Identifying private debt investment targets: an origination guide

The private debt market in the DACH region has been growing for years. More funds, more capital, more competition for the same companies.

What stands out is how unevenly attention is distributed. Ticket size ranges, return expectations and investment theses are defined down to the last detail and spelled out in investor presentations. Yet at many firms, a more basic question remains surprisingly unanswered: how do you find the right companies in the DACH Mittelstand – the region’s mid-sized, largely owner-managed businesses – in the first place?

This article shows why conventional research channels systematically fail in the sponsorless segment, which four mistakes render longlists useless, and what a robust sourcing process looks like in practice.

Why origination works differently in private debt

In traditional private equity, a substantial share of deal flow comes through established channels: M&A advisors, investment banks, structured auction processes. The debt fund is often only brought in once the sponsor wants to structure the financing. By then, someone else has done the actual origination.

For direct lending funds and mezzanine lenders, the reality looks different once they are no longer willing to wait solely for sponsor-driven transactions.

Sponsorless origination – directly approaching owner-managed Mittelstand companies that have no private equity backing – is one of the fastest-growing segments of European private debt. It is also the most labour-intensive. These companies do not put their financing needs out to tender. They do not appear in any deal database. They only become visible once a process is already under way, and by then others already have a seat at the table.

The structural reason for this imbalance lies in the ownership structure of the DACH Mittelstand. In Germany, Austria and Switzerland, a significantly higher proportion of the companies relevant to lenders are owner-managed than in the UK or France. There is simply no comprehensive sponsor ecosystem that prepares deals and brings them to lenders.

If you want these transactions, you have to go looking for them yourself.

The access problem in sponsorless transactions

Consider a typical scenario. An analyst at a direct lending fund has a clear thesis: manufacturing companies in Germany, €15–80m in revenue, an EBITDA margin above 12%, family-run, with a generational change as a possible trigger.

So where do they start?

Option 1: traditional company databases. Conventional credit reference agencies and address data providers supply raw data in bulk, but no qualification. You get thousands of hits with no context. Which companies are actually profitable? Which ones are facing a handover? Who is the decision-maker, and can they be reached? The database does not answer these questions; it merely pushes them further down the line.

Option 2: networks and intermediaries. Tax advisors, auditors, law firms and M&A advisors know their clients. But this network does not scale, is highly selective, and the same tip-off usually goes to several market participants at once. It creates no proprietary advantage.

Option 3: manual research. LinkedIn, company websites, commercial register (Handelsregister) extracts, press articles. Time-consuming, unsystematic – and it inevitably misses precisely those hidden gems that do not maintain much of an online presence. A machinery manufacturer with €45m in revenue, three patents and a twelve-year-old website is practically invisible this way.

The result is the same at almost every firm: a disproportionate share of analyst time goes into data collection rather than conversations.

Four mistakes that make longlists useless

Mistake 1: sector without a financial profile

Many sourcing approaches start with a sector definition and stop there. But a company in the right sector with too thin a margin or too little revenue is not an investment target.

The problem gets worse when the sector definition relies on official classifications such as WZ (the German industry classification) or NACE. These codes were developed for statistics, not for deal sourcing. A specialist in test benches for the automotive industry and a manufacturer of garden equipment can sit under the same code. Filtering this way produces noise while missing niches that no code list maps cleanly.

Mistake 2: lack of regional granularity

“The German Mittelstand” is not a targeting strategy. If you want to source specifically in industrial clusters – say, mechanical engineering in Baden-Württemberg, medical technology in Bavaria or logistics in North Rhine-Westphalia – you need geographical precision down to postcode level, combined with the financial data. Having the two separately does not help.

Mistake 3: outdated decision-maker data

Identifying a target is one thing. Reaching the right contact person is another. Managing directors change, shareholder structures shift, direct-dial numbers are reassigned.

The damage done by outdated contact data goes beyond the time wasted. A letter addressed to a managing director who left two years ago signals to the company the exact opposite of what you want to convey in a first contact.

Mistake 4: no signal of a financing trigger

Not every qualified company needs capital right now. Good origination means spotting the trigger early: an upcoming growth investment, a change of shareholders, a maturing bank facility, completed acquisitions that call for follow-on financing.

We look in detail at which signals can actually be read from publicly available data, and how reliable each of them is, in our article on the refinancing wave in the DACH market.

What makes a good target

Before you start searching, you need to be clear about what you are looking for.

