Description

Alpinist wrote up Adevinta in August 2020. After three years and with the stock down 50%, this situation is worth revisiting. Two apologies at the start here btw: this ended up rather long. I guess this reflects a somewhat complex group and history. Also I started writing this up in July when the share price was in the 60s, but didn’t get round to finishing it. Sorry. Perhaps I should have gone with brevity over attempting more quality. Anyway, thoughts very welcome.

It's fair to say the Ebay Classifieds Group deal has disappointed against expectations. When the deal closed in mid-2021, estimates for 2023 EBITDA were EUR 850m. Today they sit at just over EUR 650m. However, much of the repricing of the stock is simply the unwinding of a bubble-era valuation.

I believe Adevinta owns some very strong assets, has many years of double-digit top line growth ahead of it, is at the beginning of a sustained improvement in margins and cash flow performance, and is available for better than 8% FCF yield on fairly conservative 2026 numbers. And under far better operational management than it has had previously. I think this opportunity has arisen because investors continue to be sceptical on management despite increasing positive evidence, and more so because many are obsessed with the overhang created by the current ownership structure and how that will change. If you can take a slightly longer perspective and look through that, I think this stock could well double over 3 years.

The overhang

Permira bought some of Ebay's stock in Adevinta in H2 2021 (at roughly double today's share price), after which Adevinta has three big shareholders: Ebay (33%), Schibsted (just under 30%), and Permira (11%). Alongside the Permira announcement, all three agreed to a lock-up of most of their holdings until October 2023. Schibsted has made a couple of small disposals in the interim. Ebay and Schibsted have both made clear they are financial investors, not indefinite holders. So the market is anticipating selling pressure once the lock-up ends, or at least the ability to buy Adevinta in size via blocks, at a discount to market.

The incentives are rather different for the three players - Permira is rumoured to have funded half of its purchase with debt, in which case its stake has roughly zero equity value. Unless forced, Permira is highly unlikely to be a seller at close to the current share price. For Ebay, with a market cap of USD 23.5bn, the Adevinta stake (market value USD 2.8bn) is material, but still minor. For Schibsted, with market cap USD 4.3bn and Adevinta stake worth just under c. USD 2.5bn, it dominates the group. Schibsted's stub trades at a substantial discount to its likely standalone value, so Schibsted management is unsurprisingly under pressure from shareholders to resolve the situation and unlock the discount. It has promised to do so. For instance:

CFO in Mar 2023 at Schibsted CMD: "For Adevinta, the relatively low free float of the share due to the combined size of the major shareholdings and also anticipated sell-downs to the market is not optimal either to unlock the full underlying value of the Adevinta stock. So we are with this backdrop, working with strategies to reduce our ownership stake in Adevinta in a value-creative way for Schibsted shareholders. But given the size of our stake, we are prepared to use the time needed to do so."

CEO in July 2023 at Schibsted Q2 results: "we have, together with our Board of Directors, used the second quarter to further explore and develop our options, as outlined at our Capital Markets Day in March, to reduce our ownership in Adevinta in a value-creating way for our shareholders. Our ambition is still to come back with an update on a preferred solution at our Q3 results presentation in October at the latest."

So we can expect an answer on Schibsted's plans in the next 6 weeks or so. Options include a sale of the whole group (e.g. to Prosus, most obviously), a sale of Schibsted's entire stake, a distribution of some or all of the stake to shareholders, and a gradual disposal via blocks, or in the market. The cleanest solution that is within Schibsted’s control is of course a spin of the stake, but there are questions over Norwegian witholding tax implications. I understand Schibsted is awaiting a ruling from the tax authorities over the next couple of weeks.

Speaking to other shareholders, I believe management has received a pretty consistent message that a spin is the best approach. However, the Tinius Trust (27% owner of Schibsted) probably has to approve the decision, and we have little knowledge of the trust's position.

