Model ML at Work: Signals

Every investment bank maintains a pipeline tracker. This is often a spreadsheet covering hundreds of companies, updated by hand from data sources, news articles and analyst calls. Keeping it complete and current takes constant effort.

When a real opportunity surfaces, there's still plenty to do: running the valuation, understanding the timing and competitive dynamics, building the pitch deck and identifying the right buyers or sponsors. That can mean days, sometimes weeks, of work before anything is in front of a client.

Most of the conversation around AI in banking is about productivity: getting to version one faster. At Model ML, we think that's just the beginning. Signals is a new way to monitor, model and pitch automatically, making AI revenue-generating by compressing the time from idea to model to materials to outreach, so you get to a mandate faster.

How Signals Work

With Signals, the static spreadsheet becomes a Grid that tracks the same quantitative and qualitative signals your team already watches, such as stock price moves, swings in profitability or management changes. Model ML plugs into your internal and external data sources, including S&P Capital IQ, PitchBook and news, so your target universe updates itself.

You set the criteria. When the right signals trigger, your agent takes it from there.

That means:

• Coverage scales from hundreds of companies to thousands, without adding headcount to maintain the tracker

• The data stays current automatically, rather than depending on someone remembering to update it

• When a company hits your criteria, the agent runs the valuation work, whether that's an LBO, a DCF or an analysis at various prices

• If the numbers work, it builds the materials and matches the opportunity to the sponsors or clients you have relationships with

Each step triggers the next, so what starts as a signal ends as a pitch in your inbox, ready for your judgment.

Use Cases: Signals In Practice

Example 1: A company on the watchlist changes CEO. The tracker has already caught it.

In this video, Jared, former Managing Director at Evercore, walks through the pipeline tracker every bank runs. Normally, a management change at one of hundreds of tracked companies only gets picked up if someone spots it in the news and updates the spreadsheet. With Model ML, the team's target universe lives in a Grid covering thousands of companies, connected to S&P Capital IQ, PitchBook, news and the bank's internal data. Each company is tracked on the signals the team cares about, from stock price fluctuations and profitability swings to management changes, so when the CEO change comes through, the company is flagged against the team's criteria straight away. This results in a target universe that stays complete and current, with the opportunity flagged the day it appears rather than whenever the spreadsheet is next updated.

Example 2: A public company screens as a take-private candidate. The LBO and deck land in your inbox.

When a company's Signals trigger against the team's criteria, the agent moves straight into the analysis, running the LBO through to the resulting IRR and MOIC. If the returns are in the money, that triggers the next step: a take-private deck built from the model, including a company overview, the key investment thesis and an analysis at various prices. The agent then cross-references the opportunity against the sponsors the bank has relationships with, so the right idea goes to the right client. It can go further still, into mispricing and public market disconnect through a sum-of-the-parts, transaction structure with the resulting returns, and exit strategy. Once it's done, the banker receives an email with the LBO and the take-private deck attached. This produces a sponsor-ready pitch, matched to a client, before anyone had to spot the opportunity by hand.

Example 3: A company's debt starts to slip. The capital structure alternatives deck is already underway.

Not every opportunity is a take-private. For another company in the universe, the signals that matter might sit in its capital structure. The Grid tracks debt trading levels and upcoming maturities alongside everything else, and when the debt trades down to 95 cents on the dollar with a maturity cliff less than two and a half years out, the agent puts a different workflow into action and builds a capital structure alternatives deck. This creates a timely pitch for a client that's about to need advice, prepared as soon as the conditions appear rather than weeks later.

With Signals, the time between having an idea and landing a revenue-generating mandate gets dramatically shorter. Signals is coming soon to Model ML.

Want to get started? Book a demo.