The Hypergrowth Playbook - Vol. 1: Legal & Audit

Most hypergrowth advice assumes a specific kind of market: one where you can ship fast, apologize later, and let growth outrun your mistakes. That playbook doesn't transfer to legal or audit. Both professions are built around catching errors, not tolerating them, a lawyer's job is to find the exception, an auditor's job is to find the discrepancy. Selling AI into either means your buyer's entire professional training is oriented against trusting you quickly.

Vera Wienken (Director of Marketing, Libra, a legal AI workspace, part of Wolters Kluwer) and Irina Botea (VP Marketing, Cortea, an AI-native platform for auditors) have both had to hypergrow inside that constraint. Libra scaled from a 3-person marketing team to a full local presence in 12 countries in under a year. Cortea is earlier in its journey, but on a similarly steep curve: after landing its first audit-firm customers, it's now co-developing close to ten AI agents with them in parallel, and is tracking toward tripling the business by the end of the year.

Neither got there by moving fast and fixing things afterward. What follows is what they actually did instead, pulled directly from a live conversation during the Women's AI Breakfast:

1. Make trust your growth channel, not a compliance checkbox

The instinct in most AI go-to-market is to treat trust and credibility as something you build around the product: testimonials, case studies, a compliance page. Both Libra and Cortea treat it as the product itself.

Libra's clearest example: when a competitor launched a marketing campaign fronted by a famous Hollywood actor, Libra's response wasn't a bigger celebrity. It was the opposite move: real customers, from multiple countries, talking about the product on camera. Every image in Libra's marketing uses actual lawyers, never stock photography, never an AI-generated face, never a paid brand representative. Wienken's framing: "With venture capital money you can purchase anything. That doesn't necessarily build trust."

Cortea's version is more structural. Auditors are professionally trained skeptics, and they themselves get audited on the judgment calls they make. That means an auditor can't tell their own reviewer "the AI decided this." So Cortea built the accountability directly into the interface: the tool shows the full reasoning trail behind every AI output, so the auditor can explain, and defend, the decision as their own. Botea calls this "opening the black box."

The play: if your buyer has their own accountability chain above them, your trust-building has to plug into that chain, not run parallel to it. A testimonial convinces a buyer. An audit trail survives the buyer's own review process. Know which one your category actually requires.

‍

‍

2. Design your growth curve around land-and-expand

Both, Botea and Wienken, describe the same shape: slow to land, fast to expand, with almost no middle gear. Cortea's version: it takes real time to earn a single audit firm's trust on one AI agent. Once that trust exists, expansion isn't even a sales conversation anymore: Botea describes firms proactively asking to co-build the next agent, embedding their own methodology into it, no pitch required.

This has a direct implication for how you should think about your own growth math in a trust-gated category: your "time to first deal" and your "time to expansion revenue" aren't the same curve you'd model somewhere trust is assumed rather than earned. Front-load your patience into the first deal. Expect the payoff on the back end, not the front.

3. Decide what not to scale before you decide what to

Both speakers draw an identical line, and it's arguably the sharpest operating principle in the whole conversation.

Cortea's version is a diagnostic question Botea says she asks herself constantly when growth feels slow: "Is this a capacity problem, or a clarity problem?" Her explicit warning: don't try to fix a clarity problem by throwing headcount at it. In her words, "I learned this the hard way." In practice, this means resisting pressure to scale a team before testing whether the existing team, restructured, can already do the job. It also means resisting channel sprawl and adding more channels when you "haven't even figured out LinkedIn yet".

Libra's version shows up operationally. Wienken describes deliberately not scaling internal process at the same pace as headcount or country count, not from neglect, but because "if you implement and scale processes too early, it can slow you down as a company." When Libra built a new AI-education platform for customers, they tested it in Germany (their home market), watched adoption and feedback, and only then rolled it out elsewhere, rather than launching in all 12 countries simultaneously.

The play: in a hypergrowth narrative, restraint reads as caution. In practice, for both companies, restraint was the actual growth mechanism: it's what kept speed from outrunning the thing (process, channel focus, local credibility) that speed depends on.

