Depra AI vs Rankscale at the bottom of each ladder, where one entry plan includes sentiment analysis and the other marks it not included

Both of these vendors sell a genuinely cheap first plan, and the two cheap plans are not the same shape. Read on 31 October 2026, rankscale.ai/pricing publishes Essentials at $20 a month, marked yearly only at $17, with 120 credits, 2 brand dashboard slots, up to 480 answers from AI engines and 10 page audits a month. On the same day depra.ai/pricing publishes Starter at Rs 1,999 a month plus 18 per cent GST, with 16 prompt slots, 2,000 successful checks a month, 1 project and 3 competitors.

The headline numbers favour Rankscale on answers and Depra on nothing obvious. Then you read what each entry plan is allowed to do with the readings, and the picture turns over.

What the cheapest plan at each vendor is allowed to do

Capability	Rankscale Essentials	Depra AI Starter
Sentiment analysis	Not included, per its own matrix	Sentiment and average position, published on the plan
Custom dashboards	Not included	Not published as a feature either way
Data exports, CSV or Sheets	Not included	CSV download, and the websites AI cites
Competitor benchmarking	Included	Included, 3 competitors
Citations analysis	Included	Included, the websites AI cites
Local results	Not included	Any of 40 markets
MCP	Included	MCP server, named for Claude Code, Cursor and VS Code
Team workspaces with unlimited seats	Not included	Not published as a seat limit
Ranked recommendations	Rankscale Scout recommendations, included	Your top fixes, ranked
Billing term	Yearly only at this tier	Monthly, 7 days free with no card

Two lines in that table decide most entry level purchases. A monitoring plan whose readings cannot be exported is a plan whose history lives in somebody else's dashboard, and Rankscale marks data exports as not included at Essentials and included from the tier above. And a plan sold yearly only is a twelve month commitment at the moment you are least sure, which is a real condition rather than a complaint, since the page states it plainly next to the price.

The step up, and where each ladder puts the useful things

Rankscale's second tier is Pro at $99 a month, or $84 yearly, with 1,200 credits, 4,800 answers, 10 brand dashboard slots and 50 page audits, and it is the tier where sentiment analysis, custom dashboards, exports, local results and unlimited seat team workspaces all arrive. So the honest way to read Rankscale's ladder is that Essentials is a monitor and Pro is the product.

Depra's second tier is Growth at Rs 7,499 with 70 prompt slots and 8,700 successful checks, and what arrives there is different in kind: sites that mention competitors but not you, a shopping report naming the products ChatGPT shows by brand and store, and English against Hinglish results side by side. Nothing that was present on Starter is upgraded; new reports are added. A buyer who only needs the Starter reports can stay on Starter indefinitely and still export.

The agency question, which both of them price

This pair is unusual in publishing an agency position with a number on it, so the comparison is direct.

• Rankscale Growth, $385 a month or $327 yearly. 5,500 credits, 22,000 answers, 50 brand dashboard slots with extra slots addable in the app, 200 page audits, white labelled dashboard links with API, and an agency programme published as lifetime affiliate commissions and client discounts, a configurable entry in a partner agency directory, accelerator webinars and a private channel.
• Depra Scale, Rs 12,499 a month. 120 prompt slots pooled across 5 client projects, 15,000 successful checks, prompts movable between clients freely, your logo, colours and footer on client reports, per client reports, CSV exports and action lists, 8 competitors per client, and any market per client.

These two count different things, so pick by the shape of your client list. Many small clients with a handful of questions each suits Rankscale's 50 containers. Fewer clients with deeper question sets, and the freedom to move capacity between them mid month, suits Depra's pooled slots. The phrase to notice on Depra's page is move prompts between clients freely, because it is the only published statement in this pair about reallocating capacity you have already paid for.

