Applovin Corp (APP) 2021 Q4 法說會逐字稿

Greetings. Welcome to AppLovin Corp.'s Fourth Quarter and Full Year 2021 Earnings Results Conference Call. (Operator Instructions) Please note, this conference is being recorded.

大家好！歡迎參加 AppLovin Corp. 2021 年第四季及全年業績電話會議。 （操作員指示）請注意，本次會議正在錄製中。

I will now turn the conference over to Ryan Gee, Head of Investor Relations and Strategic Finance for AppLovin. Thank you. You may begin.

現在，我將會議交給 AppLovin 投資者關係和策略財務主管 Ryan Gee。謝謝。您可以開始了。

Thank you, Sherry, and welcome, everyone, to AppLovin's earnings call for the quarter ended December 31, 2021. Joining me today to discuss our results are our Co-Founder, CEO and Chairperson, Adam Foroughi; and our President, Chief Financial Officer, Herald Chen.

謝謝 Sherry，歡迎大家參加 AppLovin 截至 2021 年 12 月 31 日的季度財報電話會議。今天與我一起討論我們業績的有我們的共同創辦人、執行長兼董事長 Adam Foroughi；以及我們的總裁兼財務長 Herald Chen。

Please note our SEC filings, earnings release and shareholder letter discussing our 4Q and 2021 performance are available at investors.applovin.com.

請注意，我們的 SEC 文件、收益報告和討論我們第四季度和 2021 年業績的股東信可在 investors.applovin.com 上查閱。

During today's call, we may be making forward-looking statements regarding future events and the future financial performance of the company. These statements are based on assumptions and beliefs, and we assume no obligation to update them. Actual results may differ materially from those results predicted here today. Please review the risk factors in our most recently filed Form 10-Q as well as elsewhere in our SEC filings for further clarification.

在今天的電話會議中，我們可能會就未來事件和公司未來財務表現做出前瞻性陳述。這些陳述是基於假設和信念，我們不承擔更新這些陳述的義務。實際結果可能與今天在此預測的結果有重大差異。請查看我們最近提交的10-Q表以及其他提交給美國證券交易委員會的文件中的風險因素，以獲取進一步的說明。

We will also be discussing our non-GAAP financial measures. Reconciliations of GAAP to non-GAAP financial measures are included in our earnings press release, our shareholder letter, our 10-Q and our 10-K to be filed shortly.

我們也將討論非公認會計準則 (non-GAAP) 財務指標。 GAAP 與非 GAAP 財務指標的對帳表將包含在我們即將提交的財報新聞稿、股東信以及 10-Q 和 10-K 報表中。

Please be sure to review those reconciliations as non-GAAP measures are not intended to be a substitute for or superior to our GAAP results. This conference call is being recorded, and a replay will be available on our IR website shortly.

請務必查看這些對帳表，因為非公認會計準則 (Non-GAAP) 指標並非旨在替代或優於我們的公認會計準則 (GAAP) 績效。本次電話會議正在錄製中，重播將很快在我們的投資者關係 (IR) 網站上提供。

With that, I'll now turn it over to Adam.

現在，我將把發言權交給亞當。

Thanks, Ryan, and thank you all for joining us today.

謝謝，瑞安，也謝謝大家今天加入我們。

I'm pleased to report record fourth quarter financial results. I'm even more excited by where our software business is heading and our potential in 2022 and beyond. I want to highlight 2 reasons why we confidently raised our software revenue outlook for 2022 and why we believe we can grow to approximately $2 billion in software revenue in 2023.

我很高興地宣布，第四季的財務表現創下了歷史新高。我對軟體業務的發展方向以及2022年及以後的潛力更加興奮。我想強調兩個原因，說明我們為何滿懷信心地上調了2022年的軟體收入預期，以及我們為何相信2023年的軟體收入能夠成長至約20億美元。

First, the compounding of our software platform business exiting 2021. It's hard to fathom, but we have nearly 3x as many clients using our solution to grow their businesses today versus last year. And our software platform revenue was larger in the fourth quarter than for all of 2020 and grew 27% quarter-over-quarter. This marked our third consecutive quarter of double-digit sequential growth and significantly higher growth than all other industry peers.

