Questions, answered.
Fitting, for a product built on exactly that.

Conversational marketing analytics. ratio connects to your marketing data, and you ask questions in plain language, like "What was our CPL on Meta vs. Google last week, by campaign?", and get the answer back. Verified, traceable, in your team’s own metric definitions. No SQL, no dashboard-clicking, no slide 54 of 67.

A chatbot works with whatever you paste and produces something plausible. ratio is connected to the live source, calculates against your definitions, and shows where every number comes from. The difference is the difference between an answer and a guess that sounds like one.

No. If you can write a Slack message, you can use ratio. It was built for the people who currently have to ask someone else: marketers, founders, strategists. Not for the people who build dashboards.

It says so. When a question is ambiguous, say two metrics could match or a date range is unclear, ratio asks back instead of inventing a number. An answer you cannot trace is not an answer. That rule is not negotiable, even for ratio itself.

Keep it. Save any answer as a view your team can return to. Same question, fresh numbers. Ask once, reuse every week.

Meta, Google Ads, Google Analytics, and Google Sheets today. Your accounts, campaigns, and conversion events come with. LinkedIn and TikTok are in the rollout pipeline. Connecting takes one to two minutes; the whole account setup, about five. Your first question takes about as long as typing it.

No. Access is OAuth and strictly read-only. ratio can see performance; it cannot touch campaigns, budgets, bids, or settings. It also never sees your raw account credentials. That is the point of OAuth.

As fresh as the platforms themselves. Connectors pull live: the moment you ask, ratio queries the source directly. No cached copy, no overnight sync. The number you get is the number as of now.

Good, most teams do. That is what custom metrics are for: as a customer, you define once what counts as a lead, how CPL is calculated, which campaigns belong to which brand. Every answer is computed against your definitions, for everyone who asks.

ratio starts with sensible defaults, built from experience. They yield to your definitions the moment you set them.

Wherever you decide: if you need the guarantee, toggle EU-resident processing for in-region data handling end to end. GDPR compliance applies by default either way, and data is encrypted in transit and at rest.

No. Your numbers, your definitions, and your questions stay out of every training set, ours and our providers’. Your CPL is your business, not training material.

Then isolation matters, and you have it: every workspace runs in its own data context. Client A never meets client B’s numbers. Not in storage, not in an answer, not by accident.

They shall have it: sub-processor list, DPA, and records of processing, available on request in plain English and German. Audit-ready is a default here, not a premium tier.

You leave clean. Revoke access in one click; your data is deleted on request. No tickets, no waiting, no retention games.

Start free for 30 days. Full product, all connectors, no feature gates. And no credit card to start. Connect your data and ask your first question in under five minutes.

One prompt = one request. A question, a follow-up, a report command: each counts once, whatever the size of the answer. Plans are sized against how a team actually asks; the details live on the pricing page.

You decide. There is no card on file, so nothing gets charged by surprise. If ratio earned a place in your week, pick a plan. If not, revoke access and leave clean.

Then it deserves a better answer than an FAQ can give. Start the trial and ask it where asking is the whole interface, or reach the team through the product. Answering questions is, after all, the business model.