Lesson 8 · 18 min
The five tools cover your own site. Market and competitors are a different thing: someone collected the data and charges for it. Paying makes sense when a decision depends on data that only exists as a paid product, for a set period and with a spending cap. This lesson gives you the criterion, shows what each service delivers, and closes with your own tool plan.
Start the lesson
Goal: decide if and when to pay for data, with a criterion, a spending cap, and a price checked on the site, and build your tool plan.
Four questions before you subscribe. Click each one.
The free core stays; the paid piece plugs in, off by default.
Prices change fast: always check the vendor's site. Tap each one.
open-seo is an open (MIT) alternative to Semrush and Ahrefs, with an MCP server and skills for agents. But the data comes from DataForSEO, paid per call: the README says you need the key either way, and the hosted service charges 28% extra on every request.
It's an interface over paid data, not a data source of its own.
Scrape, crawl, map, and search: turns pages into Markdown and JSON for content audits and briefs. Pricing page read on 10/19/2026: free plan with 1,000 credits per month, 1 credit per page; paid plans start at US$ 16 per month with annual billing. Check the site.
Its search is not Google's SERP.
A store of bots (actors) that scrape SERPs with related questions, Google Maps, and more. Pricing page read on 10/19/2026: free with US$ 5 in credit; Starter at US$ 19 per month. Each actor has its own price.
Whoever runs an actor that scrapes Google takes on the terms violation; quality and price vary by author.
Volume, difficulty, competitors, backlinks, and AI monitors, with an official MCP. AI visibility monitors run from about US$ 20–100 per month with a few questions to US$ 300–1,500+ with volume, according to third parties (Jul–Oct/2026).
Share of voice is a sample of the tool's question list, not an official score.
Exact search volume doesn't exist for free. A paid AI monitor is a sample; the official, free one is Bing AI Performance. A manual battery of 20–30 questions per month is still worth it.
Fill these in before you subscribe and at the end of the test month.
# Pay for data? — <yoursite.com> — <dd/mm/yyyy> | decision the data changes | does the free tier answer it? (lesson) | missing data | who sells it | price checked on | monthly cap | |---|---|---|---|---|---| | <pick 3 topics for next month> | partly (gsc-historico, lesson 4) | exact volume | <vendor> | <dd/mm/yyyy, link> | <R$ or US$> | | <who the competitors are for term X> | no | competitor SERP | <vendor> | <dd/mm/yyyy, link> | <…> |
# Test: <vendor> — from <dd/mm> to <dd/mm> single question: <…> spending cap: <…> (turn off auto-renewal on day 1) API key: only in the local .env, outside git; never in the repository at the end of the month: - was the decision made with the data? <yes/no> - what would the free tier not have shown? <…> - continue, switch, or cancel: <…>
# Tools — <yoursite.com> free core (always): Search Console, Bing Webmaster, IndexNow, inemaseo-sistemas (5), agent via subscription week: gsc-historico + agent reading + decisoes.md month: auditor, agente-ready, bing-ia (export), botlog (filtered log), manual battery of questions in the AIs paid (optional, with a cap): <none | vendor + question + month>
Fictional conversation for practice. Pick the answer you would give; the comment appears right after.
Pick your answer. The comment shows up right after, and the conversation goes on based on your choice.
The table from lesson 1 is the starting point.
Fill in the decision table with at least 2 rows, starting from the "no" questions in your lesson 1 table.
Done when: each row has the decision, whether the free option answers it (with the lesson), the missing data and, if there is a vendor, the price with date and a link checked on the site.
Fill in "my tools plan" and generate the dated plan right below.
Done when: you have the tools plan filled in and the .md (or .ics) file from the generator saved.
Four steps and the exercises for each. Click an exercise to see its "done when".
auditor and agente-ready (lessons 2 and 3)
2 exercises
Findings in priority order and the readiness score.
View the lessongsc-historico (lesson 4)
2 exercises
Local database with the weeks and the non-branded report.
View the lessonbing-ia and botlog (lessons 5 and 6)
2 exercises
Official citations cross-checked with Google, and bots by purpose.
View the lessonagent and gate (lesson 7)
2 exercises
rotina.sh running, agent reading, and decisoes.md.
View the lessonStart with what a bot sees. Critical first; the map shows the lesson 2 exercises.
Requires the CSVs from gsc_semanal.py. The sooner you start saving, the more history you have.
Requires the Bing export and the server log; filter the log before saving it.
Requires the tools running. After that, decide what to pay for (lesson 8).
Audit, keep the history, measure AI and bots, run the routine. Pay only for the market data that a decision asks for.
The tools measure; the groundwork (course 1) and the content (course 2) are what change the number.
Choose the start date and how much time you have per week.
The explanation shows up after your answer.
Question 1 of 4
What is the first question before you subscribe to an SEO tool?skill
Question 2 of 4
open-seo is MIT. So is it free from end to end?
Question 3 of 4
An Apify actor scrapes Google for you. What's the risk?
Question 4 of 4
How do you run the first paid test?
Pick an option. The explanation appears after you try.