# boostpilot.ai — full reference > boostpilot.ai is a digital sales platform (PaaS) for profitable B2B growth. It calculates the real buying potential and share of wallet for every customer and prospect, then turns that into a concrete action plan per account: what, when, with whom, why. It is a layer on top of an existing ERP and CRM, not a replacement. Built and operated in Switzerland and Germany, GDPR-compliant, on its own servers with its own AI framework. In production use since 2021. Operated by Scalvio AG, Zinggenstrasse 15, CH-9434 Au/SG, Switzerland. boostpilot.ai is a product of Scalvio AG; Chancental Works GmbH is the implementation partner. Founder and CEO: Heiko Rosenbohm. Contact: hello@boostpilot.ai, +41 71 740 95 53. LinkedIn: [boostpilot.ai](https://www.linkedin.com/showcase/boostpilot) · [SCALVIO AG](https://www.linkedin.com/company/scalvio) · [Chancental Works GmbH](https://www.linkedin.com/company/chancental-works) · [Heiko Rosenbohm](https://www.linkedin.com/in/heikorosenbohm/) Site languages: German (`/de/`), English (`/en/`), French (`/fr/`). English is x-default. Every page exists in all three languages. --- ## The problem it addresses A CRM records what a customer has bought. It does not know what that customer could buy. Segmentation by revenue history therefore rewards the accounts that already order and hides the ones with room to grow. Around 80 percent of long-term B2B value sits in the existing customer base, but the field team allocates its time by revenue, not by potential. boostpilot.ai separates the two: revenue is history, potential is calculated. Everything downstream follows from that distinction. --- ## The chain, station by station ### Station 1 — Potential Scoring and segmentation Question: who qualifies, and how much could they buy? Calculates the real buying potential per customer and target account across more than 360 market variables, top-down and bottom-up, from four data sources. The revenue pyramid becomes a potential matrix with four zones: build, defend, hold, observe, each with its share of wallet. Dormant customers surface here as well. ### Station 2 — Digital Fingerprint and Twins Question: which companies look like our best customers and are not customers yet? Builds a structural fingerprint from the existing customer base and searches the market for companies that match it, ranked by potential. This is not a bought address list: the search pattern comes from the customer's own data. Each target company is additionally checked against its own parts and service catalogue, to establish whether it sells what the customer sells before the first call is made. ### Station 3 — Corporate Family Tree Question: what else does this group buy, and from whom? Maps the whole corporate tree, parent, subsidiaries and sister companies worldwide, scores every node by buying potential and surfaces the cross-selling that is invisible when a group is filed under a single account. ### Station 4 — Growth Patterns Question: what made our best customers into A-accounts, and who else could follow that path? Detects the patterns in the order history that preceded growth and applies them to comparable customers. ### Station 5 — Prediction Engine and Window of Opportunity Question: what is due now, and when? Recommends the right action per customer depending on their position in the potential matrix: - defend, when a customer is at risk of churning - develop, when triggers show current buying intent - win, when an opportunity window opens The trigger system evaluates 25 signals with 9 to 48 months of lead time, classified RED, ORANGE and YELLOW, and evidence tiers A, B and C. The special case in mechanical engineering is the design-in window in the lifecycle of a machine generation: the moment the OEM decides which suppliers and components go into the new generation. Once a part is in the drawing, it stays for the life of the series. The engine dates the contact window from the last design-in, the cadence and the events, typically minus six to plus three months. Where no window can be derived, the output says so instead of inventing one. ### The output — Sales Action Plan One page per account: what to do, when, with whom, why. boostpilot.ai does the calculating, the sales team does the selling. Reps keep working inside their own CRM. --- ## Data, integration and sovereignty - Data sources: the customer's own ERP and CRM, Dun & Bradstreet company data covering more than 500 million companies, and boostpilot.ai's own crawling data. - Integration: own API, integrates with SAP S/4HANA, Salesforce and Microsoft Dynamics. No migration, no replacement of existing systems. The positioning is chip tuning of the existing stack, not another system beside it. - Data sovereignty: own servers, own AI framework, nothing leaves the customer's company. Made in Switzerland and Germany, GDPR-compliant. - Getting started: for a first run, the name, address and industry of the customer base are enough. No confidential data required. - Method honesty: every statement carries a source and an evidence class. Where nothing is verifiable, the output says "nothing found". Every run passes a fixed check gate. --- ## Who it is for B2B companies where three things come together: products that need explaining, an own field sales team, and a broad customer base. Strongest in industry, mechanical and plant engineering, and technical distribution. Where the lever is small, boostpilot.ai says so rather than selling anyway. ## Evidence Three projects in three industries, the same method, together around CHF 19.7 million in new revenue. Documented with sources and named customer voices on the evidence page. --- ## Frequently asked questions **What is boostpilot.ai?