Will AI Replace Contract Professionals? The Future of Tender Analysis
Blog post description.
CONSTRUCTION BLOGS
If AI can function in the following scenario, the majority of contract professionals will be replaced.
Imagine you receive a large tender to bid. It includes civil works, steel works, MEP, finishing, commissioning, and many other components. You give it to your different teams to assess whether your firm can quote for this tender. Your team begins to study the tender, and different sections take their respective parts. It will take a minimum of 3 to 4 days to estimate whether the company should quote the bid. It is possible that some teams may advise leaving it, while others may recommend proceeding. In this case, you become the sole decision-maker to decide whether to go ahead or not.
The biggest fear roaming in your mind is: if we put in so much effort and in the end we lose, then all the effort will go in vain. What if, in the middle of preparing this tender, a new tender, one for which we are more eligible, is released, and we ignore it? And in the end, we don’t win either of them? That would be a huge loss of time and effort. In conclusion, the biggest fear in your head is the loss of time, effort, and opportunity.
Now we all know that AI has moved ahead from its initial stage and has entered the Agentic AI era, in which AI agents are trained to perform specific tasks. Now imagine an engineer has developed an Agentic AI system or platform where multiple agents, each designed to perform a specific task, work together in a single room. You have agents for quantity surveying, tender engineering, and finance. All these agents are trained with company-specific data. They know the technical and financial capabilities of the entire company.
What you do is upload the entire tender into the platform and ask them to analyze it individually according to their expertise, then discuss with each other and suggest the most appropriate option that is beneficial for the company. How much time will it take? Two hours, maybe three. And you will get a data-backed result to instantly decide whether to proceed or leave it.
Now rewind to the first option, where the evaluation was done manually. You have a lot of staff on your payroll for each task. There is a high probability that their decisions may be biased due to emotional influences or surrounding pressures. As a result, the outcome you receive might be subjective or even manipulative.
How much would agents cost? At least far less than the previous setup. You would probably need only one engineer to review the findings of the agents to confirm their compliance with the actual data, and that’s it.
It may sound like imagination to many, but if you have a little investigative mindset, you can firmly say that this reality is not far behind.