Automated product modeling for CPQ
Model updates that took weeks now take seconds.
Modelbot retrieves your product master data, turns it into a rundown you can read, and generates a complete TCX product model straight into the CPQ ticket.
01What it is
Modelbot closes the gap between the product data you already maintain and the CPQ model your sales team configures from.
It retrieves master data from your PIM and ERP systems — and from the CPQ ticket itself — processes it into a rundown you can read before anything is built, then generates a complete TCX product model with part structures, a configuration GUI and constraints. The model and its rundown are uploaded back to the ticket, with a business log for the overview and a technical log for the detail.
- Master data
- PIM · ERP · CPQ ticket
- Modelbot
- rundown → TCX
- CPQ product model
- parts · GUI · constraints
02The cost of doing this manually
Hundreds of hours go into work nobody should be doing by hand.
Four problems with the way product modeling is done today, and what Modelbot does about each of them.
Manual and error-prone
Someone reads the specification and retypes it into the CPQ tool. Every transcription is a chance for the model to drift from the real product.
Generated from the source records
The model is derived from your master data. Nothing is retyped, so nothing is mistyped.
Time-consuming
Modeling takes days or weeks, and it takes them from the people who should be working on the product itself.
Seconds, while the ticket is open
The same work runs in seconds. Your specialist reviews the rundown instead of producing it.
Does not scale
Product ranges change constantly. Every new variant or market is another modeling project queued behind the last one.
Catalogue size stops mattering
Modelbot runs per ticket, so a growing range means more runs rather than more projects.
Inaccurate quotes
Manual slips and outdated models produce quotes that damage margin first and client trust second.
Quotes that match the product
Models are regenerated from current master data, so what sales configures is what the factory can build.
03How it works
One run, four stages, nothing hidden in the middle.
Every stage leaves something a person can read. The rundown is the checkpoint; the two logs are the receipt.
-
01
Retrieve master data
Modelbot reads from wherever your product data lives — PIM and ERP systems, and the CPQ ticket itself.
-
02
Build the rundown
The master data becomes a rundown in Excel format — readable, checkable, and the last point before anything is generated.
-
03
Generate the TCX model
Rune, a package from cpqbot, turns the rundown into a complete TCX product model: part structures, constraints and a configuration GUI.
-
04
Upload and log
The model and its rundown are uploaded to the CPQ ticket, with a business log for the overview and a technical log for troubleshooting.
Data handling
Product models and their data live exclusively in CPQ. Modelbot holds them in memory only while a model is being built and never writes product data to disk. All communication with CPQ uses the OAuth 2.0 authorization code flow.
04Why Modelbot
Six reasons to stop building models by hand.
-
01
Low total cost of ownership
No modeling headcount to hire and train, and no consultancy fees to renew.
-
02
Accuracy
Human error enters during manual updates. Modelbot removes the step where it enters.
-
03
Long experience
Built by people who modelled products in CPQ systems by hand for fifteen years before automating it.
-
04
Scalability
As your product range grows, Modelbot adapts to it. The size of the catalogue stops being a scheduling problem.
-
05
Efficiency
Model updates that took days or weeks complete in seconds.
-
06
Transparency
A readable rundown before generation and two logs after it. You can always see what was produced and why.
05Customer
It’s easier for us to implement new products. When we want to approach new markets it is definitely a plus.
06Getting started
Three stages, and you can stop after any of them.
Nothing here asks for a commitment before you have seen Modelbot run against your own data.
-
1
Book a meeting
A conversation with our team about your product data, your CPQ setup, and whether Modelbot fits what you are trying to do.
-
2
Proof of concept
A kickoff workshop maps where your data lives, what format it is in, and where it needs to end up. We then build a prototype on sample data from your own ERP and demonstrate Modelbot building or extending a real product model from it.
-
3
Implement
A full implementation that syncs your sales tool and CPQ with your ERP, automated end to end if you want it. Modelbot keeps developing, and you stay on the current version.
Proof of concept and implementation are billed by the hour. Running Modelbot after launch requires a subscription.
07Questions
What technical teams ask first.
01 Is Modelbot secure? Our IT department has strict requirements.
All communication between Modelbot and CPQ uses the OAuth 2.0 authorization code flow. Two-factor authentication is supported and can be enforced for everyone in the organisation, and single sign-on works through Microsoft Entra.
02 Where does Modelbot store our product data?
It does not. Product models and their related data live exclusively in CPQ. Modelbot keeps them in memory only while a model is being built, and never writes product data to disk.
03 Our ERP holds only part of the information. Where does the rest come from?
From a Base Model or Base Rundown — where Modelbot keeps everything that stays constant or simply is not in the master data. That is the backbone of the product structure: the GUI with its question lists, sections of the part structure, and global constraints.
The base is merged with your ERP master data to produce the complete model.
04 How long does it take to get Modelbot running?
With a CPQ system already in place, one to two months. The exact duration depends on your organisation and on which systems hold your master data.
If you are at the start of your CPQ journey, we implement Modelbot alongside the CPQ rollout so both are ready together.
05 Can you also automate price data, cost data or document templates?
Those import through CPQ’s standard integration APIs. Modelbot itself is dedicated to product models, but we work with partners such as cpq.se on the rest of the integration work.
06 How reliable is it?
Modelbot runs on robust infrastructure with proactive monitoring so models can be built around the clock. You do not have to take our word for it — our status page shows the operational history of our services for the past 90 days.
08Contact
Tell us about one product family.
Describe what you would want to model first and we will come back to you.
What happens next
- 01We read what you sent and look at it against a real product family.
- 02A technical conversation about your data, your CPQ setup and whether this fits.
- 03If it does, a proof of concept built on your own ERP data.