The most expensive place to find a defect is at your customer.
Every vehicle programme creates defects. You cannot avoid that. You can only decide where you find them.
The vehicle is becoming a software system.
What used to be mechanical is now written in software. And it has to work correctly before the customer ever uses it.
Software is moving into type approval.
MoRTH has proposed new rules. They would make cyber security and software updates part of type approval.
Rule 125-T (AIS-189) would apply to passenger vehicles, commercial vehicles and tractors that have at least one ECU. Rule 125-U (AIS-190) would also cover trailers and semi-trailers.
Warranty is where a quality problem reaches your accounts.
New manufacturers pay more for warranty than established ones. This is true for two-wheelers and for cars.
Cars and SUVs are 69% of all vehicles recalled in India.
Between July 2012 and April 2024, 23 manufacturers recalled 56.7 lakh vehicles. And in India, a recall is no longer voluntary.
A service problem can become a regulatory problem.
An Indian electric two-wheeler manufacturer recalled 1,441 vehicles in April 2022, following a field incident. Within two years, 10,644 consumer complaints had reached the national consumer-protection system, and in October 2024 the Central Consumer Protection Authority acted on them.
A service problem can become a market-confidence problem.
Business Standard, April 2022 and October 2024A $2.0 billion recall charge.
A major global automaker booked approximately $2.0 billion in charges related to a battery recall, with the cell supplier agreeing to reimburse up to $1.9 billion.
Once a defect reaches the field, responsibility becomes expensive — regardless of where the defect originated.
CNBC, reporting company investor disclosures, October 2021Serious manufacturers test before the hardware exists.
“This shift-left in our software development process enables us to test earlier and more often at a much lower cost using virtual devices.”Dr. Céline Laurent-Winter
Vice President, Connected Vehicle Platforms
BMW Group · 2025
“We are keenly exploring the opportunities in Software Defined Vehicles, including the enabling vehicle architecture and technology stack.”
Girish Wagh · Executive Director, Commercial Vehicles, Tata Motors“EVs are radically different from ICE vehicles and are a relatively new technology, so it is important to thoroughly test all the components.”
Swapnil Jain · Co-founder & CTO, Ather Energy“Software is becoming increasingly crucial in vehicles, leading us to innovate in how we develop and validate it.”
Yves Bonnefont · Chief Software Officer, StellantisEvery one of those companies already had a validation organisation.
Tata Motors, BMW, Mercedes-Benz, GM and Lamborghini were not building a validation team. They already had one. They had test benches, tools, processes and experienced engineers. They were improving something that already existed.
You are asked for the same quality of evidence. But you are still building the vehicle, the supply chain, the factory, the service network and the team. You have a launch date. And investors are watching.
Three routes. Only one gives you validation capability without building the whole team first.
Build it yourself
Test benches, lab space, licences, automation and processes. And the engineers to run all of it.
- You keep full control, and the knowledge stays inside
- You spend the money before the first test runs
- Hiring and setting up take their own time, not your programme's time
- You use it least in the years when cash is tightest
Buy the tools
dSPACE, Vector, NI and MathWorks make excellent tools. That is not in question.
- Mature, proven, industry-standard technology
- But a tool is not a team
- You still need models, test development, automation and maintenance
- A test bench with nobody to run it is only a cost
Bring in a partner
- You bring: the ECU, requirements, software, product context
- We bring: infrastructure, engineers, simulation, HIL, automation, evidence
- We start on your first ECU programme, not your third
- We hand everything over when you want to bring it in-house
Why Validative
You are buying the ability to answer one question. Does this software behave as intended, and can you prove it?
Start where your programme actually is
These are capabilities, not a fixed order. Some programmes start with a virtual ECU, because the hardware does not exist yet. Others start with HIL, because the hardware is already on the bench and nothing is automated yet.
Your engineers stay on the vehicle. Our engineers stay on the testing. You can continue with us, move the work in-house, or do both together. We will not promise you a saving before we have seen your programme. Anyone who does is selling you a number, not a result.
Build your validation capability before you build production volume.
Start with one ECU, on one programme. Test the approach on real software, with a real deadline.
One ECU · one programme · no commitment beyond it