Pricing
One model, three components, one place to compute it. This page is the single source of truth for anything commercial.
Price the monthly scraping cost from the brand’s real data volume, social, reviews, news and competitors, using per-platform unit costs. Outputs the monthly price and the one-time initial build. USD / INR.
This is the single source of truth for anything commercial. If another page in this wiki implies a different model or quotes a per-generation credit cost, this page wins and that page is stale. Flag it.
The model, in one sentence
Zocket prices in three components: a fixed subscription, plus data scraping, plus usage credits.
Platform access and seats. A business decision, set with the sales lead.
The public data feeding Brand Brain and Consumer Insights, priced from the volume of mentions, reviews and filings the brand actually generates. Calculated from real per-platform unit costs.
Metered AI usage: generation, analysis and research. Estimated per account from their volumes.
Every customer conversation should use these three words. Not "a licence," not "per creative," not "per seat and some credits." Subscription, scraping, credits.
The scraping calculator
Scraping is the component people get wrong most often, because it scales with the brand's data footprint rather than with headcount. There is a live calculator for it:
→ Brand Brain, Scraping Pricing Calculator
Price the monthly scraping cost from the volume of data a brand generates. Toggle USD/INR.
How it works. One data point = one mention: a post, a review, or a filing. You enter the brand's monthly volume per platform and a scrape cadence, and it computes the monthly cost from real per-platform unit costs ($/1,000 data points, derived from actual spend, calibrated across ten finance brands).
What it covers:
- Social mentions: Instagram, X, Facebook, YouTube, LinkedIn, Reddit, Quora
- Reviews & ratings: Google Reviews, Play Store, App Store, Trustpilot, MouthShut, Amazon, Myntra, Flipkart
- Other public sources: Google News, websites, Moneycontrol
- Competitors: tracked on the same sources, priced per competitor per cadence
- Subscription & credits: your two levers, entered directly
Cadence multipliers reflect repeated scraping runs, re-scan, proxy and compute overhead: Daily 3× · Weekly 1.5× · Fortnightly 1.25× · Monthly 1×.
It also outputs a one-time initial build figure for standing the brand's graph up in the first place. Don't forget it when you quote.
When to use it
- Any ORM or Consumer Insights deal, scraping is the dominant variable cost, and volume differs enormously between brands.
- Any time a prospect names the platforms they want monitored. Put them in the calculator on the spot.
- Before quoting a competitor set. Each tracked competitor adds real cost.
How to talk about it on a call
Don't read numbers off the calculator live. Use it to build the quote afterwards, and on the call say:
"Pricing has three parts, a platform subscription, the data collection, which depends on how much conversation your brand actually generates, and metered usage credits. Give me the platforms you need monitored and roughly your volumes, and I'll come back with a firm number."
Pricing posture on first calls
Quote a range, name it as a range, and firm it up after discovery. Never improvise a number.
Acceptable to say on a call:
"Pricing is a platform subscription plus data and usage. It scales with how many modules you take and how much data we're collecting. We'll send a firm proposal once we know the volumes."
Every inconsistent number we have ever given a prospect came from someone being helpful in the moment. Quote a range, name it as a range, and send the firm number in writing.
Do not commit to per-adaptation pricing, per-generation credit costs, failed-generation billing, or exact token economics. If pushed:
"Let me confirm the exact commercial structure with our sales lead and come back to you in writing."
Trials and pilots
- Trial workspaces come with initial credits so the customer can test without a commercial conversation first.
- Scope creative pilots in masters + adaptations, not credits. That's the unit customers think in. The live Lenovo pilot is 25 masters + 100 adaptations per month.
- Feedback and iterations don't count against a creative cap: only final delivered outputs do. Say this early; it removes the biggest unspoken fear.
- Video is scoped and priced separately from images. Ask for the rate card rather than estimating.
- A paid POC with success KPIs written down beats a free trial every time. See the Onboarding SOPs.
Undefined, route these, don't answer them
These also appear on Never Commit, alongside the non-commercial things we must not promise.
These have no agreed number. Take the requirement and route it to sales leadership:
- White-label / OEM pricing (the Paytm shape): entirely undefined.
- Partner economics: Tensai has sourced a large share of the pipeline with no documented commercial structure.
- Customer-hosted, VPC / on-premise, or bring-your-own-LLM-key deployments: requested by 5paisa, Dabur, Paytm and EloElo. Commercial and solutions decision, not a call answer.
- Per-video rate cards beyond a pilot's agreed scope.
Why this page exists
Historically pricing was quoted inconsistently across calls, different monthly figures for the same module, per-adaptation numbers that didn't reconcile with token economics, and credit costs quoted from memory. That cost credibility more than it ever won deals.
The fix is this page plus the calculator. One model, three components, one place to compute it. If you find a page in this wiki that contradicts it. It's stale, say so and it gets fixed.