There's a decent chance of a positive resolution in a few weeks that prompts the market to start re-evaluating Adevinta. Regardless, looking out a couple of years, it's overwhelmingly likely the situation will have been resolved, and the overhang will not exist. So if you are able to have any patience, I think this is a major opportunity to exploit investors' myopia. If you’re a genuinely long term holder – then in the worst case, where Schibsted and/or others gradually sell down in blocks, and the share price is depressed for a period – Adevinta should be able to capitalise by buying back at low prices given the sharp reduction in leverage during 2023.

Very quick recap on the business model

Adevinta is a pureplay online classifieds business. I'm sure there are plenty of write-ups of these businesses on VIC so I won't bore you with the encyclopaedic version here. But it is an archetypal network effect business - buyers go where sellers are, sellers go where buyers are. There are nuances with different product categories, specifics of business models, and particularities of individual verticals in individual countries. But broadly, online classifieds are like newspapers of old except without the requirement to surround the classified ads with costly content or any physical product to print or distribute.

Modern online classifieds also provide a proposition that is far more valuable than newspapers were. For instance, consumers get access to much more comprehensive and detailed listings which means fewer wasted trips to view an unsuitable car or house. For sellers, leads can be qualified much more thoroughly reducing time wasted talking to inappropriate or early stage searchers, and classifieds companies can use the data they have about the products and customers to package products such as a pricing tool enabling car dealers to price inventory more accurately for quicker sale without lost revenue or enabling car dealers to advertise their inventory in a precisely targeted fashion to consumers far from the dealership.

In brief, the value creation is much larger than old-school classified advertising models, yet the cost base is much lower - hence the best online classified businesses (Rightmove, Auto Trader for instance) achieve 70%+ margins.

Adevinta doesn't achieve such attractive margins, not by a long stretch. At least, not yet.

How we've got to where we are today

Again, I'll keep this fairly brief, but it's worth looking back at the history of the group to understand the context for where Adevinta is at today, and what it needs to do to deliver on its potential.

• Adevinta comprises mostly horizontal platforms, with the exception of a couple of assets in Spain, and Mobile.de in Germany

  • Horizontal platforms built the general "flea-market" marketplace first, and then layered on the higher value verticals (property, motors, jobs)later. Therefore they were by nature initially challengers in these verticals.

  • This means that vertical market positions are typically not as dominant as some vertical specialist classified businesses which established dominance early on and have never ceded market share, such as the big UK ones, Hemnet in Sweden, REA Group and Carsales in Australia.

• The company has been strategically misguided in the past with very poor capital allocation

  • Under Rolf-Erik Ryssdal, Schibsted (pre-spin of Adevinta) did a lot of empire-building, and not a lot of fulfilling the potential of its very strong franchises - it ploughed capital and management resources into emerging markets and lower income European countries (mainly through M&A) with an over-optimistic perception of growth potential, whilst product development and innovation on its extremely under-monetised core Scandinavian and French assets was deeply lacking. I won't go through deal by deal, but there are many errors one could examine here!

    • In management's defence, what would have been an excellent deal (acquisition of Hemnet - current market cap c. SEK 19bn - for SEK 1.5bn, announced in 2015) was sadly canned by regulators

  • Schibsted dealt very poorly with the mobile transition, both technologically and strategically. Whilst the businesses were largely strong enough to weather this change without much damage, it was a very messy period for the company, and included ceding a large part of its leading horizontal marketplace position in Spain to start-up Wallapop, then throwing several hundred million Euros at copycat Shpock, which was eventually disposed for zero consideration in 2021.

  • Aside from the waste of cash, these strategic misadventures were funded partly through (undervalued) equity - the company twice diluted shareholders before the Adevinta spin, with 5% issuances in both 2015 and 2017

• Previous management was unambitious and lacked operating discipline

  • Before Adevinta spun off from Schibsted, management had a habit of describing its Norwegian business Finn (before the spin-off from Adevinta) in private meetings as "mature" in 2013-14 (since when it has grown revenue by [3x])

  • Schibsted did lots of acquisitions, but failed to integrate them effectively, particularly in terms of tech architecture

