Commercial due diligence with simulated customers

Before exclusivity, the seller controls access to the target’s relationships. Sellers worry about buyer contact with customers, suppliers and distributors for competitive reasons and to manage important customer relationships. They “may insist that access to these parties be delayed until late in the diligence process (if at all)”, Kate Sherburne wrote in a March 2018 note now hosted by Faegre Drinker. KPMG in Ireland adds that much of commercial due diligence concerns the context outside the target, so it can be researched without drawing on management’s time, through “discreet primary research and triangulation of secondary sources” (Chris Brown, July 2022).

Exclusivity, when it comes, is measured in weeks and crowded with other diligence. Reviewing the European market, Travers Smith’s Private Equity Market Insights 2026 reports that the average initial exclusivity period on private equity transactions rose from 35 days in 2024 to 49 days in 2025, with many periods then extended as cautious buyers undertook extensive due diligence, particularly on financials.

The private equity firm in Artificial Societies’ case study was at the pre-exclusivity stage, where it would typically test its thesis in-house with an expert network or panel provider, or commission a tightly scoped exercise from a commercial due diligence advisor. It wanted the research to go further, and some of its questions were commercially sensitive: putting potential ownership changes, product consolidation or future pricing directly to customers could expose the investment thesis and unsettle the market.

Artificial Societies’ client was diligencing a specialist software group with a portfolio of established brands. With a panel partner, we fielded the client’s survey to 130 software decision-makers, covering product usage, satisfaction, renewal intentions, switching barriers and appetite for new capabilities. We then built 130 digital twins, each anchored in one participant’s responses, and put follow-up scenarios to them. The table keeps observed and simulated results apart, as the study does.

Source: Artificial Societies, commercial due diligence case study, engagement undated.

The survey established a strong retention baseline, and the simulation added a more conditional picture. Artificial Societies reported the 42% who would review alternatives or leave on product retirement as a risk for further diligence, because changes to the product portfolio could prompt customers to reconsider systems they would not otherwise consider. The pricing result narrowed the value creation case in the same way: 52% of personas accepting no price increase to move to a cloud-based product suggested that migration alone would not establish a compelling case for higher pricing.

In Artificial Societies’ private equity engagement, the simulated society started from real respondents: 130 twins forming one society, Business Software Decision-Makers, made up of business owners, operational leaders and other decision-makers who select and use specialist software across a range of industries. We enriched each persona beyond the questionnaire with deep market research, proprietary industry intelligence and client-supplied expert interview transcripts covering procurement, implementation and operational workflows. The study kept observed responses distinct from that inferred context. An investment committee needs that line drawn: the 98% renewal figure is what survey respondents said, and the 42% figure is how their twins answered a product retirement scenario.

The client’s critical questions set the agenda:

• •

  How durable is customer loyalty, and what could trigger switching?

• •

  How would buyers respond to changes in pricing, ownership or product strategy?

• •

  Which new capabilities could support additional spending?

• •

  What value creation initiatives could drive growth without putting retention at risk?

Simulated follow-ups examined switching behaviour, platform migration, product consolidation, contract terms and appetite for AI-enabled capabilities, without disclosing those potential decisions to live customers or competitors. Results could be explored across the society and persona by persona, combining distributions with the reasoning behind each response. The answers remain inferred, so they tell a deal team where to direct confirmatory work, not what real buyers will do once a decision is announced.

Artificial Societies built the private equity firm’s society as a reusable research asset that could carry forward into deeper diligence and, if the investment proceeded, value creation. The client retained access to refine hypotheses, explore product priorities and inform potential post-acquisition plans, and could test new scenarios as the thesis developed without commissioning a fresh panel for every follow-up question.

Bain & Company’s Global Private Equity Report 2026 finds that typical deals now require around 10% to 12% average annual EBITDA growth to reach the benchmark 2.5x return over five years, and urges “full potential due diligence” on revenue, operational and technology levers (press release, 23 February 2026). Pricing and contract length are revenue levers, and the society had already put conditions on longer contracts: in Artificial Societies’ engagement, 48% of personas would take a three-year agreement with a 10% discount only with a price freeze and an exit clause tied to service levels. Because each twin is anchored in answers given when the survey ran, a team returning to the society a year later should weigh that date.

The twins in Artificial Societies’ engagement rest on 130 real survey responses plus modelled context, so the simulation adds a conditional picture on top of that primary research and cannot be broader than its base. With 130 respondents, segment cuts are small: split five ways, the sample averages 26 respondents per segment, which we would treat as directional rather than as a finding for an investment committee paper.

The scenario results sit two steps from behaviour. The 52% of personas who would accept no price increase for a cloud migration, and the 48% who set conditions on a three-year deal, are twins’ answers: one step from twin to person, a second from stated intention to signed contract. For that reason we would put the society’s distribution in an investment paper, not a single twin’s answer.

A first engagement with Artificial Societies follows the shape of the private equity case:

1. 1.

   The deal team sets its critical questions and the questionnaire; in the case study, the survey was the client’s own.

2. 2.

   We field the survey with a panel partner to set the baseline.

3. 3.

   We build one digital twin per respondent, enriched with market research, industry intelligence and any expert interview transcripts the client supplies.

4. 4.

   The deal team puts its sensitive scenarios to the society and reads the results in aggregate and persona by persona.

5. 5.

   The client keeps access to the society in Radiant, Artificial Societies’ platform, to re-run analyses through confirmatory diligence and value creation.

To scope a deal, book a call with the team.