We measure. We decide. We execute.
Periscopy shows where AI decides against you. The team does the work that changes the decision: technical, editorial and authority. Every action gets a date and a reading before and after.
Conversation first. Fixed number after.
We do not start with recommendations. We start by measuring what eleven AI engines and surfaces answer today when someone asks for your category, with our own software, and we keep measuring so the effect of every action is visible.
Audit
Hundreds of real buyer questions across eleven engines. Who shows up in your place, which sources decide, where to start.
Request an audit →Diagnostic
A full map of your presence, the gap against your competitors and a plan of actions across the 8 SINAL dimensions, over 4 horizons.
Talk to us →Operation
Periscopy included, execution by the team, a monthly report and proof action by action. No lock-in: the work stays with you.
Talk to us →No published prices: every project depends on questions, markets and business lines. We give you the number on a 30-minute call.
What goes into each format.
For those who want to know where they stand before investing.
- Presence audit across 11 AI engines and surfaces (150 prompts × 5 runs)
- Knowledge vs augmented: training memory and live web search, engine by engine
- Real buyer questions, by persona and funnel stage
- Technical audit for AI: schema, llms.txt, robots.txt, entity, performance
- Competitive analysis: share of voice, positioning and cited sources
- Benchmark against 3 competitors, engine by engine
- Source map of the category: the domains AI cites and where you are missing
- Sentiment and excerpts of how AI describes your brand
- Local signals where relevant: Google AI Mode and Google Business Profile
- Action plan prioritised across the 8 SINAL dimensions
- 60-minute executive presentation, in Portuguese or English
For brands building presence in AI answers. Includes everything in the Diagnostic, kept up to date.
- Periscopy: weekly audit and a live dashboard across the 11 engines and surfaces
- Technical implementation: expanded schema, llms.txt, robots.txt for AI crawlers
- Optimisation of key pages so models read and cite them
- Content optimised for citation, with briefs for each unclaimed question
- FAQ engineering in the format models prefer to cite
- Free territory: questions and sources nobody owns yet, with alerts when someone moves in
- Competitor and reputation monitoring, week by week
- Before-and-after proof: every action tied to the change in citation
- Local brands: Google Business Profile, reviews, NAP and local directories
- Monthly PDF report, weekly digest and direct access to the team
The Operation has a defined scope and cadence, and can be cancelled with prior notice. No guaranteed positions: the commitment is transparent measurement and work tied to each gap. We talk first, and the plan adapts to where you start.
We measure the AI layer. We build the foundation that holds it up.
Measurement comes first: our own technology watching the engines every week, from the Portuguese market. But we do not sell it on its own. We deliver the whole package of being found: technical SEO, local SEO, reputation, and the AI layer on top. Agencies tell you what they think; we show you what the engines answer, and deliver the plan the numbers dictate, ready to execute and with proof action by action.
Showing up when people ask AI about your category.
Measuring and building your presence inside ChatGPT, Claude, Gemini, Perplexity and Google AI Mode. Citation rate, share of voice, entity authority. It is the work most agencies still do not know how to do, and it is what put us on the map.
The foundation AI reads.
Indexing, schema, performance, extractable HTML, architecture. Without this, the best AI strategy has nowhere to be read from. Google confirmed in 2026 that its AI features rest on its core ranking systems.
For brands with an address.
Google Business Profile, reviews, consistent NAP, maps, LocalBusiness schema, local directories. What AI and Google use to answer questions with local intent.
What people say about you.
Reviews, mentions and sentiment in forums and social networks, distilled by theme. Google Reviews for brands with a physical space, Reddit and public reviews for digital ones. Models read reputation before they recommend.
For those who live online.
SaaS, software, platforms, websites, e-commerce. Citable content, topical authority, the full technical infrastructure to be found and recommended.
The same team that builds Periscopy builds your presence: measurement engineering on top, a technical and local SEO foundation underneath.
Four phases. No black box. Everything documented.
We do not sell magic. We sell a plan that is readable, executable and comparable against results. International platforms measure visibility; what is missing is rarely data: it is someone who knows what to do with it in European Portuguese, in your sector, in your sales cycle. We work in that layer, accountable for the final number.