  • Even on organic expansions - several international expansions were pursued organically, used the core platform that the Swedish side blocket.se was built on. However, the code was then developed locally such that, in Galapogan fashion, the various sites evolved to be incompatible with one another and required entirely separate development even of identical capabilities

  • Other technological screw-ups

    • As noted above, the transition to mobile was managed very poorly

    • Recognising its large tech debt and unwieldy localised code-bases, Schibsted invested heavily in a centralised global tech platform (named "Rocket") which was then abandoned and written down after 3 years

  • Slow product development cadence

  • Poorly structured pricing and packaging; failure to incorporate the fairly obvious lessons from best practice examples such as Rightmove and Auto Trader

• Adevinta has a lower quality revenue mix than many classifieds peers, with a significant contribution from display advertising

  • Display advertising is an extremely high margin revenue stream, hence rather difficult to voluntarily say goodbye to. However, it has major problems:

    • It is structurally challenged given the superior proposition of the large platforms

    • It Impairs the user experience - adds are not part of the experience, but get in the way

• Ebay Classifieds a classic "cousin from the country" - under-invested, managed for cash whilst part of Ebay Group

To summarise, the three core mistakes of the past:

• Bad strategy, leading to bad capital allocation:

  • Under-estimating the value / potential of the dominant assets it owned in high income countries France, Norway, Sweden etc. (hence dilution at low valuations)

  • Over-estimating the value of the much less dominant assets it bought and built in lower income and emerging markets (particularly those in south America and Asia, but even Spain to some extent)

  • Failing to prepare for major shifts, resulting in panics to mitigate (such as acquisition of Shpock)

• Lack of operational focus and rigour, most obviously in failing to copy the business models and product development of better-managed market leaders in other countries

• Duplicative and inflexible tech architecture

CEO change: a watershed moment

The CEO was replaced in August 2022 and has sharply changed the tone. There is a lot of work to do, and no doubt one of the biggest jobs he has to do is to transform the culture - progress here isn't easy to monitor from the outside. But the change in investor communication has been stark - the message is clear and consistent, and emphasises the need to prioritise, to make tough decisions, to maximise the value of the assets the group owns, to dispose of irrelevant businesses rather than seeking to expand the portfolio. And having run the biggest business in the group (LeBonCoin) for many years, he understands the business model from an operational perspective much more viscerally than the previous CEO. Finally, he is overwhelmingly focusing the business on its core markets of France and Germany.

He has overhauled the team with probably the most important change being the appointment of Ajay Bhatia as CEO of Mobile.de, the German motors operation. Ajay was a senior exec at Carsales in Australia for many years so he has strong credentials from a high performance operation. I spoke to him a few months ago and thought he was strategically capable, realistic about the business he is now running, and pragmatic about how to navigate the required improvement. IMO, he is an exceptional hire and just the person who is needed in that role. The CFO has also been replaced.

My understanding is that Permira was closely involved in orchestrating this management overhaul. Permira stumped up almost EUR 2bn for a 11% stake at a price of NOK 150 in July 2021. Rumour has it the position is leveraged such that the stake is worth roughly zero at the current share price. Permira is highly motivated to extract some value out of this deal, particularly given it was a relatively rare foray into a non-control investment.

Current portfolio, and changes since the Ebay deal

After the Ebay deal Adevinta had operations in 20 countries. Most of the ex-Ebay assets outside Germany and Netherlands have since been sold or exited for pretty minimal compensation in aggregate. The portfolio now comprises 100%-owned businesses in all of the largest continental western European economies (Germany, France, Italy, Spain, Netherlands), a 50% stake in a Brazilian JV with Prosus, a Canadian business (which has been designated as non-core but which management wishes to improve before selling), and a couple of other small European assets.

However, the investment case revolves around France and Germany. There is material value in the Spanish / Brazilian / Dutch assets, but France and Germany will drive the outcome. That is reflected in management's focus - as the group works to simplify its operations and advance its products, everything will be built with France and Germany in mind and then adapted to the other operations where necessary / possible / economically viable.

