AI Won't Fix Slow RFQs Until the Product Logic Is Under Control
For complex B2B manufacturers, the next real AI opportunity is not another general-purpose chatbot. It is the quote-to-delivery workflow. And it only pays off if the product logic underneath it is already under control.
The pressure is regional, not abstract. In Germany, AI use in industry has reached 58.7%, the highest of any sector. Polish manufacturers face the same customer expectations, yet only 7% of Polish SMEs rate themselves ready to implement AI. The demand is here. The foundations mostly are not.
What the data shows
Tacton’s 2026 State of Manufacturing report, based on 280 manufacturing leaders across 8 countries and 13 sectors, puts numbers on the gap. Product complexity has hit a four-year high, with 67% now describing their products as very or extremely complex. CPQ adoption is climbing, up 19 points since 2022 to 46%. And yet 43% still name customization as their single biggest quoting challenge, and 62% report moderate to severe margin erosion between quote and delivery.
The clearest tell sits underneath all of it: only 7% define their configuration rules once and reuse them across every system. The other 93% re-synchronise by hand at each handoff. Every duplicated rule set is a divergence waiting to happen.
Why quote quality beats quote speed
This is the part most RFQ projects miss. The instinct is to make quoting faster, and that instinct is understandable. Customers want quick answers, sales is overloaded, engineering is the bottleneck. But speed is not the real problem. Quote quality is.
When product rules live separately in CPQ, ERP, PLM, MES, spreadsheets and people’s heads, AI just produces wrong answers faster. A quote can look polished and still be wrong on cost, feasibility, lead time or margin. For custom-order, low-volume manufacturers that is exactly where profit leaks: rework, change orders, manual checks, customer delays, production exceptions. Rarely one dramatic failure. Usually a hundred small ones.
Cisco’s June 2026 industrial AI report makes the same case from the infrastructure side. Manufacturers are trying to move from one-off pilots to repeatable scale, and that depends on secure networks, usable operational data and systems that actually connect. The sample skews large, toward firms above $100M in revenue, so read it as direction rather than a mirror of the mid-market.
What has to be true first
So the useful executive question is not whether to use AI in quoting. It is what has to be true before AI can quote safely. The answer lives in the digital thread between sales, engineering, supply chain and production: configuration logic, pricing rules, BOM structures, option-level demand, capacity constraints, delivery assumptions. Until someone owns that, AI is just another interface bolted onto weak process control.
The ownership splits cleanly by role. Sales engineering has to cut the manual interpretation between what a customer asks for and a valid configuration. Operations has to stop quote errors from becoming production problems. IT has to make CPQ, ERP, PLM and MES data usable across the whole cycle. None of that is platform shopping. It is integration and data work, the unglamorous layer that platforms assume you already have, and that internal teams rarely have the time to build.
The takeaway
AI can speed up RFQs. It will not rescue a fragmented quoting process. The manufacturers who benefit first are the ones who treat product configuration as a governed operating asset, not a pile of rules scattered across six systems.
Before funding broad AI pilots, fix the rules, the handoffs and the data behind the quote. That is where the margin is. It is also the part worth bringing in outside help for, because it is the part no CPQ licence does for you.
So, one question for your own operation: if a quote goes out today, who can actually vouch that it is buildable, priced right and on time, before it reaches the customer?
“Content generated using AI”
We create intelligent software and AI-driven solutions to automate workflows, modernize legacy systems, and sharpen your competitive edge.