Automated quoting can be a busy engineer’s best friend. It’s easy, quick and convenient. But it can also conceal risk and lead to problems later in the production process, especially for complex part designs with strict performance requirements. AI agents can’t understand risk, nuance, trade-offs and creative solutions like their human counterparts.
The solution? Pair advanced technology with a human-first approach that applies real-world expertise to identify potential risks and opportunities.
Human-first engineering is a key component of Fathom’s “white glove” approach. It’s designed to reduce the frustrations OEMs can face when working with manufacturing partners – frustrations that slow projects, introduce risk and make already demanding jobs harder than they need to be.
When Does Auto-Quoting Work?
There is a place for fast, algorithm-driven quoting and basic design feedback. It tends to work best with:
- Simple (block-like) CNC prismatic parts, standard materials and forgiving tolerances
- Clear models and complete drawing packs
- Low-risk finishing and inspection
But it can fall short when tolerances are tight and compliance requirements are strict. Examples include:
- Tight geometric dimensioning and tolerancing (GD&T), complex datum schemes and functional fits that impact performance
- Thin walls, deep pockets and hard-to-hold geometries
- Any parts where heat treatment may cause distortion
- Complex assemblies or parts with critical interaction surfaces
Areas That Demand Caution During Auto-Quoting
Quotes That Are Too Low
Auto-quoting works well for simple parts. But it tends to fall short when evaluating more complex geometries and requirements.
Algorithms typically account for material, machine time and setup. But they may overlook costly factors such as non-recurring engineering (NRE), tolerance strategy and inspection planning.
The algorithm can’t necessarily anticipate that some parts demand perfect setup on the first attempt to hold tolerance. Even raw material pricing isn’t straightforward. Online rates don’t reveal supplier inventory depth, material availability and price volatility.
Other key considerations that the quoting algorithm may overlook include:
- Functional surfaces
- Heat treatment sequencing
- Deburring specifics
- Masking, inserts, cleaning and packaging
- Inspection programming and fixturing
Inspection is another frequent blind spot for auto-quoting algorithms. Complex parts may require dozens of programmed measurement points. A manufacturing engineer must define the CMM strategy and determine the best strategy for fixturing and gages and allocate proper time and cost for these steps.
Quotes That Are Too High
Auto-quoting platforms can also inflate pricing.
Sometimes this results when the automated tool chooses the wrong path and sticks with it. For example, an algorithm might route a part through multiple CNC workstations and fixtures, when a multi-axis solution or hybrid approach would be more cost-effective.
A human-first team asks better questions:
What tolerances truly need to be tight?
What finishes can be relaxed?
Is there a better routing strategy?
Which shop has the right equipment and expertise for the job?
Engineers also consider annual demand, revision cycles, alternate materials and efficiencies that are available later in the process. Simple design changes, such as draft angles and bend radii, can unlock significant cost savings.
Additional Risk Areas
Tolerance misinterpretation and datum errors: These emerge during assembly failures or inspection disputes. These troubles sometimes occur when the algorithm doesn’t fully distinguish prototype-friendly changes versus production adjustments. Print-first reviews, datum alignment and agreed-upon inspection methods can prevent such issues. They require discussion with your supplier’s engineering and production teams – not an algorithm.
DFM focused on cost over function: Auto-quoting systems often see a straight highway. Skilled teams see forks in the road. Multiple processes, such as EDM, milling, drilling and multi-axis machining, can create the same feature. Each method carries different cost, tolerance and lead-time implications. Algorithms may optimize machining cost while creating problems later, such as during assembly or inspection. Setup order, datum strategy and tolerance must all be considered. A human engineer can help you talk through the options and the trade-offs of each, so you make the most cost-effective decision.
Hidden manufacturability risks: A DFM report may approve geometry that later fails repeatedly in production. Experienced engineers and production people can anticipate how a part is likely to behave on the machine, not only if it’s theoretically machinable.
Auto-Quoting: Know Your Risks
Auto-quoting makes streamlined parts ordering possible. But schedules can slip when risks and potential problem areas aren’t fully evaluated. Production or inspection delays can unexpectedly extend the timeline of your project, jeopardize your product launch or cause a line-down situation.
The bottom line is that auto-quoting tools can’t replace a team of real people with deep knowledge and years of experience who can help you make the choices and trade-offs that will make your project a success.