I will similarly focus mainly on France and Germany, and then include the other businesses for valuation purposes.

Recent performance and market context

As I noted at the beginning, earnings estimates for Adevinta have come down significantly over the last couple of years. There are several drivers of this:

• Weakness in motors

  • Motors is a big part of the group revenue, and the used car market has had major volume challenges since COVID, reflecting constrained new car supply. Volumes are starting to recover, and will fully revert eventually (a year or two after new car volumes fully recover), but the constrained volumes are likely to persist for some time.

  • However, used car dealers have enjoyed a bonanza over the same timeframe thanks to very elevated used car prices, and healthy customers helps to drive strong pricing headroom - which Adevinta has been pushing into, particularly in Germany, where average revenue per listing is up from EUR13 in 2019 to EUR25 in Q2 2023.

• Tough online advertising market

  • The online advertising market generally has struggled over the last 18 months, whilst the ongoing trend of the major platforms sucking the air out of the market for other players has continued. Adevinta's (very high margin) display ad revenues have been weak, declining in most markets in recent periods.

• Ebay integration

  • Larger than expected investment to repair the under-invested Ebay assets

  • Longer than expected timeline to extract the synergies from the deal and improve the group margin

• Investment in loss-making transactional recommerce business across the group

  • Mercari in Japan was the pioneer in transactional recommerce (i.e. transaction is completed and postage arranged on-site), developing a large and highly profitable Japanese business. Vinted is the best operator outside of Japan, with a broad European presence. Adevinta was a little slow to emulate these models but started in 2018 and ramped up its efforts after COVID, launching a wallet product in 2022 and rolling out much of its better developed French model to its other sites.

  • The business is still nascent, but in Q2 2023 was a EUR 100m run-rate revenue business growing at 40% yoy in Q2 2023

  • It is loss-making as Adevinta seeks to build traction with modest take-rates and subsidised shipping, but should flip to profitability in 2024 / 2025 

However, with the used car volumes now in recovery, the business annualising the beginning of the poor online ads performance, and the new CEO almost a year into his tenure focusing on simplification and cost efficiency, I believe the trend of earnings estimate disappointments is behind us. Q2 2023 provided clear evidence for this, with the company raising profit guidance to the top end of its range. And with the stock down 60%, and the valuation MUCH lower, this means we can pay a low price to buy into a rapidly growing earnings stream which should see positive rather than negative revisions. 

Four major opportunities for the group looking forward

• More aggressive monetisation – easy to see scope for incremental EUR 700-900m of revenue without matching the performance of best practice international operators. High incremental margin on these classified revenue streams – c. 60-80%

  • There’s a couple of ways to think about the opportunity for Adevinta to increase monetisation of its assets. First, comparing against international peers, and secondly, pricing dynamics within the businesses themselves

  • Cars in Germany (EUR 350m incremental revenue) – opportunity to double revenue over a few years by reducing scale-based price dispersion

    • Mobile.de is a dominant asset. It has c. 50% advantage over the nearest competitor (autoscout24.de) in listings, and well over 3x advantage in traffic. This advantage would be much larger if we included car traffic from Kleinanzeigen (where most mobile.de listings are shown as part of a higher priced bundle).

    • Mobile.de has already repriced significantly over the last four years, almost doubling average revenue per listing (ARPL). However, the core opportunity remains significant – rebalancing pricing such that the discount obtained by scale dealers is smaller. Whilst ARPL is currently EUR 25, the smallest dealers pay more like EUR 100 – implying that the largest dealers must be paying perhaps EUR 20 or less. A scale discount is pretty standard, but classifieds sites in other countries have far smaller price dispersion. Management is committed to weight price increases in future towards larger dealer groups. Assuming price dispersion can be shifted towards a 2x range, which would be more typical, the implied revenue uplift is perhaps 150%-200% on half or more of the revenue base – i.e. roughly a doubling of the revenue (LTM revenue EUR 362m). This will no doubt take time to effect but gives a sense of the opportunity.