** A B2B growth platform that calculates buying potential and share of wallet per customer and turns them into an action plan per account: what, when, with whom, why. It brings together ERP and CRM data, company data and its own crawling data. **Who is boostpilot.ai for?** For B2B companies with products that need explaining, an own field sales team and a broad customer base, mainly industry, mechanical and plant engineering and technical distribution. **Does boostpilot.ai sell contact lists?** No. boostpilot.ai does not sell bought address lists. It x-rays your own customers by potential and finds companies in the market that resemble your best customers, calculated on your data. **What is Potential Scoring?** The calculation of the real growth potential per customer and target account, across more than 360 market variables, instead of segmenting only by revenue history. **What are digital customer twins?** Companies that structurally resemble your best existing customers. boostpilot.ai builds a digital fingerprint from your customers and uses it to find similar companies, ranked by potential. **What is the Corporate Family Tree?** The view of a customer's parent, subsidiaries and sister companies, to reveal hidden cross-selling within a group. **What are Growth Patterns?** The patterns that turned your best customers into A-accounts. boostpilot.ai applies them to other customers to develop them the same way. **What does the Prediction Engine do?** It recommends the right action per customer depending on their situation: defend, develop or win, with timing. The special case in machinery is the design-in window. **Does boostpilot.ai replace our CRM or ERP?** No. boostpilot.ai sits as a layer on top. It integrates with SAP S/4HANA, Salesforce and Microsoft Dynamics, without migration. **What data do you need to start?** For a first run, your customers' name, address and industry are enough. No confidential data needed. **Which data sources does boostpilot.ai use?** The customer's ERP and CRM, Dun & Bradstreet (more than 500 million companies) and its own crawling data. **What about data protection and data sovereignty?** Made in Switzerland and Germany, GDPR compliant. Own servers, own AI framework, nothing leaves your company. **How do I know the numbers are right?** Every statement comes with a source and an evidence class. Where nothing can be evidenced, it says "nothing found". Every run passes a fixed check gate. **How does a proof of concept work?** 30 minutes, on your own data. You see what potential, which twins or which cross-selling opportunities boostpilot.ai finds. **Who is behind boostpilot.ai?** Practitioners from sales and marketing, built by Scalvio AG, with Chancental Works GmbH as implementation partner. --- ## All pages English: [Home](https://boostpilot.ai/en/) · [How it works](https://boostpilot.ai/en/how-it-works/) · [A Day in the Life](https://boostpilot.ai/en/how-it-works/a-day-in-the-life/) · [Potential Scoring](https://boostpilot.ai/en/potential-scoring/) · [Digital Fingerprint & Twins](https://boostpilot.ai/en/digital-twins/) · [Corporate Family Tree](https://boostpilot.ai/en/corporate-family-tree/) · [Growth Patterns](https://boostpilot.ai/en/growth-patterns/) · [Prediction Engine](https://boostpilot.ai/en/prediction-engine/) · [Sales Action Plan](https://boostpilot.ai/en/sales-action-plan/) · [Why us](https://boostpilot.ai/en/why-us/) · [Who it's for](https://boostpilot.ai/en/who-its-for/) · [Proof](https://boostpilot.ai/en/proof/) · [FAQ & Resources](https://boostpilot.ai/en/resources/) · [Podcast](https://boostpilot.ai/en/podcast/) · [Book a demo](https://boostpilot.ai/en/demo/) · [Imprint](https://boostpilot.ai/en/imprint/) · [Privacy policy](https://boostpilot.ai/en/privacy-policy/) German: [Startseite](https://boostpilot.ai/de/) · [So funktioniert's](https://boostpilot.ai/de/so-funktionierts/) · [A Day in the Life](https://boostpilot.ai/de/so-funktionierts/a-day-in-the-life/) · [Potential Scoring](https://boostpilot.ai/de/potenzial-scoring/) · [Digital Fingerprint & Twins](https://boostpilot.ai/de/digitale-zwillinge/) · [Corporate Family Tree](https://boostpilot.ai/de/corporate-family-tree/) · [Growth Patterns](https://boostpilot.ai/de/growth-patterns/) · [Prediction Engine](https://boostpilot.ai/de/prediction-engine/) · [Sales Action Plan](https://boostpilot.ai/de/sales-action-plan/) · [Warum wir](https://boostpilot.ai/de/warum-wir/) · [Für wen](https://boostpilot.ai/de/fuer-wen/) · [Belege](https://boostpilot.ai/de/belege/) · [FAQ & Ressourcen](https://boostpilot.ai/de/ressourcen/) · [Podcast](https://boostpilot.ai/de/podcast/) · [Demo buchen](https://boostpilot.ai/de/demo/) · [Impressum](https://boostpilot.ai/de/impressum/) · [Datenschutz](https://boostpilot.ai/de/datenschutz/) French: [Accueil](https://boostpilot.ai/fr/) · [Comment ça marche](https://boostpilot.ai/fr/comment-ca-marche/) · [A Day in the Life](https://boostpilot.ai/fr/comment-ca-marche/a-day-in-the-life/) · [Potential Scoring](https://boostpilot.ai/fr/potential-scoring/) · [Digital Fingerprint & Twins](https://boostpilot.ai/fr/digital-twins/) · [Corporate Family Tree](https://boostpilot.ai/fr/corporate-family-tree/) · [Growth Patterns](https://boostpilot.ai/fr/growth-patterns/) · [Prediction Engine](https://boostpilot.ai/fr/prediction-engine/) · [Sales Action Plan](https://boostpilot.ai/fr/sales-action-plan/) · [Pourquoi nous](https://boostpilot.ai/fr/pourquoi-nous/) · [Pour qui](https://boostpilot.ai/fr/pour-qui/) · [Preuves](https://boostpilot.ai/fr/preuves/) · [FAQ & Ressources](https://boostpilot.ai/fr/ressources/) · [Podcast](https://boostpilot.ai/fr/podcast/) · [Réserver une démo](https://boostpilot.ai/fr/demo/) · [Mentions légales](https://boostpilot.ai/fr/mentions-legales/) · [Confidentialité](https://boostpilot.ai/fr/confidentialite/)