Diagnostic
We map your presence across 11 engines and surfaces with 150 prompts from your sector. We identify where you are, where your competitors are, and the exact gap.
Build
Schema, llms.txt, citable content, optimisation of key pages. We implement directly or guide your team.
Authority
Placement in the sources models read: technical directories, comparisons, sector publications. Active building of external citations.
Operation
Weekly monitoring, monthly report, continuous adjustment. Direct access through a shared Slack or Teams channel.
Ten fronts of work. All measured.
Appearing in AI answers is not one job. It is technical work on the site, authority built outside it, and measurement that tells the two apart. A partner who only does the first will show you a clean audit and a flat curve.
A website that keeps up with AI.
We build AI Native websites: fully functional, readable by the engines and ready to change in hours.
See AI Native websites →The measurement is ours. The tool in front of it is yours.
An agency account with as many clients as you like. Periscopy with your logo, your colours and your domain. Whoever logs in to read the numbers sees the agency, not us.
Talk about a partnership →The same product brands use. White-labelled, with the agency's logo.
AI already shows products, prices and ads.
Answers stopped being only text: Google AI Mode shows product panels with prices and buttons (live in Portugal), ChatGPT shows shopping carousels, and in the United States the purchase already completes inside the conversation. Next to the organic answer sits paid space. We cover both, and we say plainly what has reached Europe and what is still on its way.
Agentic commerce readiness
A shopping agent reads price, stock and delivery time from your data, not from your well-designed site. We audit and prepare the layer engines actually read: a complete and fresh Google Merchant Center feed, Product and Offer schema with shipping and returns, Merchant Center's new conversational attributes, and registration in assistant feeds as they open to Europe. The feed is the richest and freshest structured source a shop publishes, which is why optimising it moves more than one surface. Third-party measurements point to a high overlap between ChatGPT carousels and Google Shopping organic results (Search Engine Land, March 2026), which we treat as a pointer and not as a premise: what holds the work up is the mechanism, not someone else's percentage.
Paid visibility inside AI answers
On Google, ads enter AI Mode and AI Overviews from the campaigns you already run; on ChatGPT, they reach Europe from August 2026. We manage paid presence inside the answers and measure both columns in one table, paid and organic kept apart, so you can see what the ad adds to the presence you already earned. In half the engines we measure (Claude, Perplexity, DeepSeek, Mistral) there are no ads: there, organic is the only route, and that is why no media buy replaces GEO.
The machine behind our studies will run the question you bring.
Eleven AI engines and surfaces, personas, and every answer read one by one. The same collection behind the five studies we published, pointed at the question in your market.
11 engines and surfaces. Two routes each.
ChatGPT, Claude, Gemini, Grok, DeepSeek, Mistral and Perplexity. Each one that supports it runs in two modes: what the model answers from training memory, and what it answers with live web search. The two are won differently, and reporting them together hides which lever is working.
Google AI Overviews, Google AI Mode and Microsoft Copilot. Queried from your country and in your language, because they return different answers per market. Measuring a British brand from a Portuguese location would produce numbers that look right and are not.
GEO takes months, not weeks.
The training-memory route moves in months, because it depends on sources that third parties publish and models absorb. The live-search route moves in weeks. Anyone promising ChatGPT presence in thirty days is improvising.
We measure what we can measure.
Citation rate, share of voice against direct competitors, average position, sentiment, and the sources each engine cites. We do not report estimated impressions or potential reach, because those numbers are not measured, they are imagined.
Our depth is Portuguese.
The published research, the case studies and the Tier-1 press relationships are Portuguese. The method transfers; the local source landscape has to be rebuilt for your market, and we would rather say that now than discover it together later.
Before you sign.
How long until there are results?
Where there is live search, weeks. In the model's memory, months. We do not promise a deadline: we measure every week and show the before and after.
Does this replace our SEO?
No. AI engines read what is indexed. We do both layers, and the technical foundation comes first.
Who does the work?
Technical delivery is in house. For one-off productions we work with trusted freelancers, always disclosed.
And if we leave?
You keep the implemented work, the full documentation and the measurement history. No hostages.