Questions about the Decision Map.

One buying decision, followed all the way. The Decision Map shows what AI tells your buyers, the Work Plan ranks the work those answers point to, the Action Pack gives you the material for each item, and Measure Again, on the same decision, shows what changed in a Before/After Snapshot. One price covers the whole loop.

A complete study of how AI answers present one offering, in one buying category, to buyers in one country. Five buying questions you approve go to ChatGPT, Claude, Gemini and Perplexity, three times each. The map shows the options put forward, your place in each answer, the words used about you and the pages cited or found by search, with the answer behind every finding.

You’d get one answer, shaped by your own account, history, settings and location. Ask again and it may change, and it tells you nothing about the other three engines. A Decision Map asks five approved questions on four engines, three times each, in your buyers’ market, and keeps every answer for you to read.

That’s why every question is asked three times on every engine. You see where an answer holds and where it moves. Every repeat is kept, nothing is averaged away, and each finding shows how many answers stand behind it.

Yes. Every finding leads back to the answers behind it, and each answer can be read in full with its engine, model and date. The Evidence & Trust page in the example map shows how it works.

Your website, your buyers’ country, and the offering and buying category you want studied. CitePack suggests those last two from your site. Then you say in your own words what the offering is best at and who it’s for, and plan your five questions: a buying situation for each, with up to two buyer specifics. If you like, add who the offering isn’t for and facts you stand behind, such as a price, a limit or a certification.

Any country, one per map: the one your buyers are in. The field starts at the United States and you can change it. A company that sells in two countries runs two maps.

Five points in a buying decision. In category consideration, the buyer is just starting to look. In job suitability, they describe the work they need done. In mandatory eligibility, they have a condition they won’t bend on. In balancing priorities, one thing matters more to them than another. In change of solution, they’re moving away from what they use now. Each question covers one situation, and the five together cover the whole decision.

Details that make a question sound like your buyer: the work to be done, a must-have, a preference or what they use today. A question carries up to two, or none when it should stay broad. CitePack suggests them from your website, and you can write your own.

CitePack drafts them from your study. You have the final word: change any wording you like, then approve all five together. Once approved they’re sealed. Every engine gets them exactly as written, and they don’t change for that study.

CitePack drafts one question for each of the five buying situations, covering the whole decision. Each is framed for your offering, its buying category and your buyers’ market. It carries the specifics that matter, like the work to be done or a must-have. It is never a fixed prompt from a library. You read every question, change any wording and give the final go. Each goes to ChatGPT, Claude, Gemini and Perplexity, three times on each, every repeat clean. That makes 60 observations of one decision, every answer kept word for word.

ChatGPT, Claude, Gemini and Perplexity. Each one is asked through its official API, in your buyers’ market, on a fixed model with its web search on.

By default, two ways on every repeat: each question names your buyers’ country, and each engine is set to it. ChatGPT, Claude and Perplexity take it through their own location setting. Gemini gets a location point in your country. Your approved question is never changed, and your map records how the market reached each engine.

Yes, by default, and we recommend it: that way all four engines read your market in the same words, inside the question itself. It’s your call. If you turn it off before you approve, the questions don’t name your market and CitePack sends no location to any engine, so each one answers from its own default. Your map shows which you chose.

It’s the pattern behind what they see. Each engine is asked through its API, on a model chosen to match its free app, with web search on, in your buyers’ country. In the app, each person’s history and settings shape their own screen, so no two match exactly. The map shows what holds across 60 controlled observations. It doesn’t replay any one person’s session.

A fixed model per engine, chosen to match the one its free app gives buyers. Every answer records the model that produced it, and answers from different models are never compared as if they were the same.

Yes. Web search is on for all four engines, so answers can draw on pages from the web. The map keeps the pages an answer cited apart from the pages its search found, and shows you both.

An Overview of how often you were put forward and named first across all 60 observations. A page for each question, with every answer. AI engines, on how each engine built its answers and which pages it used. Who gets picked, on the other names and who comes first. What AI says, which sets the answers’ words about you beside the headline phrases on your pages, and checks the prices, free plans and amounts they state against your website. And Evidence & Trust, with every answer exactly as it came back. Beside the map, at its own address, you get the Work Plan.

A ranked list of the work your answers point to, beside your map. Each item names the answers and pages behind it, who does it and what to check next time you measure. Where another company comes first and no page of yours was used, you get the outline of the page to build.

Yes, when the answer gives one. If an answer lists options in a clear order, the map shows your place: first, second, third and so on, with how many answers held each place. If the order isn’t clear, the map says so instead of guessing.

Yes. Who gets picked shows every other name in the answers, who comes before or after you and who is listed first, each with the answers behind it. It’s a record of what the answers said, not a ranking of companies.

You see the disagreement, not an average. You can read every answer, compare the repeats and see how many answers sit behind each finding.

No. An answer can mention your company without presenting your offering as an option. A citation is different again: it is a reference to a page. The map records each one separately, so a mention is never reported as an offer and a cited page is never reported as a recommendation.

The material for each Work Plan item, ready to use: a pitch email, a page draft, a rewrite, a correction note or a technical fix, whichever the item calls for. Each piece says why it exists. Copy it or download it as Markdown, then mark the item done. It comes first to founding companies.

Yes. The loop includes one Measure Again: the same five questions, the same four engines and the same market, on Day 28 or X, the day you choose. We recommend Day 28, 28 days after your first map. The Before/After Snapshot puts the two maps side by side, with each item’s check beside what you did. It reports movement or none and never promises a change. It is not a daily tracker or a score.

No one can promise that, and we don’t. You get the work your answers point to: the Work Plan ranks each job and ties it to the answers and pages behind it. It doesn’t guess at an engine’s reasons, and it never promises a recommendation, a citation, or that a change will move the answers.

Neither. One Decision Loop is a complete study of one offering in one buying category, for one buyer market: the Decision Map, its Work Plan and Action Pack, and one Measure Again when you choose. No single score, no league table of companies, no continuous monitoring.

Not as results. Crosstrow is an illustrative example. Every example here, the full example map included, shows how to read a Decision Map. None of it is a study result, a customer study or evidence of a commercial outcome.

The founding cohort is open: ten seats, first come, first served. Join the waitlist, and if a seat is yours, we’ll send you your own sign-in.

The first ten companies to run a Decision Map. Each gets its own sign-in, sets up its own study and pays the founding price for one Decision Loop: $99, or ₹4,999 if you pay from India. After the founding ten, one loop costs $149 or ₹7,499. Every loop includes the Work Plan, the Action Pack and one Measure Again.

Join the waitlist. The ten founding seats are first come, first served, and each one comes with its own sign-in. You sign in, set up your study and pay at checkout. Then you read the five questions CitePack drafted, change any wording and approve them. Your map is in your account 30 minutes later on average, and we email you when it’s ready, with a link to your Work Plan. If it isn’t delivered within 24 hours, you get a full refund.

For each of the ten founding companies, one Decision Loop costs $99, or ₹4,999 if you pay from India. After those ten, one loop costs $149 or ₹7,499. Already a customer? Each further loop is $99, or ₹4,999 if you pay from India. A loop covers one offering in one buying category and one buyer market. A different offering, category or market is a separate loop. If your map isn’t delivered within 24 hours, you get a full refund.

30 minutes on average after you approve your questions. We email you when it’s ready. If it isn’t delivered within 24 hours, you get a full refund.

You get a full refund. If your map isn’t delivered within 24 hours of approving your questions, we refund what you paid in full.

In your CitePack account. Only you can open it.