Credits against checks, in one paragraph

Rankscale's unit is a credit and it publishes the weight of each engine, Google AI Mode, ChatGPT, Perplexity and Gemini at 0.25, DeepSeek at 1 and Anthropic Claude at 2, with an FAQ line stating that a query typically costs 0.25 credits per engine per prompt. So a credit is worth four cheap checks or half a Claude check. Depra's unit is a successful check, defined on its page as one question asked to one AI engine, with a prompt slot defined as one buyer question in one language and one market. Depra's four engines are flat priced inside the plan and anything beyond them is an Enterprise add on. Rankscale's 17 or more engines are all spendable from the same pool at different weights. Both publish enough for you to model your own set, which is more than most of this category does.

Where we are, and where we are not

We sell in this category and we should be clear that we do not compete at the bottom of these two ladders. Our pricing page, read 31 October 2026, publishes a free scan on your own domain and then a paid ladder whose cheapest plan that watches AI apps at all is GEO Growth at $299 a month, or Rs 11,799, for 100 buyer questions, 1,200 scheduled answers a month and all five engines with Claude included. A buyer with $20 or Rs 1,999 to spend should buy from one of the two vendors above, and this page is not going to pretend otherwise.

Two further published lines go against us. Rankscale puts team workspaces with unlimited seats on its paid tiers and we publish Logins included, 1, full access. Rankscale also publishes a changelog, and we do not publish one. What our page does publish that neither of these two does is the writing side, 20 pages a month on Growth and 40 on Scale, written and published to your own site, plus the per surface credit weight, one credit for a standard check and three for Claude, which is a three times multiple where Rankscale's published weights make Claude an eight times one.

On the published record behind the Rankscale half of this pair, including what it prints instead of a rating of its own, see Rankscale reviews compared.

Questions about the entry tiers at Depra AI and Rankscale

Is Rankscale Essentials really 17 dollars a month

At $17 it is the yearly rate, and the comparison table marks the tier Yearly only, with $20 shown as the monthly figure. Every tier above it publishes both a monthly and a yearly price.

Can I export my readings from the cheapest plan

From Depra Starter, yes, CSV download is published on the plan. From Rankscale Essentials, no, its matrix marks data exports in CSV and Sheets as not included at that tier and included from Pro.

Which cheapest plan collects more answers

Depra Starter, on published figures, 2,000 successful checks a month against up to 480 answers on Rankscale Essentials. Rankscale's 480 can be spent across 17 or more engines; Depra's 2,000 are spent across its four.

How many client brands can each entry plan hold

Rankscale Essentials publishes 2 brand dashboard slots with extra slots addable in the app. Depra Starter publishes 1 project and 3 competitors. Neither is an agency plan and both vendors publish a proper one higher up.

Does either entry plan include Claude

Rankscale does, at a published weight of 2 credits, which means 120 credits on Essentials buys 60 Claude checks and nothing else if you spend it all there. Depra holds Claude behind an Enterprise add on, so its Starter plan covers ChatGPT, Gemini, Perplexity and Google AI Overviews only.

What is the cheapest honest way to find out whether this is worth buying at all

Use the free routes first. Depra publishes 7 days free with no card on Starter, Growth and Scale. Rankscale publishes a free trial of Pro, which is the tier where exports and sentiment live, so the trial shows you more than the Essentials plan you might then buy. Run both, write down what each produced, and then pick a plan against that rather than against a feature list.

How we know this

We read the assistants rather than guessing what they do. In our own audit of 1,260 saved answers from ChatGPT and Gemini across the United States, the United Kingdom and India, 82.0% of every citation landed on a site the business itself owned, and review directories took 0.7%. Counted as one domain per reading, 16,516 of 20,143.

That is the whole asymmetry: the pages you own are the pages an assistant quotes back to a buyer. See the dated readings and the method on our proof page.

Find out whether assistants name your business, on your own domain and your own buying questions.

By Aditya Goel, Founder, AI Knows Us. Reviewed 1 November 2026.