | Warmup clears. Copado displacement campaign live. Second cold campaign (evaluation-phase accounts) live. | Launch notifications. First cold replies. |
| Week 4 | Read results, kill underperformers, scale what works. Dreamforce attendee campaign built and queued against the September 15 window. | Lead report with first funnel read. 30-day review call. |
Daily 15-minute standup with Girish during the three-week build, then the cadence moves to the weekly rhythm below.
Accounts are tiered by intent, and messaging intensity follows the tier. These are live examples pulled from your actual market during our evaluation, not hypotheticals.
| Level | Signal | Real example from your TAM | Messaging |
|---|---|---|---|
| L3 | Evaluation phase: job post names multiple DevOps tools, or public complaint about an incumbent | A JD listing Copado, Gearset, Jenkins and Azure DevOps as alternatives; named operators publicly on record about painful Copado deployments | Direct and immediate: the account is choosing right now |
| L2 | Incumbent friction: hiring dedicated engineers to run a competitor tool; new Agentforce adoption creating backup risk | A CMS-serving health-IT firm hiring a Copado engineer in July | Displacement or trigger angle, principal-led |
| L1 | Firmographic fit: 5,000+ employees, Salesforce install confirmed by 2-of-3 triangulation | Your upmarket ICP with the full blocking committee mapped: platform owner, release manager, vendor management, finance | Patient multi-touch, outcome-led |
This was your open question, so here is the full answer. A name becomes a lead your team should call only after it clears three checkpoints, in order.
| Checkpoint | Question it answers | How |
|---|---|---|
| 1. Fit | Right account, right person, before we ever send? | Every account and contact graded A/B/C/D against your ICP. Salesforce confirmed by 2-of-3 triangulation: job-post evidence, AppExchange footprint, technographic provider. Title matched to the buying committee; company above your size threshold. C and D grades are never contacted. |
| 2. Timing | Why now, and not six months ago? | The signal ladder above. Where a direct signal exists (an evaluation-phase job post, a public complaint about an incumbent) we use it. Where none exists, we infer from proxies (hiring patterns, tech adoption, review sentiment) plus an AI research read of the raw evidence. Signals older than 60 to 90 days expire, so nobody is worked on stale intent. |
| 3. Response | Real forward intent, or just a polite reply? | A reply alone is not qualified. Qualified means a genuine forward signal: asked to meet, asked for pricing or a demo, asked a product question, named a need or timeline, or routed us to the decision-maker. A warm-but-vague reply is an MQL and stays in nurture. A phone number in a signature is never, by itself, a qualifying signal. |
What reaches your India team is the output of all three: SQLs with the evidence attached, not raw MQLs to dig through. That is the difference from inbound, and it is why the 5-minute response SLA matters. These people just raised their hand.
Illustrative math at steady state, from week 4 onward, at your stated $50K average deal size:
At a 25-35% meeting-to-opportunity rate and $50K ASP, that is roughly $100K-150K of new pipeline per month at steady state, before the Dreamforce campaign and before LinkedIn volume from your rep accounts is layered in.
Copado alone has roughly 1,700 customers, and every one is a confirmed displacement target. Add the Gearset and AutoRABIT install base, the change-sets and DevOps Center graduation pool, and Salesforce's enterprise segment (over 16,000 accounts above 1,000 employees, 80% of the Fortune 500). Your ICP is deep. The only constraint is reaching those accounts with the right message at the right moment. That is the machine's entire job.
You asked this on the call and it deserves a clear answer. These are in-class benchmarks, not guarantees, which is exactly what the 60-day gate is for. Success is a ramp, and we reset the target together at every 30-day mark.
| Window | Target | What it proves |
|---|---|---|
| Days 1-30 ramp | First qualified meetings on the board. Infrastructure live and healthy, churned-accounts and first cold campaigns launched. Month one is a ramp, not a peak: infrastructure is still warming through the first two weeks. | The machine works and produces real SQLs |
| Days 31-60 build | Full volume across multiple campaigns, plus LinkedIn from your rep accounts. Qualified meetings climbing toward steady state as winners emerge. | The motion scales |
| Days 61-90+ steady | Roughly 10 to 15 qualified meetings a month as winning campaigns scale and LinkedIn compounds (the funnel above shows 8 to 9 from the core email motion alone; LinkedIn and multiple campaigns lift the total). At your $50K ASP, that is pipeline your 20 reps feel. | A repeatable engine, not a one-off push |
"Qualified" means the three-checkpoint definition above, not raw replies. And at every 30-day review we set the next window's target together, on the funnel data we actually have, not a number picked today.
