Link Leads Blog · August 20, 2026
Most agents size a lead order by budget, not by goal — whatever's left in the marketing line that week. Flip it around: start from the revenue number and work backward through the funnel to a lead count. Here's the math, worked two ways with real numbers already published on this blog.
A normal funnel runs forward: leads in, dials or texts out, some contacts, some appointments, some closes, some revenue. Reverse it and it's just division instead of multiplication: pick the revenue number, divide by your average commission per deal to get deals needed, divide by your close rate to get appointments or contacts needed, divide by your contact rate to get leads needed. The formula doesn't change — what changes is which number you already know and which one you're solving for.
The catch is that "your close rate" doesn't exist until you've run a batch. Everyone's first pass at this math has to borrow someone else's funnel and scale it, then swap in real numbers once they have their own. Below are two funnels this blog has already published in full, scaled toward a $10,000 goal with the arithmetic shown at every step — treat both as starting models to test against, not guarantees.
The home page's own illustrative example prices 10,000 aged leads at $150, texting the full list at roughly $120, and lands 2 closed deals at an average $750 commission each — $1,500 in revenue, about 6x the $270 combined spend. Scale that example up 10x and it lines up almost exactly with the Growth monthly plan (100,000 leads/month for $1,000, a third cheaper per lead than one-time pricing):
That clears $10,000 with room to spare in the model — which is really a statement about the Starter plan too: at 50,000 leads/month ($500), the same linear scale lands around $7,500 in modeled revenue, short of the goal. The plan-sizing decision isn't "which is cheaper per lead," it's "which volume, run at this conversion rate, actually clears the number you're solving for." Run the free 100-row sample through your own texting sequence before committing to a monthly tier — the linear scale-up assumes your close rate holds flat as volume grows, which is an assumption, not a fact, and it's the first thing worth stress-testing.
The real week one agent published on this blog ran 500 calling leads ($500 at $1/lead) through 1,636 dials to 80 booked appointments, 37 shows, and 12 closed deals worth $14,827 in annualized premium (ALP). Scale that whole funnel 10x and it lines up with the largest single order the calling-lead product allows — 5,000 leads, priced at $4,375 after the 12.5% volume discount that kicks in at the top of the ladder:
ALP is not revenue in your pocket — it's annualized premium, and a commission is a percentage of it that depends entirely on your contract, product mix, and carrier. That's a deliberate gap in this model: plug your own commission rate into $148,270 of ALP and you'll know within a few minutes whether 5,000 leads clears a $10,000 goal for you, because the answer genuinely depends on numbers only your contract has. What the model does tell you regardless of contract level: 120 modeled closes off one $4,375 order is the kind of ratio that makes the question "how many leads" answerable in leads, not just dollars.
Both paths above pick a clean multiple of a real, already-observed batch instead of solving the algebra directly to a $10,000 target. That's on purpose. Solving directly — dividing $10,000 by a revenue-per-lead ratio — produces a lead count with three decimal places of false precision, built on one data point that was never meant to predict anyone else's month. Scaling a real batch by a round number keeps the fact that this is a model, not a forecast, visible in the math itself. Use whichever framing gets you to an order size you're willing to test — the round numbers exist to keep you honest about how much the answer is still a guess.
Both paths above are scaled from single real batches — one agent's week, one illustrative example — not an average across buyers, and results on aged data move with state mix, age band, script, and speed to lead. The way to replace "modeled" with "known" is to run one batch at test size, log it, and let your own conversion rates replace the borrowed ones. The cost-per-contact spreadsheet is the tracker that turns a first batch into real numbers, and the ROI calculator lets you plug in your own close rate and commission instead of anyone else's the moment you have them.
Start smaller than either path above. Pull the free 100-row sample to see the fields, then size a first order — SMS or calling — against a goal you can actually check against real results inside a week or two, not a projection four scales removed from anything you've run yourself. Code PACT takes 20% off a first order at the order builder.