首先，我們的軟體平台業務在2021年即將結束時將實現複合成長。這很難理解，但與去年相比，目前使用我們解決方案來發展業務的客戶數量幾乎增加了三倍。我們的軟體平台營收在第四季超過了2020年全年，季增了27%。這標誌著我們連續第三個季度實現兩位數的環比增長，並且增長顯著高於所有其他行業同行。

AppDiscovery continues to perform very well and consistently delivers against our clients' ROI goals. That is why we had over 200% net dollar-based revenue retention for our existing clients, which provides a solid foundation for a robust organic growth rate to persist in '22 and beyond. Also consider our sales force is still ramping and that AXON just turned 1 year old last quarter.

AppDiscovery 持續表現優異，並始終如一地達成客戶的投資報酬率目標。正因如此，我們現有客戶的淨美元收入留存率超過 200%，這為 2022 年及以後保持強勁的有機成長率奠定了堅實的基礎。此外，我們的銷售團隊仍在不斷壯大，AXON 上季剛成立一週年。

Second, the integration of MoPub. We've seen amazing progress to date integrating MoPub features, publishers and demand partners. The number of apps monetizing with MAX is up over 60% since we announced the deal. And over 90% of the largest publishers for MoPub are actively integrating our unified offering.

第二，MoPub 的整合。迄今為止，我們在整合 MoPub 功能、發行商和需求合作夥伴方面取得了顯著進展。自從我們宣布合作以來，使用 MAX 獲利的應用程式數量增加了 60% 以上。超過 90% 的 MoPub 大型發行商正在積極整合我們的統一產品。

The AppLovin Exchange will enable many new DSPs and bidders to work with MAX. These demand partners generated the vast majority of MoPub's revenue, but they didn't have direct access to MAX publishers before. As they come online through our unified platform, which will be over twice as big as MoPub, we expect these partnerships to deliver a compelling revenue opportunity in '22 and beyond.

AppLovin Exchange 將使許多新的 DSP 和競標商能夠與 MAX 合作。這些需求合作夥伴創造了 MoPub 的絕大部分收入，但他們之前無法直接接觸 MAX 出版商。隨著他們透過我們統一的平台（規模將是 MoPub 的兩倍多）上線，我們預計這些合作關係將在 2022 年及以後帶來極具吸引力的收入機會。

We believe to operate a fast-growing business successfully over the long term, we need to be executing against our current opportunities and also be building products that can be catalysts for growth years into the future. I'm very proud of our team because in addition to executing on all of our current initiatives, we're building products in big markets, such as blockchain and NFTs for mobile content, marketing solutions for mobile OEMs and carriers, and enabling performance marketing in the fast-growing CTV market. These categories have similarities to the business and technologies we've already built and success in any one of these can meaningfully expand our business. We look forward to sharing our progress on these initiatives and others as they develop.

我們相信，要長期成功經營快速成長的業務，我們需要抓住當前機遇，並打造出能夠在未來數年成為成長催化劑的產品。我為我們的團隊感到非常自豪，因為除了執行我們目前的所有計劃之外，我們還在大型市場中構建產品，例如用於移動內容的區塊鏈和NFT、面向移動OEM和運營商的營銷解決方案，以及在快速增長的CTV市場中實現效果營銷。這些類別與我們已建立的業務和技術相似，其中任何一個領域的成功都能顯著拓展我們的業務。我們期待分享這些計劃以及其他計劃的進展。

With that, let me turn it over to Herald to provide color on our numbers and outlook.

有了這些，請容許我把話題轉到《先驅報》上，為大家介紹我們的數據和前景。

Thanks, Adam, and thanks to everyone for taking the time to join us.

謝謝，亞當，也謝謝大家抽空來加入我們。

As Adam noted, we had a very solid Q4 and an outstanding 2021. In 2021, we not only had great financial results, where revenue grew more than 90% and adjusted EBITDA grew more than 100%, we were able to fully ramp our AXON machine learning platform, build critical mass in our first-party apps, complete strategic acquisitions in both tech and content, access to public equity markets and private debt markets, and importantly, build out our global team and Board. This positions our company well for future growth.

