Speya researches suppliers for you. You describe what you need in plain language, and Speya searches across public sources, checks each company against your requirements, and returns a list where every answer links to the source it came from.
It works the same way whether you are looking for suppliers you have never heard of or researching the ones you already buy from.
In this article
Meet Speya
How Speya thinks
What you can do with Speya
How you work with Speya
How to read a Speya list
FAQ
Meet Speya
Speya is the chat you work in throughout the platform, and it is the orchestrator behind everything the platform does. You describe what you want in plain language, and Speya interprets the request, picks the right tool for the job, and runs it. You do not switch between tools yourself.
The platform is built around company qualification. Behind the chat sit your relevant tools. Describe the outcome, and Speya reaches for the right tool:
Discovery finds new suppliers from what you describe: a product, a service, or a set of capabilities.
Enrich researches suppliers across a list at once, one question per column.
RFx asks suppliers directly for the answers only they can give.
Monitor re-checks the columns you choose on a schedule and tells you when something moves.
Spreadsheet draws on the case description, the context you have shared, and the results to produce a document that tells the story behind the decision.
How Speya thinks
These things shape everything Speya does, whether you are discovering new suppliers, researching a base you already have, or asking a general question.
It works from public data. Speya reads across 250+ public sources: company websites, trade fair listings, certification registries, public company records, trade data, marketplaces, the open web.
It reasons over what suppliers publish; it does not keyword-match. Speya reads how suppliers describe themselves in whatever language and detail each one uses. That is why it surfaces small workshops, foreign-language sites, and suppliers with thin websites that a normal search would bury.
It thinks before it acts. Most of what Speya does begins with research and reasoning rather than a direct action. It studies the domain, plans the work, and only then executes. That is why context beats phrasing: the more you share about your goal and constraints, the better it performs. Write to it the way you would brief a colleague.
It keeps working toward your stated intent. Speya treats each result as a step, not an endpoint. It reads what has come back against what you set out to do, and offers the move that gets you closer, such as a deeper research pass, another column, contacts and outreach, or a summary to share.
Everything for the case is in one view. Because Discovery, Enrich, RFx, and Monitor all write to the same list, the full context of a case (who the suppliers are, what you have researched, what they have replied, and what has changed) is available to Speya at any moment. You do not have to re-brief or re-paste. When you ask a follow-up, Speya already has the case in front of it.
What you can do with Speya
Explore new suppliers and markets
You describe a category, a market, or a specific part. Speya turns that into a Discovery Plan: a search plus the requirements it will verify on every candidate. You edit the plan before it runs. What comes back is a researched list of companies, each backed by sources, that you can sharpen with further passes.
Start here whether you are entering a new product category, looking for an alternative to a current supplier, or mapping a market to understand the players and dynamics before you plan your strategy. What you get back is not only a shortlist but a picture of the market you can act on.
Work with your existing supplier base
Start by uploading your supplier list. Speya matches each company to its record, and then you ask questions across the whole base at once: certifications, ownership, risk signals, sustainability criteria. Answers come back as research columns with sources attached. Where a data point changes over time, set a monitor and Speya re-checks it on a schedule and tells you what moved.
Start here if you are cleaning up supplier data, screening a base against a framework, running due diligence, or watching for risk across a portfolio too large to review by hand.
Most teams end up using both. A risk finding on a current supplier turns into a search for an alternative. A supplier you discover today becomes part of the base you monitor next quarter.
How you work with Speya
You work with Speya in one of two modes. Which one fits depends on what you are doing.
Speya Auto. is the default when you start a new project. You give Speya a brief, answer a few follow-up questions so it can scope the case, and step away. Speya runs several searches in parallel, brings the candidates into one list, generates the research columns that fit the case, ranks the results into tiers, and writes a summary of what it found. You come back to a shortlist ready to act on. You can steer mid-run if you want to widen the search, add a column, or move on. Auto is for Discovery cases.
Step-by-step. puts you in control of each move. Use it for two things:
Researching a supplier base you already have. You ask specific questions across the base, one column at a time, and read what comes back before deciding what to research next. Auto is not built for this. If you are working with an existing base, work step-by-step.
Exploring new suppliers when you want to steer. You review the Discovery Plan before it runs, sharpen the list with further passes, and add columns as questions come to you. Useful when you are still framing the case.
Both modes work on the same list, so a Discovery case run in Auto can be refined step-by-step afterwards.
How to read a Speya list
A few things that help you read a Speya list well. They apply whether you are exploring new suppliers or working with a base you already have.
Every result has a source. Click any cell to see the reasoning behind the answer and the sources Speya used. Alongside the reasoning, Speya shows how strong the evidence is; when it is limited, that is usually a signal to ask the supplier directly rather than treat the answer as settled.
Match and no match. In Discovery, every supplier candidate is checked against the requirements you set. A match means Speya found public evidence that the supplier meets the requirement. A no match means Speya could not confirm it from available sources, often because the supplier simply does not publish that detail, not because they fail it.
Candidates are grouped into tiers. When a sourcing case completes, Speya reasons across the whole list, weighing the requirements and the evidence found for each candidate, and groups the results by how well each supplier matches what you set out to find. Tier 1 is the strongest match against your requirements; Tier 2 is worth reviewing; Tier 3 is where the fit is weaker or the available information is thinner. Tiers describe fit against your criteria, not how good a company is.
Results can vary between runs. This is the nature of researching across public sources: what is publicly available shifts over time, different framings of a search reach different parts of the market, and search agents work within a time budget rather than reading the entire web. Most cases capture what you need in a single pass; some benefit from a second pass framed differently. Because every prompt, instruction, and result is visible to you and to Speya, running additional passes is intentional work, a way to build a fuller picture of the market.
FAQ
How is this different from ChatGPT or Google?
They give you answers to verify yourself; Speya gives you traceable suppliers. One instruction researches every supplier to find a match, every result links to its source, and when Speya cannot confirm something it says so. Results are never ranked by who paid to appear. Speya reads what suppliers publish, which is why small and lesser-known suppliers surface rather than the best-marketed ones. You get a comparable list built for a sourcing decision, not a page of results or a paragraph to fact-check.
Do I need to write perfect prompts?
No. Speya researches your case before it acts, and it asks when it needs to narrow things down. Write to it the way you would brief a colleague: what you are sourcing, what kind of partner you want, and why you are looking. Context matters far more than phrasing, and you can steer the results in plain language at any point.
Can I use it in my own language?
Yes. You can write to Speya in over 200 languages with peak performance for English, French, German, Spanish and Portuguese. Speya reads foreign-language supplier sites for you, so your results are not limited to the language you searched in.
Is my work private?
Yes. Your supplier data, searches, and documents are never used to train any AI model. Everything is stored and processed in the EU, encrypted at rest and in transit, and covered by a signed DPA. The platform is ISO 27001 certified and SOC 2 Type II audited.
Within your workspace, supplier lists are shared with your team so colleagues can build on each other's research, while your chat with Speya stays private to you. For how sharing, roles, and access work, see How to manage your workspace. For the full picture on how your data is handled, see Security.
Can I bring in the suppliers we already work with?
Yes. Upload an Excel or CSV file, or add suppliers by name in chat, and research and monitor them like any discovered supplier. See How to import and track suppliers with Monitor.