Funnel Planner and ROI Modeling for Agencies — SoloCMO
Ads 03 · plan

Do the math before you spend the money.

Drag the stages onto a canvas, wire them together, set the rates — and watch visitors, revenue and return cascade through the whole funnel before a single dollar moves.

See pricing The Ads system
The problem

Most ad budgets are set by feel and defended afterwards.

Two thousand a month sounds reasonable, so two thousand a month it is. Nobody works out what that has to produce at each stage to be worth doing, because doing it properly means a spreadsheet nobody wants to build.

Then results come in thin and the argument starts, with neither side able to say whether the plan was wrong or the execution was — because there was never a plan precise enough to be wrong.

The cascade

Set the rate at each stage. Watch what survives.

Every stage takes what the one before it produced and passes on a percentage. Change any number and everything downstream moves — which makes the fragile stage obvious long before it costs anything.

1 Paid trafficFacebook · $2,000/mo at $0.80 CPC — 2,500
↓Landing page converts at 6%
2 Opt-in pageFree in-home assessment 6% 150
↓40% reachable and qualified
3 Qualified leadsAnswered and fits the service area 40% 60
↓35% close
4 Booked jobsCalendar 35% 21
↓$850 average job value
$ Monthly revenue21 jobs × $850 — $17,850
$2,000Monthly spend
$13.33Cost per lead
$95.24Cost per booked job
$850Average order value
$15,850Monthly profit
8.9xROAS

Illustrative model for a local home care client. Every figure here is an assumption you can argue with — which is the point.

Scenarios

Build the version where it doesn't work.

Two tabs, same funnel, different assumptions. The optimistic model is the one you'd like to present; the conservative one is the model that tells you whether the campaign survives a landing page that converts at three percent instead of six.

Present both and you're the only person in the room who has thought about it honestly.

Scenario A — conservative

Page converts at 3%

Leads
75
Booked
10
Revenue
$8,500
ROAS
4.3x

Scenario B — target

Page converts at 6%

Leads
150
Booked
21
Revenue
$17,850
ROAS
8.9x

One assumption changed. Both models still profitable — which is the answer you want before committing, not after.

Seventeen elements

Not just opt-in and thank you.

Traffic sources, a dozen page types and three revenue steps, dragged onto the canvas and wired in whatever order the funnel actually runs.

Which matters because a webinar funnel, an application funnel and a straight lead capture have completely different arithmetic — and modelling all three the same way is how agencies end up surprised.

  • Multiple traffic sources feeding one funnel, each with its own volume
  • Branching paths — upsell and downsell running from the same step
  • Products and expenses set separately, so profit isn't just revenue
  • Undo, redo, zoom and save on a full canvas
  • Every funnel kept and reusable as a template for the next client
Traffic 2 sources
  • Traffic entry
  • Retargeting
Pages 12 types
  • Opt-in
  • Sales
  • Content
  • Survey
  • Application
  • Calendar
  • Webinar reg
  • Webinar live
  • Replay
  • Thank you
  • Chatbot
  • Custom
Revenue 3 steps
  • Order form
  • Upsell
  • Downsell
Where it earns its keep

It's a planning tool. It sells like a proposal.

A budget request backed by a model showing what has to be true is a different conversation from a budget request backed by confidence. And it changes what happens three months later.

At the pitch

Justify the number

Show the client what their spend has to produce at each stage. Nobody argues with arithmetic they watched you build.

At the start

Agree the targets

The model states the conversion rate every stage needs. That's a shared expectation rather than a promise you'll be held to vaguely.

At the review

Point at the stage

When results come in under, the model shows which assumption was wrong — and the conversation is about a landing page instead of about your competence.

Every module is on every plan. Plans differ only in how many clients you run.

See pricing
Shared context

What feeds it, and what it feeds.

Feeds the planner

Keyword Value
Job value
Offer
The entry point
Ad Library
Real CTR & CPC

The planner feeds

Funnel Diagnostics
The targets
Meta Ads
What to build
Content Report
Plan vs actual
Nearby modules

What operators run alongside it.

Funnel Diagnostics

The same stages, measured against what the model predicted.

 Meta Ads

The creative that fills the top of the funnel you just planned.

 Ad Library

Logged click-through and cost figures worth modelling from.

 Keyword Value

Average job value and close rate, already worked out.
Questions

Funnel Planner, answered straight.

What does the planner actually calculate?
Visitors cascading through each stage at the conversion rate you set, ending in leads, revenue, profit and return. It also gives cost per acquisition, average order value and ROAS, so the unit economics are on screen rather than in someone's head.
Aren't the conversion rates just guesses?
At the start, yes — and that's the point. The model shows what each stage would have to achieve for the campaign to pay, which turns a guess into a testable assumption. If it needs a landing page converting at twelve percent, you've learned something before spending anything.
Can I compare different versions?
Scenarios sit side by side on their own tabs, so a conservative model and an optimistic one can be built from the same funnel and compared. That's usually the honest way to present a budget to a client.
What kinds of funnel can it build?
Traffic sources, a dozen page types — opt-in, sales, content, survey, application, calendar, the three webinar stages and more — and revenue steps including order form, upsell and downsell. Simple lead capture through to a full webinar or application funnel.
Is it useful for selling the work?
That's often where it earns its keep. A budget request backed by a model showing what has to be true is a different conversation from a budget request backed by confidence — and when a stage underperforms later, you both already agreed what it needed to hit.
Get started

Every plan includes every module.

There are no feature gates. Plans differ only in how many clients you run and how much you generate for them.

See pricing The Ads system