[{"data":1,"prerenderedAt":267},["ShallowReactive",2],{"blog-post:introducing-signaldig:en":3},{"id":4,"title":5,"authors":6,"body":13,"category":252,"categorySlug":253,"date":254,"description":255,"extension":256,"locale":257,"meta":258,"navigation":259,"path":260,"published":259,"readingMinutes":261,"seo":262,"slug":263,"stem":264,"translationKey":263,"updated":265,"__hash__":266},"blog\u002Fblog\u002Fintroducing-signaldig\u002Fen.md","SignalDig: SEO, GEO and social research",[7],{"name":8,"to":9,"avatar":10},"SignalDig Team","https:\u002F\u002Fgithub.com\u002Fdaily-growth-signals",{"src":11,"alt":12},"\u002Fbrand\u002Fsignaldig-avatar.svg","SignalDig team avatar",{"type":14,"value":15,"toc":241},"minimark",[16,20,23,26,36,41,44,84,87,91,94,97,100,103,107,115,138,141,145,148,170,190,197,201,204,207,211,214,217,221,224],[17,18,19],"p",{},"SignalDig is a research product for people and AI agents. It brings SEO research, search results, GEO visibility, competitor research, backlinks and public conversations into a workflow that keeps the evidence close to the conclusion.",[17,21,22],{},"That matters because growth questions rarely arrive as neat reports. Someone may be deciding whether a topic deserves a new page, checking how a domain appears in AI search, comparing competitors for a query or trying to understand the language used in public conversations. The relevant evidence usually sits in different places, and the context does not always travel with it.",[17,24,25],{},"We built SignalDig to make that research easier to inspect. A result should show where it came from and when it was observed. It should also make clear what the data can support—and what still needs to be checked elsewhere.",[27,28,29],"blockquote",{},[17,30,31,35],{},[32,33,34],"strong",{},"Product update — September 1, 2026:"," SignalDig now organizes its public capabilities into five research areas and supports a complete Agent setup with an API key, MCP and matching Agent Skills. This article reflects the current product surface.",[37,38,40],"h2",{"id":39},"what-signaldig-can-research","What SignalDig can research",[17,42,43],{},"The product currently covers five areas:",[45,46,47,56,63,70,77],"ul",{},[48,49,50,55],"li",{},[51,52,54],"a",{"href":53},"\u002Ffeatures#search-keyword-research","SEO search and keyword research"," for keyword context, search demand, related queries and current search results.",[48,57,58,62],{},[51,59,61],{"href":60},"\u002Ffeatures#geo-ai-search-visibility","GEO and AI search visibility"," for inspecting how a domain or topic appears in supported AI-search observations.",[48,64,65,69],{},[51,66,68],{"href":67},"\u002Ffeatures#search-competitor-research","Competitor research"," for examining competing results and ranking visibility around a defined query.",[48,71,72,76],{},[51,73,75],{"href":74},"\u002Ffeatures#website-backlink-research","Backlink research"," for backlink and traffic-related observations about websites.",[48,78,79,83],{},[51,80,82],{"href":81},"\u002Ffeatures#social-content-demand-research","Social and content-demand research"," for public conversations and content signals from supported platforms.",[17,85,86],{},"These areas are related, but they are not interchangeable. Google is the default search engine for search research, and Bing can be selected when it is the source you want to examine. Social retrieval is based on supported public sources, not a complete survey of a platform. The product keeps those distinctions visible instead of turning different observations into one unexplained score.",[37,88,90],{"id":89},"how-a-research-request-works","How a research request works",[17,92,93],{},"The starting point is a question with enough detail to make the result useful. That might be a query and market, a domain and search engine, or a topic and social platform. Add a language, time range or other filter when it changes what you need to compare.",[17,95,96],{},"SignalDig then retrieves the relevant observation from its search, website or social-data capabilities. The output describes what was available at the time of retrieval. It is useful for deciding what to examine next, but it is not a permanent statement about a market.",[17,98,99],{},"The interpretation stays tied to the returned data. For example, a missing impression count does not mean that a post had no impressions; it means that the result cannot support a conclusion about impressions. Pagination, retrieval time and the selected source can change what the result tells you.",[17,101,102],{},"From there, the result can support a content test, a competitor comparison or another research question. The next step still belongs to the person or agent using the evidence. SignalDig makes the reasoning easier to review; it does not make the decision on the user’s behalf.",[37,104,106],{"id":105},"see-the-workflow-in-real-examples","See the workflow in real examples",[17,108,109,110,114],{},"The public ",[51,111,113],{"href":112},"\u002Fuse-cases","Use Cases library"," contains redacted examples based on defined retrievals:",[45,116,117,124,131],{},[48,118,119,123],{},[51,120,122],{"href":121},"\u002Fuse-cases\u002Fseo-research\u002Fseo-keyword-opportunity-demand-research","Researching keyword-opportunity needs