Poeton Industries completes successful AMRC sponsored AI trial – the rollout begins
Instro AI are collaborating with AMRC to help facilitate digital transformation in the manufacturing and engineering sector. Through a series of trials, participants can establish the efficiency and commercial benefits an AI tool can provide. Poeton, the UK's leading independent surface treatment expert has recently completed a trial.
Founded in 1898, Poeton Industries has become the UK's leading independent surface treatment expert providing coatings and treatments to components and materials, to improve properties like wear resistance, corrosion resistance or friction.
Instro AI Solutions develops tailored generative AI assistants for UK manufacturing which deliver instant, trustworthy answers that speed decisions, reduce errors, and free specialists for higher-value work.
AMRC, Advanced Manufacturing Research Centre are part of the UK's High Value Manufacturing Catapult network. They work with companies of all sizes (from SMEs to large firms) to develop, test and de-risk new manufacturing technologies and processes.
Norfolk County Council part funds the trials through the Growth Through Innovation Grant Scheme.
The issue
Poeton Industries handle up to 4,000 customer emails per month and receive around 3,000 RFQs each year enquiring how surface processes should be performed. As the specifications can vary greatly between different customers and different projects, a labour intensive, manual gap-analysis is undertaken each time by a full-time employee to identify differences and assess whether Poeton Industries can meet the requirement. Any mistakes in this manual process can lead to compliance risks and delays can impact commercials.
The issue
Poeton Industries handle up to 4,000 customer emails per month and receive around 3,000 RFQs each year enquiring how surface processes should be performed. As the specifications can vary greatly between different customers and different projects, a labour intensive, manual gap-analysis is undertaken each time by a full-time employee to identify differences and assess whether Poeton Industries can meet the requirement. Any mistakes in this manual process can lead to compliance risks and delays can impact commercials.
Objectives of the trial
- To explore whether Instro AI could demonstrate the potential to deliver ongoing benefits in the accuracy and speed of sourcing technical information
- To establish whether the overall speed and accuracy of technical customer email responses could be improved
The AI system deployed
Using a variety of customer, business system and process information, we developed a bespoke AI solution for Poeton Industries with 2 key interfaces:
- An email handler. Approved Poeton Industries users would create or forward an email to an Instro AI inbox which triggered a response containing any relevant information, along with a draft email response that could be sent onto a customer
- A web-based chatbot. A general purpose, Poeton Industries branded, secure password protected web app to respond to queries from users in Q&A, chat-based language
Specifically
- Reduce manual effort in specification review and gap analysis
- Speed up response times to technical customer enquiries and specification changes
- Improve accuracy and consistency in identifying compliance and capability requirements
- Support commercial growth by enabling faster and more reliable qualification of new business opportunities
The trial period
The live trial ran for a month, and data was gathered through catch up meetings and through a feedback mechanism set up to capture response accuracy, response completeness, and the time saved per query.
Source data used for the trial
To create the AI system, data was accessed from the following:
- ERP Data – data exported from Poeton Industries ERP System
- Customer Specifications – historic and current versions of customer orders/test specs
- Test Matrix – full test matrix of what processes and tests Poeton Industries currently performs
- Website data
- Business capability statements – internal documents describing Poeton Industries's current process capabilities, equipment and test availability.
The trial results
User engagement.
Significant usage and system engagement –
530
separate engagements over 26 working days (82 email & 448 chatbot)
Speed.
Customer email response time improved by
58.7%
saving a massive 8 minutes per customer email enabling more enquiries to be handled each day
"In just four weeks, Poeton Industries achieved 58.7% faster email drafting and strong accuracy ratings, proving the value of AI-assisted technical response handling."
Phil Sanders, Commercial Director, Instro
Completeness.
87%
of answers were deemed 'complete' by users ie. nothing missing and needed no manual interventions
"Phase 1 showed strong early value, especially in faster, more consistent technical responses, and the tool was well received by our teams. Phase 2 will scale that progress, deepen the use cases, and test how it embeds into daily operations. We're pleased to continue the partnership."
Lee Mason, Group Digital Transformation Manager, Poeton Industries
What this means for the business
The clear appetite to embrace the system from a user perspective is underpinned by a solid business case with time savings alone. In addition, there is a high level of confidence that technical responses will be thorough and can be relied on in terms of accuracy. And with the addition of more Poeton Industries source data post-trial and system integrations, accuracy and completeness levels are expected to be pushed beyond 90% for the rollout.
Does your business face similar issues accessing technical information?
Get in touch and find out if an Instro AI part funded trial could be right for you.
We use cookies to improve your experience. By using our site, you agree to our use of cookies. Learn more.