> For the complete documentation index, see [llms.txt](https://tembi.gitbook.io/tembi-knowledge-base/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://tembi.gitbook.io/tembi-knowledge-base/product-news/ai-agents/prospecting-agent.md). # Prospecting Agent The Prospecting Agent helps your sales team figure out whether a webshop is worth pursuing, and why. Rather than manually piecing together signals from the webshop, you get a structured assessment the moment you open a webshop - grounded in your organisation's own definition of an ideal customer. This means your team spends less time researching and more time reaching out to the right accounts, with the right angles.
### What you see when you open a webshop When you open any webshop in Tembi, and activate the Prospecting Agent, it runs and returns an assessment in the webshop's AI summary view. Here's what's included: **ICP fit score** \ A numerical rating - for example, 7/10 - that reflects how well the webshop matches your ideal customer profile. The score is generated based on the criteria and priorities your admin has defined, so it's specific to your business rather than a generic ranking. **Plain-language summary** \ A concise overview of the webshop covering the signals most relevant to prospecting: what platform they're on, which delivery providers they use, their product positioning, OOH options, and any cross-border or growth indicators. The goal is to give you a fast, structured picture of who you're looking at - without having to navigate multiple tabs manually. **Reasoned assessment** \ Below the score, the agent explains *why* it rated the webshop the way it did. Each point in the reasoning is tied to a specific signal - for example, platform alignment, current delivery partners and methods, or industry fit - so you understand what's driving the score and can use that context when you reach out. **Data source transparency** \ The assessment shows which data points were used in the analysis and flags any that weren't available. This helps you understand the confidence level behind the output and know where you might want to do additional research before engaging. ### Generating outreach from the assessment Once you've reviewed the assessment, you can generate outreach directly from the same view. Two formats are available: **Email draft** \ A ready-to-send prospecting email built around the specific webshop's setup, the fit rationale, and your commercial context. It's not a generic template - the content reflects what the agent actually found, so it reads like it was written with this account in mind. **Pitch outline** \ A set of structured talking points you can use to prepare for a call or meeting. This covers the strongest fit signals, the most relevant commercial angles, and any context worth addressing upfront. Both formats can be generated in your preferred language, making it straightforward for local teams or those doing cross-border outreach to work from the same setup. ### How the agent is configured The Prospecting Agent is set up by your organisation's Tembi admin through a guided configuration flow in Agent Studio. It works by answering a series of questions that help the agent understand your ideal customer profile - no technical knowledge required. These questions cover things like: * What types of webshops you target (industry, size, geography) * Which platforms or delivery configurations signal a good fit * Where your organisation tends to win, and whys * Any commercial context that should inform how fit is assessed The answers to these questions are used to build the agent's reasoning logic. Every time the agent assesses a webshop, it applies that logic consistently - so the quality and approach of your prospecting doesn't depend on who's doing it. If your ICP evolves, your admin can update the configuration and all future assessments will reflect the change. *** ### *A practical example* Here's how this plays out in practice. You (or your team) create a list of prospects, your're researching a new webshop - say, an Italian sportswear brand operating on Shopify. You open them in Tembi and the Prospecting Agent returns a 7/10 score. The reasoning explains that they're on Shopify (which aligns with your target platform), that they use Klarna (signalling a modern, flexible checkout adopted to consumers), current delivery price is withing your target and they don't offer parcel lockers (which you do). The agent also notes that their industry and product category are a strong match for the solutions you offer. From there, you click to generate an email draft. It references their setup, the delivery gap, and frames the outreach around the specific opportunity. You review it, make a few adjustments, and copy paste the content to your email client. That whole process - from opening the webshop to having a draft ready - takes a minute rather than the better part of a half hour. *** ### Getting started The Prospecting Agent is available to all Tembi users once a template has been configured by your admin. * **Admins**: Go to Agent Studio and select **+ New Template** under Prospecting Agent to begin. You'll be guided through a set of questions about your ICP and commercial context. * **End users**: Open any webshop in Tembi and look for the AI summary view. If it shows "Not configured," reach out to your admin to get the agent set up for your team. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://tembi.gitbook.io/tembi-knowledge-base/product-news/ai-agents/prospecting-agent.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.