    • It’s worth noting that much of this uplift should eventually come through purely from volume reversion – listing volumes are currently a long way below pre-COVID normalised levels. Once new car registrations recover (and the supply chain constraints appear to have essentially disappeared in the last few months), used car transactions will follow, albeit with a lag of a couple of years. German car registrations are +14% yoy for 2023 through July.

  • Cars in France (Eur 150-200m incremental revenue) – monetising more in line with international benchmarks

    • In the motors vertical, LeBonCoin is also pretty dominant, although less so than Mobile.de in Germany. It has over 2x the listings of lacentrale.fr, the nearest competitor, and about 2.5x the traffic.

    • France has a little over 5m used car transactions per year, down from c. 5.6-5.7m pre-COVID. This compares to a current run-rate of c. 6m per year in Germany, and 7m per year in the UK. Average used car pricing is also materially lower in France than either the UK or Germany (new car prices are lower, and the average age of the fleet is a little older). I estimate the total gross profit pool for French car dealers is perhaps 60% that of Germany.

    • French motor vertical revenue is not regularly disclosed but was c. EUR 140m when last disclosed for LTM Q3 2021. I estimate 2022 motors revenue in France was c. EUR 160m. This suggests LeBonCoin is monetising materially less as a proportion of dealer gross profit than Mobile.de is in Germany, or Auto Trader in the UK. I estimate LeBonCoin revenues are c. 2-3% of the transactional gross profit pool in France, vs. c. 4-5% for Mobile.de and Auto Trader in Germany and the UK, and closer to 10% for Carsales in Australia.

    • This is not particularly surprising when examining the site. The user interface is shabby when compared to Auto Trader in the UK for instance. The content is limited – for instance many ads have just three photos compared to a majority with more than twenty on Auto Trader. I understand also that the services provided to dealers are far more limited than those offered by the best practice sites such Auto Trader and Carsales in Australia. The site seems to lack a well-structured prominence advertising model, and the data proposition is nascent.

    • France should be able to at least double its revenue based on these international benchmarks – however this will require significant product evolution. Under the leadership of Ajay Bhatia (who is head of the mobility vertical within Adevinta as well as his role running Mobile.de), I am confident this product development is now underway.

    • The French motors business will also enjoy a volume tailwind once recovered new car registrations flow through to used transactions: the French new car market is up 16% yoy through July 2023.

  • Property in France (EUR 150-300m incremental revenue) – monetising more in line with international benchmarks

    • In the property vertical, LeBonCoin is the market leader but with a less dominant position. In the 2021 capital markets day, Adevinta claimed 1.8x the web traffic of Seloger, the main rival. Again, the LeBonCoin user interface would benefit from some professionalisation / more photos / better aesthetics. The estate agent offering also needs evolving towards best practices in other countries.

    • In the 2021 capital markets day, Adevinta disclosed French property revenue of c. EUR 135-140m. I estimate this had grown to EUR 150m in 2022.

    • French estate agents achieve relatively generous commissions – typically around 4-5% or so. As a result, the commission pool in France is large, even though properties change hands fairly infrequently. In the last twelve months there were just over 1 million transactions. I estimate roughly two thirds were intermediated by agents. Assuming an average price of almost EUR 300k and a commission rate of 4%, this suggests the commission pool is perhaps EUR 7.5bn or so. Hence LeBonCoin is only taking around 2% of industry profit pool.

    • This compares to more like 7-8% for Rightmove in the UK, and materially higher for REA Group in Australia.

    • Again, the product in France needs significant evolution to fulfil the potential. But the opportunity in property is large – perhaps to triple or quadruple the revenue. I will include a doubling in my thinking for now.

