This page shows, for each plan, what a customer pays us vs what it costs us to run them — so you can see the profit. Drag the sliders to play with prices and credit amounts and watch every number update live. Hover any ? for a plain-English explanation. Nothing here is locked in — it's a sandbox for testing pricing ideas.
Pick a plan to model?
Try different prices · drag to test?
Drag each plan's monthly price. The profit you keep, the profit margin, and the big totals above all recalculate live using the same cost-to-serve numbers.
Solo light-usage agent
$49CAD / MO
$0$99$199$299
Price
$49
Cost to Serve
$0
Profit Kept
$0
Profit Margin
0%
Pro full-time agent
$99CAD / MO
$0$99$199$299
Price
$99
Cost to Serve
$0
Profit Kept
$0
Profit Margin
0%
Team brokerage seat
$149CAD / SEAT
$0$99$199$299
Price / Seat
$149
Cost to Serve
$0
Profit Kept
$0
Profit Margin
0%
What your plan gets you?
Every plan includes a monthly bucket of credits. Each tool spends some credits — a listing costs 10, a photo edit 3, a social post 1. Drag a plan's slider to change the credit bucket and watch what it buys (and the profit numbers above) update live. Chat is free and unlimited — it never costs a credit.
Solo part-time / starter agent
75credits / mo
05001,0001,500
What that buys each month — pick any ONE, or mix & match
…all from the same bucket. Mix & match however you like.
How much a typical Solo agent actually uses0%
Pro full-time producing agent
300credits / mo
05001,0001,500
What that buys each month — pick any ONE, or mix & match
…all from the same bucket. Mix & match however you like.
How much a typical Pro agent actually uses0%
Team brokerage seat
900credits / seat / mo
05001,0001,500
What that buys each month — pick any ONE, or mix & match
…all from the same bucket. Mix & match however you like.
How much a typical Team agent actually uses0%
If a customer ever uses up their whole bucket, they don't get cut off — they just buy top-up credits ($0.25 each, sold as $25 = 100). But as the bars show, a typical agent uses only a slice of what they pay for, so most never need to.
If we had this many customers?
1,000CUSTOMERS ON THE PRO PLAN
101001,00010,000
?
Money In · revenue
$99,000 CAD
All customer payments / mo
Everything customers pay us in a month — each plan's monthly price × number of customers, plus any top-up credits the heavy users buy. Tap to collapse. (The finance word for this is "revenue".)
?
Cost to Serve · COGS
$21,767 CAD
$21.76 / customer · AI + tools + card fees
What it costs us to actually run the product for a month. The bills we pay other companies: Anthropic (the Claude AI), Google Gemini, Resend (email), Cloudflare (hosting), plus the fee Stripe takes on every card charge. The per-customer figure is the cost to run one agent at the chosen plan. Does not include your own time, marketing, or support. (Finance word: "COGS" — cost of goods sold.)
?
Profit You Keep · gross profit
$77,233 CAD
Money in − cost to serve
What's left after paying the bills to run the product — money in minus cost to serve. This is the cash the business actually keeps before your own salaries, marketing, or office costs. (Finance word: "gross profit".)
?
Profit Margin · gross margin %
78%
Higher is better · aim 70–80%
Of every $1 a customer pays, how much is profit after the running costs. So 78% means 78¢ of every dollar is kept. Higher is better. AI tools like ours aim for 70–80% (the AI itself costs money every time it's used, so we don't hit the 80–90% of older software). Above 80% is excellent; under 60% means something needs fixing. (Finance word: "gross margin".)
Where the running costs go?
Why we don't do flat unlimited pricing
Why AI-native pricing isn't flat-unlimited
The structural risk we're designing around
Flat Unlimited · Margin Erodes
No usage governor
Heavy user vs light user compute~10× cost
What each paysSame flat fee
Margin under our controlNo
Credit Governance · Margin Defended
Tier allowance + auto overage
~80% of usersNever hit cap
Heavy 20%Pay in proportion to use
Per-tier margin floorKnown & bounded
The three plans · monthly price + a bucket of credits
Anthropic offers 90% off cached input tokens. Not wired anywhere yet. Every repeated system prompt across Listings, Chat, Social, Landing etc. compounds the saving — payoff scales with volume. Caching the Listing prompt alone is the easiest first pass.
Lever · Available
Model Routing
Route cheap deliverables (social captions, subject lines, chat) to Haiku 4.5 ($1/$5 per M); reserve Opus 4.7 ($5/$25 per M) for CMAs and Carousels. Agent never sees the difference — same UX, ~30% cheaper aggregate COGS at typical mix.
Risk · Watch
Vendor Concentration
Anthropic = ~57% of variable COGS (up from ~52% — v2 stripped FAL + web_search, so the spend that remains is more Anthropic-weighted). Any pricing change directly compresses margin. Opus 4.7 just dropped 3× from Opus 4.5, but a future hike has full pass-through risk. Mitigation: prompt caching + model routing absorb a 20-30% Anthropic hike before it touches margin.
Risk · Future
Customer Support at Scale
Today: founders (Braden + Kelley) handle every ticket — free labour, doesn't hit COGS. Once volume requires a dedicated support hire (probably ~500+ paid agents), customer support becomes 7-10% of revenue, dropping Pro margin from ~76% to ~68%. Plan: hire from revenue, not from raise.
BUILT BY BRADEN KOOP & KELLEY SKAR
Building Cheat Sheet
One agreed name for every page in Qlarify — so when you say “the X page,” we both know exactly which one. This is the shared map we build from.
Reference = the file/route under the hood (what Claude searches). Say the bold name.
The shareable listing / CMA link you send a client. (unique link per presentation)
/p/:slug
Published Website
The shareable property or neighbourhood site. (unique link per site)
/w/:slug
Unsubscribe Page
Where an email recipient opts out (CASL-compliant). (unique link per recipient)
/u/:token
Is my information safe? (the plain-English version)
Short answer: yes — here’s why, without the tech-speak.
Think of Qlarify like a building where every agent gets their own locked room. When you sign in, the system checks your key (email + password). Your password is never kept as-is — it’s run through a secret scrambler so even we can’t read it; we only ever check that your key fits the lock.
Once you’re inside, you only ever see your own room. Every time the app pulls up your contacts, presentations, or branding, it’s hard-wired to ask only for the things tagged to your account — there’s no hallway from your room into anyone else’s. We even run an automatic guard that blocks any new code from breaking that rule before it can go live.
Everything travels through a locked tunnel (HTTPS). Your data is scrambled on the way to and from the app, so no one can read it in transit.
Outside keys are double-locked. If you connect another tool (like your CRM), its password is encrypted before it’s stored, and it’s never shown back on screen or written into any log.
Your clients’ privacy is built in. Contact lists follow Canada’s anti-spam + privacy laws — every contact has a consent record, every marketing email has a real unsubscribe link, and if someone asks to be removed we can erase them (while keeping the legally-required proof they once opted in).
The links you share are clean. When you publish a presentation or website for a client, the system strips out anything that could run sketchy code, so the link is safe to open.
It’s been professionally audited. The whole system went through a security review, and an automatic checker re-tests the riskiest rules on every single update.
Bottom line: your data is yours, it’s walled off to your account, it travels encrypted, and the parts that face the outside world are sandboxed and double-checked. Put your real client info in with confidence.