The 6-phase outreach workflow that fills a solo agency pipeline

A solo agency pipeline works when every phase has a fixed time budget and nothing is allowed to exceed it: 10 minutes to prospect, 15 to research, 5 to write, 5 to follow up, an hour for the proposal call, an hour to onboard. At 30 prospects a week that produced 4 to 6 meetings and 1 to 2 closed engagements for me, and it did so without a CRM that needed maintaining.

I ran ModernWebSEO as a solo agency for the better part of two years. The pipeline ran on a six-phase outreach workflow that I refined to the point where it works without me thinking about it. This is that workflow, with real numbers.

The system is built for one person on one channel. It does not scale to a team without significant rework. For a solo founder doing services or productized services it is the cleanest pipeline I have run.

The six phases at a glance

The six-phase outreach workflow moves a lead through prospect, research, first-touch email, soft follow-up, proposal, and onboarding, with fixed time budgets of 10, 15, 5, 5, 30-60, and 60 minutes per phase. Run end to end, 100 prospects entering phase 1 produce roughly 5 closed engagements.

The funnel math from this: 100 prospects in phase 1, ~80 qualified, ~12 first replies, ~6 second replies from follow-up, ~18 active conversations, ~9 proposal opportunities, ~5 closed engagements.

In practice I ran 30 prospects per week, which produced 4-6 meetings and 1-2 closed deals per week. At my agency price point ($1500-4000 per engagement) that was a sustainable rate without burnout.

Phase 1: Prospect (10 min per lead)

The prospect phase of the outreach workflow is identifying who to contact, on a budget of 10 minutes per lead. The output is a qualified prospect matching the ICP, drawn from inbound interest signals, referral introductions, or outbound discovery. Roughly 80% of prospects survive this filter and move to research. The three sources, in order of conversion quality:

• Inbound interest signal. Someone followed me, replied to a tweet, or downloaded a free resource. They self-identified as the right audience. 30-40% of my weekly leads come from here.
• Referral introductions. An existing client introduces me. ~10-15% of leads. Highest close rate.
• Outbound discovery. I search for companies matching the ICP (industry, size, tech stack, recent funding signal). ~50% of leads. Lowest per-lead conversion but the only one that scales beyond my existing network.

For outbound, the ICP for SEO services was: companies between $1M-$10M revenue, in software/B2B SaaS, with a published blog that had not been updated in 4+ weeks, on a marketing team of 1-3 people. This filter alone eliminated 95% of false positives.

Tools used for phase 1: a saved Twitter list, Crunchbase for funding signal, a free LinkedIn search, a simple Python script that pulls company blog last-modified dates. Total tooling cost: zero. Total time per qualified lead: about 10 minutes.

Phase 2: Research (15 min per lead)

The research phase of the outreach workflow builds the personalized hook for the first email, on a budget of 15 minutes per lead. The output is a hook specific enough to prove the email is not a blast template. Each hook needs three components:

• Recent specific event. A funding round, product launch, hiring spree, conference talk, recent blog post the founder wrote. Something that happened in the last 60 days that I can reference.
• One concrete problem. A specific weakness in their current marketing presence I could improve. "Your blog ranks for X but not Y" or "Your meta descriptions are missing on these three pages".
• One personal connection. Something we share. Mutual contact, common tool, similar past project. Real, not invented.

The output of phase 2 is two or three sentences I can drop into the first email that proves I am not blasting templates. If I cannot produce these three things in 15 minutes, I drop the lead. The hook is non-negotiable.

Phase 3: First-touch email (5 min per lead)

The first-touch email phase of the outreach workflow is one short, personalized cold email, written in 5 minutes per lead because the hook comes from the research phase. The output is a reply, not a sale, and across the funnel this email converts 10-15% of qualified prospects into conversations.

The first email is short. Three to five sentences. No attachments. No links to a portfolio. Subject line under 60 characters.

The template I used most often:

Subject: Quick question about [their blog / their last launch / their hiring page]

Hi [first name],

I noticed [specific thing from phase 2 research]. [One sentence explaining why it caught my eye, ideally with a specific observation, not a compliment].

I help small SaaS teams with [one-line description of service]. If [specific outcome that addresses their situation] is interesting, I would love to chat for 15 minutes, no pitch, just to understand what you are working on.

[Name]

Reply rate from this template across about 800 sends in 2024-2025: 11.2%. The variance was mostly driven by phase 2 research quality. When the "specific thing" was genuinely specific, reply rate climbed to ~16%. When it was a generic compliment, it dropped to ~4%.

The 15-minute call ask is deliberate. "30 minutes for a discovery call" gets fewer accepts. "Just to chat" gets quick yeses that flake. "15 minutes, no pitch" lands the right buyers.

Phase 4: Soft follow-up (5 min per lead)

The soft follow-up phase of the outreach workflow is one two-sentence bump email sent seven days after the first touch, on a budget of 5 minutes per lead. The output is a reactivated thread: the follow-up converted roughly 30% of non-responders into replies. I consider it the single most undervalued phase in cold outreach.

The template:

Subject: Re: [original subject]

Hi [first name],

Just bumping this in case it got buried. If now is not the right time, no worries, happy to circle back next quarter.