• Recommerce in France, Germany, Netherlands (some overlap with Kleinanzeigen discussion): EUR 400m+ of incremental revenue. Incremental margin less clear, but best practice is 30% (Mercari in Japan)

  • France and Germany combined see well over 500m listings per year, of which around 150m are sold based on high level management guidance on sales conversion. Some of these are too bulky or too low value relative to shipping cost to be addressable under a transactional model, but most could be sold under this model

  • Let's assume 100m unit sales, at an average price of EUR 50 (based on Nov 2021 CMD disclosure) - this gives transactional GMV of c. EUR 5bn

  • Mercari achieves a 10% take rate ex-shipping, and c. 30% operating margin - i.e. EBIT is c. 3% of GMV

  • At the same 10% take rate, this would imply revenue of EUR 50m, vs. EUR 69m in 2022

  • Given the more competitive market, it's reasonable to assume Adevinta will achieve a lower margin. Let's assume 1.5-2%  of GMV instead - this suggests recommerce is a EUR 75-100m EBIT opportunity for the group (and significantly more than this in terms of improvement vs. 2023 given recommerce is currently materially loss-making)

  • With its operations in France, Benelux and Germany, Adevinta is well placed to offer a cross-border recommerce proposition which should increase platform liquidity and ability to monetise - this is likely to come in 2024

• Kleinanzeigen (wide range of outcomes – but at least a couple of hundred million EUR opportunity)

  • Kleinanzeigen is a very large web property with an active user base - it has over 40m unique users, around 1bn visits per month and well over 20m monthly app users. It is Germany's 6th biggest website by monthly unique visitors. However, under Ebay it was run on a shoestring - the total cost base in 2019 was EUR 67m. As a result, the product was unsophisticated and there are many low hanging fruit to be picked, aside from the transactional recommerce opportunity discussed above.

  • A couple of examples:

    • Google AdSense for Shopping

      • The Kleinanzeigen CEO estimates that in Germany the vast majority (c. 70%) of Google's AdSense for Shopping business was driven by ads on Kleinanzeigen

      • This involves Google placing Shopping ads on Kleinanzeigen's site, for which it takes a 32% commission!

      • Kleinanzeigen is planning to bring this operation in-house, eliminating the large Google take rate

      • Not only this - Kleinanzeigen argues it can offer a superior product to Google AdSense by utilising more detailed data than Google had access to for targeting purposes (such as responding to a consumer's historical and current session searches), so should be able to generate a pricing uplift as well

    • Serving SMB retailers properly

      • Kleinanzeigen has c. 250k users it regards as "professional", of which only c. 35k are currently paying customers

      • These are mostly relatively unsophisticated SMEs, in many cases using Kleinanzeigen as their primary or only channel

      • Kleinanzeigen is thoughtful about the profile of professional users it is targeting / willing to accommodate. It is consciously seeking to avoid following Ebay in losing the authenticity of the user experience by allowing business sellers to overwhelm the marketplace with commoditised inventory. It targets professional sellers with "unique" inventory

      • There are two ways to monetise this user base - through subscriptions, and through shifting them to adoption of the transactional model. Transactional will likely be the primary monetisation engine.

      • A key reason for the low level of monetisation is the extreme historical under-investment in the salesforce. Kleinanzeigen has more than doubled its employee base since the beginning of 2022, mainly due to investment in the sales force, and there is a lot more to come. The entire business only had 160 employees at the beginning of 2022!

  • Finally, Kleinanzeigen has begun to penetrate the property classifieds market, as Leboncoin did from a similar position fifteen years ago.

    • Germany's biggest property classifieds operator is Scout24 (publicly listed), and the #2 is Immoweb / Immowelt, which is owned by Axel Springer (privately owned since KKR buyout in 2019]

    • Given the structure of the German market, with only a little over half of transactions intermediated by agents, Kleinanzeigen has a strong opportunity to build a position in this market

    • So far it has signed up 10k agents (vs. 22k for Scout24, the market leader) and achieved a revenue run-rate of EUR 16m in Q2 2023

    • There is a realistic prospect of developing a #2 position in German property, which could sustain a EUR 200m+ revenue business. The UK #2, Zoopla, achieved property revenue of GBP 140m in LTM Mar 2018 before it was taken private, which was a little over half of Rightmove’s revenue run-rate. Scout24’s LTM revenue is EUR 500m.

