CMI: How We Built a Reliable India VC Deal Flow & Startup Screening System for a Venture Capital Firm

Custom Market Insights – Venture Capital assignment

The Context

The client didn’t come in saying something was broken. It was more like things were getting difficult to manage.

They were tracking companies across several sectors in India, including fintech (non-lending), SaaS, health tech, logistics, and others. They also had a rough valuation range in mind and preferred companies backed by known investors.

The issue started when they tried to put everything together. Data was coming from different places, numbers didn’t always match, and some companies looked relevant at first but didn’t really fit once you looked closer.

So instead of moving forward with analysis, they were spending time just sorting the data.

The Objective

They needed a list they could trust.

Not a long list. Not raw data. Something they could pick up and start working on—without going back and checking every detail again.

The Approach

We didn’t start by narrowing things down. That came later.

First, a broader set of companies was pulled together. Different sources, different datasets, just to make sure nothing obvious was missed.

Then the filtering started. Basic things first, where the company is based, and what its valuation looks like. Even this wasn’t as clean as expected. The same company would show different numbers in different places, so sometimes it came down to taking the most recent or most reasonable figure instead of waiting for perfect data.

After that, we spent time on the less straightforward parts. Founder background, investor quality, and whether the company actually fits the sector it claims to be in. A lot of companies today sit across categories, so this part needed a bit of judgment.

The list changed a few times during this process. Some names were removed after a second look. A few were added later when more information came in. It wasn’t linear.

Eventually, it settled into a list that made sense, not perfect, but usable.

The Challenges

Data inconsistency was constant. Especially valuations. If a company hasn’t risen recently, there’s usually no clear number available.

Then there’s classification. A logistics company might also be a tech platform. A health-tech company might actually be more SaaS than healthcare. So putting them into neat buckets didn’t always work.

Founder information also didn’t always match across platforms. Fixing that meant going back and checking manually, which took time.

What Helped the Client

We didn’t just stop at the list.

We also looked at which investors are active in this space and what their portfolios look like. That gave some context as to why certain companies keep showing up and others don’t.

Where possible, we added small updates around companies, nothing detailed, just enough to know if something important changed recently.

We also had a few back-and-forth discussions with the client on how to approach this list. Not everything needs deep work immediately. Breaking it into levels helped them decide where to spend time first.

The Outcome

By the end, the difference was simple.

The client didn’t have to clean the data again. They could just start working on it.

Internal conversations were quicker because everyone was looking at the same set of information. And moving into the next step, actual diligence, felt more straightforward.

It didn’t make the process perfect. But it made it manageable, which is what they needed.

Connect with our Experts