Assumptions and dashboard

A model in Vector is one project’s forecast: a set of assumptions you control, and a 36-month trajectory calculated from them. You change the assumptions; the dashboard shows where they lead.

​

Change an assumption

The assumptions sit on the project’s Dashboard, grouped by what they describe:

• Acquisition funnel — marketing budget, conversion rates and word of mouth, which together decide how many customers you bring in each month.
• Pricing — price per seat, seats per workspace, and the share of seats that pay. Add more packages for tiers, or turn on a credit tier.
• Cost of goods (COGS) — what it costs to serve customers, such as AI inference per meeting and infrastructure.
• Team — the roles on your team, their salaries, and how each one grows.

Edit any field directly. Changes save on their own a moment after you stop typing — a dot next to the project name means some haven’t saved yet. Click Save now to save at once.

​

Let AI suggest values

Several sections can ask AI for values grounded in real data instead of your guesses:

• ✦ Suggest channels & CAC in Acquisition funnel — cost per visit, conversion rates and budget, from benchmarks for your market.
• ✦ Suggest TAM in Acquisition funnel — the size of your market.
• ✦ Suggest from competitors in Pricing — prices, from what your competitors charge.
• ✦ Suggest benchmarks in Base & MRR breakdown — churn and seats, from market benchmarks.
• ✦ Suggest COGS in Cost of goods (COGS) — inference and infrastructure costs, from current AI model and infrastructure prices.

Each proposal shows its reasoning, a source link and how confident the AI is. apply the ones you agree with — or Apply all when there are several. Nothing changes in the model until you apply it. Each run uses your AI key. With pricing packages turned on, the pricing and benchmark suggestions aren’t offered.

​

See where the plan leads

Summary — outcomes at month 36, at the top of the project, shows the headline results at the end of the horizon: workspaces, paid seats, annual recurring revenue, gross and operating margin, breakeven, blended customer acquisition cost, and unit economics such as LTV to CAC and CAC payback. The charts below show how you get there, month by month. To see how much money the plan needs, look at All costs, payroll & P&L: the lowest point of the cumulative line is the capital you’d have to raise. The model doesn’t soften a declining trajectory, so you see the real shape of the plan you’ve built.

​

Test a what-if

Change one assumption and the KPIs and charts update at once — no reload, no waiting. Try a different conversion rate or salary and watch the 36-month trajectory reshape. To see which assumptions matter most, look at Sensitivity: it shifts each input a little and shows how much your annual recurring revenue at month 36 moves. The inputs at the top are the ones worth getting right; the ones at the bottom you can estimate roughly. Unusual or extreme inputs never break the display: they produce flat or zero trajectories, not errors or blank charts.

​

Check that the numbers are realistic

Plausibility check compares what your model implies with real market norms. Green means plausible, amber borderline, and red means the curve demands something rarely seen in practice — a signal to revisit the assumption behind it.

​

Work back from a goal

Instead of tuning assumptions until the numbers look right, set the destination and let AI find ways to get there. Open the 🎯 Planner tab and pick a goal:

• Hit a target — reach a value, such as paid users or annual recurring revenue, by a given month.
• Reach breakeven — become operating-profitable by a given month.
• Maximize — push a metric as high as possible.

Optionally set a Cash floor ($) — how deep in the red a strategy may go — and click ✦ Find strategies. AI proposes several distinct strategies, and each is run through your model, so the results are calculated, not guessed. Apply to model → loads the one you like into your assumptions.

​

Trace how a number is calculated

When a number looks wrong, open the Data model tab. It shows the whole calculation as a map: inputs on the left, formulas in the middle, results on the right. Click any value to see its formula, its current number, what it depends on and what it feeds into, traced all the way back to your inputs.

​

Where to go next