Quick answer

A new AI operating partner needs a fast, defensible, measurable win on a line the investment committee can see — which points to revenue, not diffuse efficiency. The strongest first 90 days: spend the first month finding the revenue already recoverable in one or two portfolio companies' existing data (dormant referral and partner relationships), the second month reactivating it to produce a real P&L result, and the third standardising the method so it can roll out across the portfolio. It lands quickly because it runs on data the companies already have.

Key takeaways
If you only read 30 seconds of this article.
  1. Lead with revenue, not efficiency — it's the line the IC can see and you can defend.
  2. Start where data already exists — recoverable revenue needs no new stack or long build.
  3. Prove it on one or two companies before scaling — a real result beats a broad pilot.
  4. End the quarter with a method, not a one-off — repeatability is the real deliverable.

The first-90-days mandate

The job of a new AI operating partner is, bluntly, to make AI mean something on the financials before the goodwill runs out. The temptation is breadth — launch pilots across many companies and many use cases to look active. Breadth is exactly the trap. It produces motion without a provable result, and 90 days later there is activity to report but no number. The mandate that actually protects credibility is narrow and deep: one defensible, measurable win on a line the investment committee already watches. That line is revenue, and the fastest revenue win is the one hiding in data the portfolio already owns.

Days 0–30: find the revenue

The first month is diagnosis, not deployment. Pick one or two portfolio companies with meaningful referral or partner-driven revenue and run the recovery diagnostic on their existing history — no new software, no IT project. The output is a located, quantified opportunity: which relationships have gone dormant, how much is recoverable, and where concentration risk sits. Crucially, this can be done in days because the data already exists, so you end month one with a number you can take to the company's leadership rather than a roadmap of things you intend to try.

30d
To a located, quantified revenue opportunity
34%
Median dormancy the diagnostic surfaces
P&L
Where the result lands — a line the IC watches
Benchmark figures from 1,000+ direct company audits.
Source: Innovation Park Revenue Lens Benchmark · Q2 2026.
Day one of the 90:

Get a recoverable-revenue number for a portfolio company now.

The free Revenue Recovery tool gives a benchmark-calibrated range in 90 seconds. No email required.

Open the free tool

Days 30–60: prove it

The second month turns the number into a result. Working inside the company's existing CRM and revenue-cycle stack, the reactivation sequence begins on the highest-value dormant relationships, and the first recovered revenue starts to appear. The goal of this window is not to finish the recovery — that continues well beyond 90 days — but to demonstrate the cause-and-effect clearly enough that no one disputes it: these named relationships were dormant, we reactivated them, this revenue returned. A small, undeniable, attributable result in month two is worth more to your standing than a large projected one, because it converts the diagnosis into proof.

Days 60–90: make it repeatable

The final month is where a good operating partner separates from a lucky one: turn the win into a playbook. Document the method, standardise the reporting, and sequence the rollout across the rest of the portfolio prioritised by recoverable-revenue potential. The deliverable that earns the next mandate is not "we recovered revenue at one company" but "we have a repeatable, P&L-visible AI value-creation method and here is the portfolio-wide plan." That reframes you from someone running a pilot to someone operating a program — which is the whole point of the role. On qualifying $30M+ engagements the recovery work carries our 3× fee recovery guarantee: we recover at least three times our fee, or we keep working at no additional fee until we do.

FAQ.

What should an AI operating partner prioritise first?

A fast, defensible, measurable win on a line the IC can see — which points to revenue over diffuse efficiency. The highest-probability first move is finding the revenue already recoverable in one or two portfolio companies' existing data, because it lands on the P&L, shows results quickly, and establishes a repeatable method rather than a one-off pilot.

Why not start with efficiency pilots?

They're easy to launch and hard to prove. Gains are diffuse and hard to attribute to a P&L line, so even a success struggles to demonstrate value in 90 days, and they often need lengthy customisation. A revenue win that's visible and quick serves a new operating partner better than an efficiency project that may not pay off until after credibility is needed.

How fast can revenue recovery show a result?

A first read can come in days because it runs on existing data, and reactivation typically begins producing within the first engagement. That speed is exactly what a first-90-days mandate needs: a credible, P&L-visible result inside the window, plus a method to roll out across the portfolio.

JV
Julia Vorontsova & Tyler Opsahl
CEO & COO · Innovation Park · Antwerp & Denver

Julia and Tyler built Revenue Lens from 1,000+ direct company audits and deploy it as a portfolio-wide value-creation lever for private-equity owners. Articles are drafted with a bench of industry writers, partner-network operators, and AI specialists experienced in regulated industries such as healthcare and finance.

How to start.

Here is the fastest path to a real answer. No leap. A stair.

What we won't ask for in the first conversation: portfolio-company data, an engagement letter, your team's time, or a signed NDA before we've confirmed fit.