Manny Del Val
Founder, KeptDo · Operator, Level 7 Capital
From IT to multifamily
I've worked in IT my whole career — from systems engineer to Director, most recently as an IT Manager. Twenty-five years of building systems, leading teams, and solving the kind of problems where you can't hand-wave the details.
In 2024 I really got into AI. Not the demo-day AI — the kind that solves real business problems. And around the same time I shifted my own focus from residential real estate to commercial multifamily, where I was struggling to analyze deals as quickly and thoroughly as I wanted.
Two converging things: I was learning what AI could actually do, and I was personally bottlenecked by the same workflow every active syndicator has. So I decided to do something about it.
The workflow that breaks everyone
It's not one moment. It's a thousand small ones.
A broker drops a packet in your inbox. You hope it's in a format your model can read. It isn't. So you spend 30 minutes to 4 hours formatting it, depending on how rough the package is. Then you copy values into your model. Then you match T-12 figures. Then you align the rent roll. Then — finally — you have a basic analysis you can work with.
But you're not done. Now you need market analysis. Sensitivity. A real defensible buy decision. That's another half day. Sometimes a full one.
The PDF-to-Excel grind is what did it for me. Rent rolls coming in with every layout under the sun, and me manually aligning them so my Excel model could read them. That's when I said: enough of this.
3 AM is not a strategy
I'm the type of person that won't stop. That's just how I'm built. So I was constantly up until 3 or 4 in the morning underwriting deals. I had to, to compete.
But then I asked myself the real question: what if I took that same energy and worked smarter on something that would dramatically cut the late nights? And what if I could help other operators do the same?
That's how the concept came about.
Why existing tools didn't solve it
CoStar is a great tool. It's genuinely amazing. It's also ridiculously expensive, and it doesn't do what I needed it to do — like show the investor what their journey looks like over the hold.
Excel templates? You've seen the math above. They're flexible and they're mine, but they cost me hours per deal.
There was nothing on the market that did what I needed in one product at a price an active syndicator running $5M–$200M could actually justify.
So I built it.
Why the name “KeptDo”
KeptDo is built on my motto and how I live my life: we keep our word, and we do what we say we're going to do.
That's where Said. Done.comes from. It's the operating principle of the company and it's the standard the product has to hold itself to. Every analysis, every LOI, every LP deliverable.
The first real test
I turned KeptDo on Level 7 Capital's own pipeline. Midwest, 20–100 units, Class B assets in Class A or B areas. Sixty deals — from broker networks, on-market pipeline, off-market relationships.
We analyzed all sixty in under two weeks. 1 to 2 hours per deal in KeptDo, including sensitivity and best-base-worst scenarios. The old workflow would have been roughly 3 hours per deal on a good day — 180 hours across the screen, or $36,000 in analyst time at my rate.
Three deals cleared our return thresholds (4.5–7% cash-on-cash, 10–12% IRR, 1.85–2.0× equity multiple over 5 years). We submitted LOIs on all three. Made best-and-final on all three. Got outbid by other groups on all three.
That's fine. We're conservative because that's how we protect investor capital and deliver the returns we promise.
What comes next
The Multifamily Deal Analyzer Pro is the first product. It's the one that solved my own problem. There are more coming.
KeptDo isn't a software company in my head. It's a solutions company — we build AI tools for real problems that real operators, investors, and underwriters actually have. If you've got one you'd like us to look at, tell me.