[drop_cap]T[/drop_cap]he prospect who filled out your gift’s opt-in form is warm, not qualified. Warm means they’re curious enough to trade an email for a teardown. Qualified means fit, budget, authority, and timing all line up well enough that a call between you two is worth both people’s time — and the only way to know that before the call is to ask, not to find out forty-five minutes into a discovery conversation that this deal was never real.
This is Stage 3 of the Attract band, and it’s the stage that protects everything downstream of it. Every hour spent on a call with someone who can’t sign, doesn’t have budget, or needs the work done by a date you can’t hit is an hour not spent on a deal you can actually close — and worse, it’s an hour that trains you to expect low-quality calls, which quietly changes how you show up on the good ones too.
The Deal You Decline Protects the Margin on the Ones You Take
Every agency founder has a story about the client who seemed fine on the call and turned into a scope-creep nightmare that ate the margin on three good projects trying to keep them satisfied. Almost always, the warning signs were visible before the call — in the intake form, in the way they answered the budget question, in who they said would be involved in the decision. The discovery call didn’t reveal the problem; it just delayed noticing it by two weeks and a proposal.
Qualification isn’t a filter you apply to be difficult. It’s the mechanism that makes your average client good instead of average. An agency that takes every inbound lead to a call has a distribution of clients — some great, some fine, some disasters — and the disasters cost far more than the great ones earn back, because a bad-fit client doesn’t just fail to renew, they consume senior time, damage team morale, and sometimes become the case study you have to explain away in the next pitch. Declining early is the cheapest risk management available to a services business, and it costs nothing but the willingness to say no to revenue that was never going to be good revenue.
The Short Intake Form — Four Fields That Actually Route
Most intake forms ask ten questions and route nobody, because most of the questions (“How did you hear about us?”) are for your marketing dashboard, not your qualification decision. Four fields do the actual routing work:
What they want done. Not “tell us about your project” — a specific enough answer that you can tell whether it matches your specialism from Stage 1. A prospect who describes a need that’s nothing like what you do is either confused about what you offer (fixable, with one clarifying email) or hoping you’ll be flexible about your positioning (not fixable, and a preview of every scope conversation to come).
What’s driving the timing. “We need this live before our Series B closes” and “we’ve been meaning to get to this for a while” are different prospects wearing the same intake form. The first has urgency that will make the project move; the second may stall at every decision point because nothing external is forcing a choice. Neither is disqualifying on its own, but the answer changes how you structure the proposal and the follow-up cadence.
Who signs. Ask directly — “who else is involved in this decision?” — because the single most common cause of a stalled deal is discovering three weeks into a “yes” that there’s an unmentioned budget approver who was never on the call. A prospect who answers “just me” and means it moves fast. A prospect who answers “just me” and turns out to mean “me, my co-founder, and our board” cost you nothing to identify on the form and would have cost you weeks to discover later.
What range they have in mind. This is the field agencies are most afraid to ask and the one that saves the most time when they do.
The Honest Budget Question, Asked Without Losing the Lead
The fear behind not asking about budget is that naming a number scares off a prospect who might have paid more, or that a low answer means you should have quoted lower. Both fears are backwards. A prospect who won’t give you even a range is usually a prospect who hasn’t done the internal work of deciding this is a real priority — and a prospect who names a number below your minimum is a gift, because you just found out for free instead of after a proposal.
Ask it as a range, not a demand, and frame it around outcomes rather than your rate card:
“Projects like this typically run $X–$Y depending on scope. Is that in the range you were expecting, or should we talk about what’s realistic for your budget?”
This does three things at once. It anchors them to your actual pricing before either of you has spent time on a proposal. It invites an honest answer instead of a guessing game, because you’ve made it clear you’d rather know now than find out at contract stage. And it gives a prospect who’s genuinely below your range a graceful way to say so — some of them will still be worth a smaller-scoped engagement, and you’d rather know that upfront than build a full proposal for a budget that was never there.
The Scoring Table
Score every qualifying lead against the same four criteria before the call gets booked, using your own thresholds:
| Criterion | Strong signal | Weak signal |
|---|---|---|
| Fit | Matches your named specialism from Stage 1 | Adjacent to it, or a category you’ve never actually served well |
| Budget | Named range meets or exceeds your minimum | Vague, or below your floor with no flexibility stated |
| Authority | Names the actual signer, or is the signer | ”I’d need to check with someone” with no name offered |
| Timing | A real external driver with a date | ”Whenever,” or a date that conflicts with your current capacity |
Three or four strong signals: book the call. Two: book it, but flag it for a shorter, more pointed discovery than your default. Zero or one: this is a decline or a referral, not a discovery call — going to the call anyway rarely improves the odds and always costs the time you could have spent on a stronger lead.
Routing Rule
The rule that keeps this simple: route by the weakest link, not the average. A prospect who’s a strong fit, has real budget, and real timing, but has no identifiable signer, is not a “pretty good” lead — it’s a lead that will stall at the exact moment you need a yes, and no amount of strength on the other three criteria fixes that. Ask the missing question before the call, not during it. A five-minute follow-up email — “before we book time, who else needs to be part of this decision?” — resolves more stalled deals than any amount of discovery-call polish.
The Decline Script That Keeps the Relationship
Declining well is a skill, and it pays back later — a prospect you decline gracefully today is a referral source or a return client in eighteen months, when their situation has changed. The wrong way to decline is silence or a vague “we’re not the right fit” with no specifics. The right way names the actual mismatch and, where honest, offers a next step:
“Based on what you’ve described, I don’t think we’re the right fit — we specialize in [your specialism], and this sounds like it needs [what they actually need]. I’d rather tell you that now than take your budget for work we’re not the best at. If it’s helpful, [a specific referral, or: keep us in mind if the scope shifts toward X].”
That script does the same thing your Stage 1 hook did: it names the mismatch specifically enough to be credible, and it protects the relationship by being honest instead of evasive. A prospect who’s declined this way remembers you as the agency that was straight with them — which is worth more, over a long enough time horizon, than the margin you’d have lost forcing a bad-fit project through.
AI Earns Its Place Here
The intake form’s routing logic and the decline script are both structured enough for a model to draft well — once you’ve decided your own thresholds, which is a judgment call no model should make for you.
Feed it your actual specialism, your real minimum budget, and your capacity constraints, and ask it to draft the four intake questions in your own voice, plus a decline script template you can fill in per prospect. It’s also useful for a first pass at scoring a borderline lead against your table — summarizing the intake answers against the four criteria so you’re deciding from a clear readout rather than skimming a form under time pressure.
Never let it make the actual go/no-go call. The scoring table exists so a human decides against explicit criteria, not so a model decides on your behalf — the cost of a wrong “let’s take this one” is measured in months of margin, and that judgment stays yours.