with Reddit and X"," follows conversations across two platforms to identify demand questions and feature priorities.",[48,125,126,130],{},[51,127,129],{"href":128},"\u002Fuse-cases\u002Fsocial-research\u002Fdeepseek-hardness-x-discussions","Finding current DeepSeek Harness discussions"," shows what a bounded social sample can contribute to a content plan, along with the risks that remain.",[48,132,133,137],{},[51,134,136],{"href":135},"\u002Fuse-cases\u002Fsocial-research\u002Flinkedin-post-engagement-analysis","Analyzing a 95-post LinkedIn history"," shows why pagination, post age and missing impressions matter when reading engagement data.",[17,139,140],{},"These examples explain how the product handles a research task. They are not representative surveys, customer endorsements or promises of ranking, reach or revenue.",[37,142,144],{"id":143},"set-up-signaldig-for-an-ai-agent","Set up SignalDig for an AI agent",[17,146,147],{},"An agent needs three things to use SignalDig:",[149,150,151,154,162],"ol",{},[48,152,153],{},"Create a SignalDig API key in your account. It authenticates requests and attributes credit usage.",[48,155,156,157,161],{},"Configure ",[51,158,160],{"href":159},"\u002Fmcp#quick-install","SignalDig MCP"," in a compatible client. Start with the SEO Data server, then add Social Data or SEO Decisions when the task requires them.",[48,163,164,165,169],{},"Install the matching ",[51,166,168],{"href":167},"\u002Fskills#install","SignalDig Agent Skills",". Skills do not contain live data. They explain how to choose tools, set a sensible scope, preserve evidence and handle partial results.",[17,171,172,173,176,177,181,182,185,186,189],{},"For a manual installation, download the latest release ZIP from the ",[51,174,175],{"href":167},"Skills installation guide",", extract it and copy the Skill directories into the directory used by your client. Codex uses ",[178,179,180],"code",{},"~\u002F.codex\u002Fskills",", Claude Code uses ",[178,183,184],{},"~\u002F.claude\u002Fskills",", and other compatible clients can use ",[178,187,188],{},"~\u002F.agents\u002Fskills",". Restart the client or open a new session after copying the files. The guide includes exact commands and Windows instructions.",[17,191,192,193,196],{},"Cursor and VS Code users can use the one-click server import in the ",[51,194,195],{"href":159},"MCP guide",". Store the API key in the client’s protected credential field or an environment variable. Never put it in a repository, prompt, screenshot or shared log.",[37,198,200],{"id":199},"mcp-and-agent-skills-have-different-jobs","MCP and Agent Skills have different jobs",[17,202,203],{},"MCP is the connection between an AI agent and SignalDig’s current data and analysis tools. Agent Skills are open, inspectable instructions for using those tools. The API key authenticates a request; it does not decide what the agent should research or how it should interpret a missing field.",[17,205,206],{},"Keeping these parts separate means the data connection and the research instructions can be updated independently. It also lets users read the workflow before connecting live data.",[37,208,210],{"id":209},"what-signaldig-does-not-provide","What SignalDig does not provide",[17,212,213],{},"SignalDig does not replace first-party analytics, revenue records, customer interviews or direct experiments. It does not continuously monitor the entire web, guarantee complete platform coverage, or promise rankings, traffic, engagement or revenue.",[17,215,216],{},"Search, backlink and social results depend on the source, selected scope and freshness of the retrieval. When an important field is missing, its value is unknown—not zero.",[37,218,220],{"id":219},"why-we-are-building-it-this-way","Why we are building it this way",[17,222,223],{},"Good growth research is not just a matter of collecting more fields. The useful question is whether someone can understand the evidence well enough to decide what deserves a closer look.",[17,225,226,227,231,232,235,236,240],{},"That is the role SignalDig is designed to play: give people and AI agents a clearer starting point for SEO, GEO, competitor, backlink and social research, while keeping the evidence and its limits in view. Start with the ",[51,228,230],{"href":229},"\u002Ffeatures","five research capabilities",", explore the ",[51,233,234],{"href":112},"real use cases",", and ",[51,237,239],{"href":238},"\u002Fmcp","connect MCP with the matching Skills"," when you are ready to run a research task.",{"title":242,"searchDepth":243,"depth":243,"links":244},"",2,[245,246,247,248,249,250,251],{"id":39,"depth":243,"text":40},{"id":89,"depth":243,"text":90},{"id":105,"depth":243,"text":106},{"id":143,"depth":243,"text":144},{"id":199,"depth":243,"text":200},{"id":209,"depth":243,"text":210},{"id":219,"depth":243,"text":220},"Product","product","2026-08-12","SignalDig brings SEO, GEO, competitor, backlink and social research into one place, with evidence-backed use cases and setup guides for MCP and Agent Skills.","md","en",{},true,"\u002Fblog\u002Fintroducing-signaldig\u002Fen",8,{"title":5,"description":255},"introducing-signaldig","blog\u002Fintroducing-signaldig\u002Fen","2026-09-01","-Rr3g-XVQ0W1PrA_2WVmbdtm2vn4aYZAtukaYWZNLnI",1788262611126]