Company structure

• Owner-managed or family-controlled, with no private equity sponsor
• Clear operational responsibilities and professionalised management
• Stable core business in a defined niche

Financial profile

• Stable or growing EBITDA over at least three years
• Recurring revenue from maintenance, service or long-term framework agreements
• Manageable ongoing capital expenditure, so that free cash flow after capex covers debt service
• Net financial debt to EBITDA within a range that leaves headroom for refinancing

Situational trigger

• Succession or generational change
• Growth or internationalisation financing
• Add-on acquisition, even without a sponsor
• Replacement of existing bank lines or first-time alternative financing

Regional specifics in the DACH region

The DACH market is less transparent than the UK and US markets. That increases the research effort while at the same time reducing competitive pressure. For firms that systematise the effort, this lack of transparency is not an obstacle but the very reason the segment works.

How ProxDeal changes the sourcing process

ProxDeal is an AI-powered origination platform for the DACH Mittelstand. Its database covers more than 7 million DACH companies and over 992 million data signals: 592 million for Germany, 180 million for Austria and 220 million for Switzerland.

For private debt investors and debt advisory boutiques, this means four things.

Natural-language search instead of filter logic. You enter the target profile as free text. The Hyper-RAG AI translates it into a search across the entire dataset. A thesis such as “special-purpose machinery for the food industry, southern Germany, €20–60m in revenue, owner-managed” works straight away, without first having to be translated into industry codes. Codes can be used as an additional filter, but they are not the starting point. This is precisely where list providers fall short: their search begins and ends with the code list.

Combinable financial, structural and geographical filters. Revenue, earnings, equity, leverage, shareholder structure, holding structures, year of incorporation, federal state, postcode radius. Freely combinable, in a single pass.

Direct contacts at management level. No switchboards, no info@ addresses. For initial outreach in sponsorless direct lending, this is where response rates are won or lost.

A longlist with more than 100 parameters per company, exportable to Excel. More than a hundred AI analysts working in parallel evaluate public registers, annual financial statements, company websites and media reports, and match each company against the defined profile. The result is a prioritised list, not a pile of raw data.

Practical example: from thesis to longlist

Starting point. A mid-sized direct lending fund with a DACH focus is looking for borrowers for senior secured loans. Investment thesis: manufacturing, €20–80m in revenue, an EBITDA margin above 11%, family-run, southern Germany and Austria.

Without structured data. The analyst spends two to three days working through database extracts, search engines and LinkedIn, and asking around their network. The end result is a list of around 300 entries. After initial qualification, perhaps 50 of them remain relevant. Time went into 250 entries that yielded nothing – and the contact details are still missing.

With ProxDeal. The profile is described in free text and supplemented with the financial filters (revenue of €20–80m, EBITDA margin above 11%), the geographical filter (Bavaria, Baden-Württemberg, Austria) and the structural filter (owner-managed, no identifiable sponsor).

Result. Between 80 and 150 qualified companies with financial metrics, shareholder structure, managing director contacts and identifiable signs of a financing need. Processing time: 5 minutes.

The difference lies less in the time saved than in what that time can then be used for. A research exercise spanning several days becomes a task you finish before your first coffee. The analyst starts outreach the same morning, not on Thursday of the following week.

For debt advisory boutiques

Debt advisory boutiques face a twofold problem. They need to know the right lenders, which comes with time, and they need to find the right borrowers before those borrowers become active themselves. The second part is the harder one.

In debt advisory, mandate origination is, at its core, a deal origination problem.

If you approach a Mittelstand company before it has communicated its financing needs, you are having an advisory conversation. If you only react once the mandate has been put out to tender, you end up pitching against eight other firms and negotiating over fees.

In concrete terms, this means the following for advisory teams:

• Proactive identification of companies with a recognisable trigger, such as growth, a change of shareholders or expiring financing
• Sector-specific screening in precisely the industries in which the boutique has a track record
• Decision-maker contacts for outreach by letter, email or phone
• Repeated screening of the same thesis over time to detect changes in structure and key metrics

Our guide to direct lending in the Mittelstand describes which instruments are used in the subsequent structuring and which terms are realistic in the current market environment.

Conclusion

The private debt market in the DACH region is becoming more competitive. More funds, higher allocations, falling risk premiums in the established segment. If you continue to rely solely on sponsor deal flow and chance introductions through your network, you see the same slice of the market as everyone else – and compete accordingly.

The sponsorless segment is the alternative, but it does not come for free. It requires a process that you can repeat, measure and improve.

The question is no longer whether structured company data is necessary for deal sourcing. The question is who will be the first to use it systematically.

Create a longlist and identify your first qualified investment targets in 5 minutes.

Frequently asked questions

ProxDeal is the AI-powered origination platform for M&A advisors, private equity, private debt and search funds in the DACH region. More than 7 million DACH companies, 992 million data signals, longlists in 5 minutes. Made in Munich, hosted in Frankfurt, GDPR- and EU AI Act-compliant.

Last updated: 8 August 2026.