Office & managementVerified Real deployment at a named or precisely described company, with a measured result from a primary source.

AI drafts answers to customer questionnaires and RFPs from a library of approved answers and proof documents (Haworth, Ingredion, Cuisine Solutions)

Added 2026-10-27 · Last updated 2026-10-27

Manufacturers in this case: Cuisine Solutions · Haworth · Ingredion · all manufacturers using AI

Role

Office (sales support, sustainability and quality: answering customer questionnaires, RFPs and information requests) · Typical users: sales or customer service representative, proposal or bid manager, sustainability (ESG) manager, quality manager, the experts who are asked the same questions again and again

The problem

Customers keep sending questionnaires: sustainability surveys (EcoVadis, CDP, their own supplier forms), RFPs, supplier qualification forms, and detailed technical questions. Many questions repeat from one customer to the next, but the answers and proof documents are scattered, so each questionnaire means emailing the same experts, digging up certificates and copying last year's answers, which may be outdated. At Haworth, an office furniture manufacturer that sells through more than 600 independent dealers, dealers used outdated files and emailed corporate experts for answers, those experts were flooded with repeated questions, and proposals were delayed. At Cuisine Solutions, a sous-vide food producer, several sustainability questionnaires a year had to be answered without a central database or proper collaboration between departments. At Ingredion, data for EcoVadis, which rates both the group and individual plants, was collected largely by hand, with no central place for proof documents.

How AI helps

The company builds one library of approved answers, data and proof documents (policies, certificates, energy and waste figures, product data sheets). When a questionnaire or RFP arrives, the AI matches each question to the library and drafts an answer with the supporting document, and assigns the questions it cannot answer to the right expert. A conversational assistant lets sales people or dealers ask a question and get an answer with its source. Each new answer that people approve goes back into the library, so the next questionnaire is faster. People review and approve every answer before it is sent.

Step by step

1. Collect the last 5 to 10 questionnaires and RFPs you answered, with your final answers, and count how many questions came back more than once.
2. Build the first answer library from them: one approved answer per recurring question, with its owner, its proof document and a review date.
3. Put the proof documents in one place: certificates (ISO 9001, ISO 14001), policies, energy, water and waste figures, product data sheets.
4. For the next questionnaire, let the AI draft from the library and mark each answer as found in the library, adapted, or missing; send the missing ones to the right expert.
5. Have the owner approve every answer before it goes out, and add each new approved answer to the library.
6. Review the library every year, or when a certificate, figure or policy changes, so that old answers are not reused.

Tools

• Responsive (AI response management with a content library and the Ask assistant): used at Haworth
• Sunhat (AI platform for sustainability questionnaires, ratings and proof documents): used at Ingredion and Cuisine Solutions
• Other tools to evaluate: Loopio and other RFP response tools, and the questionnaire features of sustainability platforms
• For a small company, a first version: a shared folder of approved answers and proof documents plus any LLM (ChatGPT, Claude, Copilot, Gemini) with the prompt below, keeping confidential data in an approved company tool

In the AI tools index: Responsive · Sunhat · all AI tools used by manufacturers

As reported by the source; most figures are vendor or self-reported and not independently verified.

Human review: cautions

• A questionnaire answer is a commitment to a customer: the owner must approve every answer, and the AI must not invent a figure, a certificate or a policy.
• An answer library goes stale: review dates and expiry dates matter, or last year's figures get reused.
• Sustainability ratings such as EcoVadis check your proof documents, not the quality of the wording: the tool speeds up the work but does not improve the practices behind the answers.
• Figures are published by the vendors (Responsive, Sunhat); the Haworth time saving is for dealer partners, and none of the results is independently verified.

You are a proposal and sustainability coordinator at a manufacturer.

I will paste: (1) a new customer questionnaire or RFP [list of questions]; (2) our library of approved answers from past questionnaires [question, approved answer, owner, date approved, proof document name]; (3) a list of our proof documents [certificates, policies, figures, with dates].

Do this:
1. For each new question, find the closest approved answer and label it: reuse as is, adapt (say what to change), or missing.
2. Draft the answer for 'reuse' and 'adapt' questions, and name the proof document to attach.
3. Flag answers older than 12 months, certificates past their expiry date and figures from an old year, so the owner updates them first.
4. For 'missing' questions, write the question to send to the right owner, in one sentence.
5. Give a summary table: question number, status, owner, proof document.

Never invent a figure, a certificate or a policy that is not in what I pasted. Every answer is reviewed and approved by its owner before anything is sent to the customer. Use only the data I paste in; if something is missing, write 'not found' or list it as a question instead of guessing. Do not send, order, post or change anything in any system: produce drafts for my review and approval.

New questionnaire:
[paste]

Related cases

• AI reads customer tenders and specification packages clause by clause and builds the deviation list (ROBEL, Bosch Manufacturing Solutions)
• AI answers customers' 'where is my order?' emails from live ERP data (promotional-products maker, medical device maker)
• AI supplier contract review and renewal tracking
• AI screens and schedules hourly plant applicants around the clock, so recruiters stop living on the phone (Twin City Fan, General Motors, Nestlé)

All 11 Office & management use cases →

Sources

• https://www.responsive.io/customer-stories/haworth
• https://www.getsunhat.com/customer-stories/ingredion-datamanagement
• https://www.getsunhat.com/customer-stories/cuisine-solutions-sustainability-data-efficiency-increase