| Rhythm | What happens | Your involvement |
|---|---|---|
| Shared Slack channel | Launch notifications, campaign status, hot-reply alerts in real time. The system reports continuously; you never wait for a meeting to know what is happening. | Read as it suits you; flag anything, anytime |
| Weekly lead report | Written report: sends, replies, positive replies, qualified leads, campaign-level funnel, kills and scales, next week's plan. | 5 minutes to read |
| Weekly strategy call | One call with Girish: review the report, decide priorities, unblock inputs. | 30 minutes |
| Daily standup | A written standup every working day, two minutes to read, on your schedule instead of another 8-to-11 call. A live 15-minute call daily through the three-week build; after that the written update continues and the live call moves to weekly. | 2 minutes daily; 15 min live during build |
| Reply handoff | Positive replies hit your Slack with full context within minutes. Your team works them; recommended SLA is 5 minutes on qualified replies. Weekly A-prospect phone list with numbers and signal context for your callers. | Your team responds and logs outcomes |
On daily contact specifically: you get a touchpoint every working day. The only thing that tapers after the build is the live call, and only because the written update plus live Slack will already show you more than a standup could. If at day 30 you still want a standing daily call, we scope and price it then, on real experience rather than a guess.
| ControlFlow owns | Flosum owns |
|---|---|
| Infrastructure, deliverability, and sending reputation | List inputs, compliance sign-off, brand assets |
| Signal engines, list building, grading, campaign builds | Salesforce access and the CRM integration point (your engineers) |
| Copy, sequencing, monitoring, kill/scale decisions (with your veto) | Reply handling, calls, and meetings (your reps and team) |
| Weekly reporting and the funnel numbers | Deal execution from meeting onward |
Volume and automation stay inside deliverability guardrails ControlFlow sets. Pushing past them requires written sign-off from both sides. This protects your domains and your brand, which outlive any single campaign.
At day 30 and day 60 we review together: ICP accuracy, signal performance, campaign mix, message performance, and pipeline against cost.
The day-60 gate is written into the agreement and cuts both ways: if the pipeline does not justify the cost, Flosum exits cleanly with 7 days notice, keeps all data, lists, and copy, and owes nothing further. Equally, if Flosum-side inputs block delivery, ControlFlow can invoke the same clean exit. No lock-in on either side until the numbers have earned it.
| Item | Terms |
|---|---|
| Retainer | $8,000 per month, flat. Not hourly: you are buying an outcome and a machine, not a timesheet. |
| One-time setup | $2,500: domains, inboxes, warmup, Salesforce integration, DNC and suppression build. |
| Tools and data | At cost, full transparency. Your existing ZoomInfo and Cognism plug in. Your data remains yours. |
| Payment | Setup + month one due at signing. Months thereafter invoiced in advance, net 7. |
| Agreement | One-page MSA with ControlFlow. 60-day mutual gate as above; thereafter month-to-month with 14 days notice. |
| IP | The system, tooling, and methods remain ControlFlow's. All Flosum data, lists, copy, and campaign results are Flosum's, exportable at any time. |
| Account ownership | Jawad Haider personally owns the Flosum account, with senior oversight from ControlFlow's second principal on strategy and QA. |
| Validity | These terms and the start dates below are held through July 31, 2026. After that, the calendar reflows around other commitments. |
Context for the math: one closed deal at your $50K ASP covers roughly half a year of this program. You said it yourself: no matter how much pipeline we generate, your 20 reps will consume it. Day-0 cost is $10,500. One deal returns it nearly five times over.
Your own hiring names the target: scaling from roughly $30M to $100M ARR. That is a pipeline problem before it is anything else. Priced against a $70M growth mandate, an engine that feeds it at $8K a month is a rounding error, with a 60-day gate so it either earns its place or exits clean.
| Action | Owner | When |
|---|---|---|
| Approve this proposal and sign the MSA | Girish | Day 0 |
| Setup + month one invoice paid; clock starts | Flosum finance | Day 0 |
| Kickoff standup; churned-account list criteria agreed | Jawad + Girish | Day 1, 11:00 AM PT |
| Salesforce access + suppression rules confirmed | Flosum ops | Day 2 |
| Domains purchased, warmup running | ControlFlow | Day 2 |
| Churned-accounts campaign live | ControlFlow | Week 1 |
| Cold campaigns live on warm infrastructure | ControlFlow | Week 3 |
| 30-day review | Both | Day 30 |
| Gate review: scale, adjust, or clean exit | Both | Day 60 |