# SEO Map settings {% embed url="" %} ### Map Name This is just for your reference. Name it whatever you like! ### Niche The broad "parent topic" of your website, business, brand, etc. If you're a beginner blogger or not sure where to start, [see this guide on choosing a niche.](https://docs.topicalmap.ai/tutorial/what-makes-a-good-niche) Example niches: * digital marketing * table tennis * dog food * pet snakes * computer game shortcuts * VPN tools **TIP:** We suggest keeping the niche REALLY broad. It's better to let the AI come up with TOO MANY ideas, rather than too few. ### Language The language that the AI *should* return. Under the hood, this basically functions as a *translation* and does **NOT** affect the actual topics generated. ### Country The topics, subtopics, etc, will be *slightly* geared towards this country. **Note 1:** This SHOULDN'T affect the output language, but sometimes the AI gets confused here. **Note 2:** This SHOULD be a "slight" targeting, but sometimes the AI goes overboard and tries to make ALL topics related to the country. **Note 3 for 'Cluster Views': Choosing a target country does NOT affect the search volume metric shown. TMAI shows the global monthly search volume.** (This is as of August 2024, and will hopefully change as we update this!) ### Amount of Topics The general size of the map. Choosing fewer topics yields fewer subtopics & fewer microtopics, but the results will likely be more relevant. Choosing *more* topics yields more subtopics & microtopics, but the results might be more 'creative.' 🧧 **CREDITS:** The amount of topics you choose will cost less/more credits. ### Model The AI model that will be used under the hood. Generally speaking, the more expensive & larger models (Claude 3 Sonnet & GPT-4o) should perform better, although a bit slower. We believe the smaller models still perform excellently for most tasks, but you should test for yourself & your niche! 🧧 **CREDITS:** The AI model you choose will affect credits. * Claude 3 Haiku & GPT-4o-Mini = 1 credit * Claude 3 Sonnet & GPT-4o = 2 credits ### Generate cluster view with metrics? If you select yes, you will get **both** the normal view with topics, subtopics, and microtopics, **as well as** a completely new map 'view.' This view will show a bullet point list (with child lists), with... * Broad clusters of topics, with a meta title example and URL structure (hard silos) * Subtopics, with a meta title & URL structure * A primary keyword & related keywords for each subtopic * Global monthly search volume for each keyword, pulled from Google Keyword Planner * An AI-generated 'relevance' score. **NOTE:** The search volume metric also includes Clickstream data, in addition to Google Keyword Planner data. 9 times out of 10, this doesn't make a difference. For some keywords (such as NSFW keywords that Google might not show data for at all), this might give a more accurate search volume. 🧧 **CREDITS:** Generating a cluster view increases map credit cost. ### Edit subtopics before generating map? If you select yes, the AI will generate a short list of subtopics (under your main niche topic), that you can edit. You can... * Edit subtopic text (as well as the parent topic text) by clicking the text and typing * Remove subtopics (using the delete icon) * Add a subtopic manually (it'll have placeholder data which you can click to edit) * Generate more subtopics with AI (which you edit further) 🧧 **CREDITS:** Generating more subtopics marginally increases the map credits used.