[Name]

Two sentences. Reactivates the thread without guilt-tripping. Closes the loop politely if the answer is no.

Across the same 800 first-touch sample, the follow-up converted ~30% of phase 3 non-responders into replies. Not all positive (some were polite declines), but the conversion-to-call rate of phase 4 replies was about the same as phase 3, in the 35-40% range.

Two-touch outreach beats single-touch by roughly 40% in total response rate. Three-touch outreach beats two-touch by another 5-10% but burns goodwill. I stop at two.

Phase 5: Proposal call + document (30-60 min per lead)

The proposal phase of the outreach workflow is a 30 to 60 minute call plus a one-page written proposal, with the goal of getting to a clear yes or no. Proposals close at 40-60%, and the strongest predictor of closing is whether the prospect gave a concrete budget range on the call.

The 15-minute call expands to 30 if it goes well. The agenda:

1. What are you trying to accomplish? (10 min, mostly listening)
2. What have you tried? (5 min, identifies the dead ends I should not propose)
3. What does success look like in 90 days? (5 min, gives me the success metric for the proposal)
4. Is there a budget range you have in mind? (5 min; an explicit ask saves both parties from a bad-fit proposal)
5. Next step (5 min; I propose to send a written proposal within 48 hours)

The written proposal is one page. Three sections: situation summary, proposed scope (3-5 line items), price + timeline. Sent as a Notion link with view permission, not a PDF.

Proposals close at 40-60%. The variance is driven by the budget-range question in the call. When the answer was concrete ("we have $3-5k for this"), close rate was ~70%. When it was hedgy ("we are flexible"), close rate dropped to ~30%. The hedgy answers signal the budget is below my floor and the proposal will get sticker shock.

Phase 6: Onboarding (60 min per lead)

The onboarding phase of the outreach workflow converts a closed deal into an active client, on a budget of 60 minutes per lead. Each engagement gets a one-hour kickoff call, a shared Notion and Slack workspace, a weekly written report every Friday, and a 30-day check-in calendar invite.

About 95% of closed engagements complete the onboarding and become active clients. The 5% who do not are usually buyer's remorse before any work happens, so I refund without argument and we both move on.

The tools used for the whole pipeline

The full outreach pipeline at ModernWebSEO ran on six tools: Google Sheets as the CRM, Gmail with saved replies, a simple Python script for ICP filtering, Cal.com for scheduling, Notion for proposals and client workspaces, and Stripe for invoicing. Total spend was about $50 per month. The stack, in detail:

• Google Sheets for the CRM. One row per prospect, six timestamp columns, status enum, notes field.
• Gmail plus a saved-replies folder for outreach. No automation tool.
• A short Python script for phase 1 ICP filtering. Pulls company blog last-modified dates from a list of domains.
• Cal.com or Calendly for call scheduling.
• Notion for proposals and the ongoing client workspace.
• An invoicing tool once the engagement starts.

Total tooling spend: about $50/month including Stripe transaction fees on incoming payments. The cost is negligible because the system is built around one person doing focused work, not around tools doing it for them.

Why these reply rates look impossible

The 10-15% first-touch reply rate above is three to five times the published benchmarks, and I want to be precise about why rather than let it read as a claim about email copy.

Published 2026 benchmarks put average cold email reply rates around 3.1% (Cleanlist), with 3-5% described as a realistic target for a well-run campaign and 8-12% as top-performer territory (Apollo). The trend line is also going the wrong way, with reported averages falling year on year as volume rises.

Those benchmarks measure a different activity. They are averages over campaigns sending hundreds or thousands of emails, where research per lead is measured in seconds. This workflow spends 25 minutes per lead before the email is written, and sends 30 a week. Same word, different job.

That is the actual trade, and it is not flattering to either side. Volume outreach buys reach and pays for it in reply rate. Researched outreach buys reply rate and pays for it in ceiling: 30 a week is the number one person can sustain, and no amount of discipline turns it into 300. Anyone promising both is selling software.

Which means the honest question is not how to raise your reply rate. It is which constraint you are actually up against. If you cannot find 30 qualified prospects a week, volume tooling will not save you, because the ICP is the problem. If you can find 3,000 and cannot reach them, research per lead is the wrong lever and you should be running the other playbook.

Where this workflow breaks

The six-phase outreach workflow is not a growth-hack playbook. The reply rates above are unremarkable and the conversion math is the same math every B2B services business runs. What makes the system work for a solo operator is the discipline of the time budget per phase: no 45-minute research sessions, no 7-email sequences, no CRM that needs maintenance.

The leverage is in cutting the workflow down to what one person can actually do for 30 leads a week, indefinitely, without dreading Monday.

For a system that automates the parts that can be automated (ICP filtering, email drafting from templates, CRM pipeline tracking), the B2B Growth Machine is what I built to handle the parts I no longer wanted to do by hand. For the broader agency context (services pricing, retainer transition, onboarding playbooks), AI Agency Launch Kit is the wider system this workflow fits inside.

Part of the solo founder series. The conversion math behind the funnel is in the CRO experiment post. The pricing decisions for service engagements use the same Van Westendorp framework the digital products do.