The letter of intent is signed, the champagne is back in the fridge, and a buyer's team opens your data room. For the next two or three months, their job is to find out whether the company works the way you described it.
Most founders picture diligence as an audit of the financials. In a software company, the financial review is only one of several workstreams running at once, and the others are where first-time sellers tend to be caught unprepared. I spent five years as a corporate lawyer working on these transactions before I spent about 25 years building technology companies, and the request lists have not changed much.
1. The data room is read as a signal
Before anyone reads a contract, the buyer forms a view from how the data room is organised. A clean index, documents that answer the question asked, and requests turned around in days tell them the business is run with discipline. Folders of unlabelled exports and answers that take three weeks tell them the opposite, and they start looking harder.
Speed matters for another reason. Every week diligence runs long is a week in which your results can soften and the buyer's enthusiasm can cool.
2. Technical diligence: the code and the people who maintain it
A serious buyer will have engineers review the product, sometimes their own and sometimes an outside firm. They look at the architecture, how much technical debt sits in the code, how releases are tested and deployed, and whether the documentation would let a new developer become productive.
The question underneath all of it is who can maintain this. If one senior developer is the only person who understands the core of the product, the buyer sees a key-person risk as real as founder dependence, and it will show up in retention terms or in the price. Documented systems and a second person on every critical area are worth more than another feature on the roadmap.
3. Security and privacy
For a company that holds customer data, security is now a standing item, not an afterthought. Expect questions about who has access to production systems, how credentials are managed, whether there has been an outside security assessment, and what happened in any incident you have had.
Privacy sits beside it. A buyer will ask what personal information you collect, where it is stored, and how you meet federal privacy law, and Quebec's stricter regime if you have customers there. A short, honest written answer prepared in advance is far better than one improvised under deadline.
4. Customer diligence and the reference calls
Near the end, most buyers want to speak with some of your customers. Founders tend to dread this, but handled well it is one of the strongest parts of the process: a customer who explains in their own words why they renewed is more persuasive than any deck.
Handled badly, a call can reveal a pricing complaint or a competitive evaluation the buyer did not hear about from you. Choose the timing with your advisor, prepare the customers who agree to speak, and make sure nothing they are likely to say contradicts what you have already disclosed.
5. Run your own diligence first
The best preparation is to do the buyer's work before they arrive. Build the data room early, using a typical request list as the outline. Have someone independent look at the code and your security practices, and fix what they find. Read your largest customer contracts again. Where a problem cannot be fixed in time, write it up and disclose it early, because a disclosed issue is negotiated calmly and a discovered one is negotiated with suspicion.
The takeaway
In a software sale, diligence tests whether the company works without surprises: the code, the people who maintain it, the data it holds and the customers who pay for it. None of that can be fixed in the six weeks after a letter of intent. A founder who has run the buyer's checks a year ahead usually goes through diligence on schedule, at the price agreed.



