Photo: Ron Lach / Pexels
Implementing artificial intelligence for contract review costs at least $1,000, estimates Valeryan Brunin, head of IT company ILAVISTA (digitalbusiness.kz). He says the benefits should be measured in the value of staff time or sales growth. This means choosing a task and setting a target before implementation begins.
What happened
Brunin outlined which tasks businesses operating offline can hand over to AI and how to assess the results. He has seen more companies seeking help with implementation, but many come without a specific task in mind. The expert suggests starting by mapping out workflows and identifying routine work that takes up specialists’ time.
One example is a construction company. Five accountants manually processed certificates recording the percentage of work completed, each handling 10–20 documents a month. At half an hour per document, this took a total of 25–50 hours a month.
Another example is comparing versions of contracts. After negotiations, a lawyer checks whether the agreed terms have made it into the final document. AI can highlight differences, while the value of this automation depends on the volume of documents and the lawyers’ workload.
What this means for costs
Brunin estimates that setting up contract review requires at least 40 hours of a specialist’s time at $25 an hour. That puts the starting cost at $1,000. He says this covers studying how the lawyers work, documenting the process, and integrating and configuring AI.
The expert estimated that building a training system for a sales department would cost around $10,000 and take three months. The first month goes into preparing the knowledge base, the second into implementation and the third into testing. He says every employee in the department needs to be involved through in-depth interviews to gather their knowledge.
How to measure returns
Brunin suggests choosing a way to measure results in advance. The first option is to calculate the time spent on a task and the employee’s hourly cost. The second is to track a specific metric, such as how many sales proposals turn into deals.
In his own work, he cut the time needed to prepare a sales proposal from eight hours to one. He says the proportion of deals closed then rose from one in ten to three in ten.
“In both cases, the target figures are set in advance,” Brunin stressed. The hours freed up still need to be directed towards tasks that deliver a financial return. For example, an accountant could review bank fees and suggest ways to cut costs.
When to assess the first results
After choosing a task, the next step is to implement it in small stages and measure the results. Brunin suggests assessing the initial changes every two weeks: checking the financial benefits, then moving on to the next stage.