4. Know which roles need potential, and which need proof

Both speakers landed on the exact same hiring rule: for most early-stage roles, hire for curiosity and adaptability. For roles like product marketing and positioning, hire for direct experience.

Their reasoning matched too. Right now, both of them are personally the ones translating high-level positioning into everyday messaging and content, and both are wary of staying the only person who can do that as the company keeps growing. Curiosity can be taught into most roles over time; the specific skill of turning positioning into content is harder to grow from scratch, so that's where they hire for experience instead of potential.

Two concrete filtering techniques worth stealing directly:

Wienken's interview question for gauging real curiosity: "What's a marketing or brand campaign that genuinely inspires you?" It sounds like an easy question to answer, which is exactly why it works so well as a filter: a surprising number of candidates can't actually answer it.

Botea's method: an on-site trial day with a broad, open-ended assignment. Since candidates are physically in the building with the people who actually built the product, curiosity shows up in whether they walk over and ask that team directly, or default to guessing (or asking an AI) instead of using the people right in front of them.

‍

‍

5. Get acquired without losing your speed

Getting acquired by a giant has a reputation for slowing everything down: the speed, the autonomy, the parts that made the startup worth building in the first place. Libra's experience argues the opposite, but only because they negotiated the right conditions upfront.

When Wolters Kluwer (190 years old, with roughly 3,000 people in Libra's division alone) acquired the company, Libra's ask wasn't just resources. It was protection. Wienken describes it plainly: "We don't want people to come and do safaris at your office." That protection is what let a 15-person team keep operating like the speedboat of the bigger corporation, instead of getting slowly absorbed into a bigger, slower structure.

The acquisition also handed Libra real acceleration it couldn't have built alone: Wolters Kluwer already had local offices in all 12 countries Libra needed to launch in, and its legal content library gave Libra's AI outputs the credibility lawyers needed to actually trust them.

None of this was friction-free. Wienken is direct about it: "there are also clashes." Bringing along hundreds of existing salespeople and local marketing teams, all suddenly representing a new product, took as much deliberate work internally as winning any external customer did.

The play: if you're negotiating an acquisition, negotiate for insulation before you negotiate for resources. The resources (infrastructure, credibility, market access) are the upside everyone already expects. The room to keep operating at startup speed instead of being absorbed immediately, is what determines whether you actually get to use that upside at all.

6. Turn a mistake into proof, not a liability

Both speakers treat mistakes as inevitable rather than as failures to hide, which matters because their buyers are professionally trained to hunt for exactly this kind of error, a wrong clause, a wrong number. In that context, a mishandled mistake doesn't just annoy a customer, it confirms the buyer's original skepticism about trusting AI with judgment calls at all.

Botea's clearest example: a skeptical seven-person pilot group at an audit firm experienced a real product mistake mid-pilot. Cortea's response was to get on a call, explain plainly what happened, and ship a fix within hours, with no blame shifted onto "the model updated overnight" as an excuse, just an acknowledgment and a fix. When the pilot group later voted on which vendor to select, all seven chose Cortea anyway.

Wienken frames this as a cultural value at Libra, not just a crisis-response protocol: talking openly internally about mistakes, "the fuckups," rather than treating them as things to manage around. Her reasoning: a culture that punishes visible mistakes produces people who hide risk instead of catching it early.

The underlying thesis

Hypergrowth in a trust-gated vertical isn't a slower version of normal hypergrowth, it's a different optimization problem entirely. In a category where the buyer doesn't need much convincing, the constraint on growth is usually distribution: can you reach enough people fast enough. In legal and audit, the constraint is trust velocity: how fast you can earn the right to be believed, one accountable decision at a time. Every principle above is really the same idea applied to a different function: match your growth mechanism to what your buyer's own scrutiny actually requires, and treat restraint as a design choice, not a delay.

‍

This piece draws on a live conversation between Vera Wienken (Libra), Irina Botea (Cortea) and Nicole Büttner (Merantix Capital) at the Women's AI Breakfast, hosted at Merantix AI Campus.