• Cost efficiency – if verticalisation plan gets traction, low-40s margin target should prove unambitious

  • The main cost of a classifieds site is employees, and the largest (and most expensive, per head) function within the employee base is technology and product

  • Adevinta has over 3,000 technology employees, and c. 5,700 employees altogether. This equates to revenue of EUR 305k per employee – vs. c. EUR 540k at Auto Trader and c. EUR 600k at Rightmove.

  • Clearly Adevinta is not an efficiently run business. This is not surprising when one hears the CEO discussing the technology architecture of the group - shortly after starting as CEO, he mentioned that Adevinta has been replicating code 15-20 times in some cases. There are numerous different search technologies, several different chat functionalities, etc.

  • However, the efficiency varies significantly across the group: in Germany, Adevinta operates at c. EUR 625k per employee; in France and most of the rest of Europe c. EUR 320k, and Spain by far the bottom of the pack at c. EUR 130k per employee.

  • It will of course take time, but by simplifying, gradually removing duplicative technologies, and bringing the organisation to a position where much of the code base is shared across countries and businesses, and the sites run in a modular fashion, there is a lot of scope for efficiency gains. I expect this to come through by the group continuing to grow its revenues pretty rapidly, but without adding many employees, rather than through large cost-cutting programmes.

Summary of outlook

Adevinta management has provided, and reiterated several times, guidance for 2026: 11-15% revenue CAGR from the 2023 base, and 40-45% EBITDA margin in 2026. If delivered linearly, and assuming cash flow conversion of c. 100% of net income, this would leave the business on EV / EBITDA of c. 8.5x in 2026, or a FCF yield of over 8%. Given online classifieds assets generally trade at FCF yields of 2.5-4.5% in public markets today, clearly the market doesn't believe the guidance. So in essence, much of the discussion above comes down to validating the guidance. If we can believe it, the stock is a no-brainer on a 2-3 year view.

It seems pretty obvious that the potential is there. The question is more whether the capability is. The beauty of the businesses that Adevinta owns is that they are so robust that whilst the opportunity has not been properly seized for over a decade, it remains largely undiminished - in the sense that the businesses' competitive position is largely unchanged. Well established network effects are extremely tough competitive barriers to overcome.

My take is that with the current management we have decent odds of Adevinta transitioning, eventually, to a high-performing classifieds operation. If that is the case, management's margin guidance should be conservative. As product development accelerates and the sites' value-add to their clients improves, pricing power should strengthen, and I believe that this is possible without adding much to the cost base. If the businesses really fulfil their potential, there are many years of double digit, high margin growth ahead.

Valuation

By 2026, Adevinta should be able to generate perhaps EUR 2.6bn of revenue, and EUR 1.1bn of EBITDA - assuming c. 55% incremental margin. I believe this should translate to c. EUR 600m of FCF. If we model assuming that the business runs with 2x leverage, and uses excess cash for buybacks (optimistic, but one can dream!), then the share count could reduce by almost 15% by the end of 2026 as well. This would leave Adevinta generating over NOK 6 of FCF per share in 2026, approaching NOK 8 of FCF per share in 2027. With sustained double digit revenue growth and improving margins, I think the stock would be pretty attractive at NOK 160 under those circumstances at the end of 2026.

With these assumptions, the return from today's NOK 79 price would be low-mid 20s IRR.

Alternatively, consider the discount to more mature, slower-growing, but admittedly more consistent and better managed businesses such as Rightmove / Auto Trader / Carsales etc. Adevinta trades at a discount to these businesses, adjusted for weaker cash conversion, despite materially higher earnings growth potential.

Risks / concerns

• Growth in France has been somewhat disappointing for a few periods

• Transactional generalist business may struggle to achieve decent economics given the strong competition from Vinted and others

• "Verticalisation" project could become another tech transition debacle

• Ajay Bhatia could leave – I think he is important to the transformation programme

Catalyst

• Continuing operational delivery, resulting in earnings upgrades

• Removal of overhang / increase in free float and stock liquidity

• Potential for bid from e.g. Prosus