正如Adam所言，我們第四季業績穩健，2021年更是精彩。 2021年，我們不僅取得了優異的財務業績，營收成長超過90%，調整後EBITDA成長超過100%。此外，我們也全面提升了AXON機器學習平台的水平，在第一方應用中累積了一定規模，完成了科技和內容領域的策略性收購，成功進入公開股票市場和私募債市場，更重要的是，我們組建了全球團隊和董事會。這為公司未來的成長奠定了堅實的基礎。

Instead of reading our financial and operating metrics for Q4 and 2021, which are well laid out in our earnings release and shareholder letter, I do want to highlight our outlook for '22 and software target for 2023. First, we plan to grow our scaled software business at over 100% for 2022. We're projecting software platform revenue for 2022 at just over $1.4 billion, at the midpoint of our range, representing 111% year-over-year growth. This is a sizable upward revision from the $1 billion target we gave you earlier in 2021.

我不想細數我們第四季和2021年的財務和營運指標（這些指標在我們的財報和股東信中都有詳細列出），而是想強調一下我們對2022年的展望以及2023年的軟體目標。首先，我們計劃在2022年將規模化軟體業務的成長速度提高100%以上。我們預計2022年的軟體平台營收將略高於14億美元，處於預期區間的中點，年增111%。這比我們在2021年初設定的10億美元目標有了大幅調高。

Our outlook for continued robust growth stems from: one, our strong finish to 2021 and record Q4 '21 results, setting us up well for organic growth in 2022; and two, as Adam mentioned already, our outlook benefits from the solid progress integrating MoPub into MAX.

我們對持續強勁成長的展望源於：第一，我們在 2021 年取得了強勁的成績，並創下了 2021 年第四季度的業績記錄，為 2022 年的有機增長奠定了良好的基礎；第二，正如亞當已經提到的，我們的前景受益於將 MoPub 整合到 MAX 中的穩步進展。

While going forward, we won't be reporting on MoPub stand-alone since it is being fully integrated into the MAX platform, we continue to believe the MoPub acquisition should be a highly accretive and strategically impactful investment for our company in the near term and over the long term, as evidenced by our software guidance for '22 and '23, which I'll come to in a second.

雖然未來我們不會單獨報告 MoPub，因為它將完全整合到 MAX 平台中，但我們仍然相信，收購 MoPub 對我們公司來說應該是一項具有高度增值性和戰略影響力的投資，無論是在短期還是長期，這從我們對 22 年和 23 年的軟體指導中可以看出，我稍後會談到這一點。

I do want to call out that we expect to pay approximately $200 million in migration fees to publishers coming on to the MAX platform, mostly in the first quarter. Given these costs are related to the MoPub transaction, we plan to add back a substantial portion of that amount to our adjusted EBITDA. Post the MoPub acquisition, we would not be adding back those fees in the ordinary course.

我想特別指出的是，我們預計將向加入 MAX 平台的出版商支付約 2 億美元的遷移費用，大部分將在第一季支付。鑑於這些成本與 MoPub 交易相關，我們計劃將其中很大一部分加回到調整後的 EBITDA 中。收購 MoPub 後，我們通常不會再加回這些費用。

With regard to our '23 outlook for software, given all the progress we've made to date and the many opportunities we see ahead, we are working hard to reach the $2 billion of software revenue in 2023, as Adam mentioned. On the app side, our outlook is for $2.2 billion in 2022, at the midpoint of our range. It is amazing to say we had negligible apps revenue just 4 years ago and now have a multibillion-dollar business with an awesome set of studios and developers around the globe.

關於我們對2023年軟體業務的展望，鑑於我們迄今為止取得的所有進展以及我們預見的眾多機遇，正如Adam所提到的，我們正在努力實現2023年軟體收入20億美元的目標。在應用方面，我們預計2022年的應用收入將達到22億美元，處於我們預期範圍的中間值。令人驚訝的是，就在四年前，我們的應用收入還微不足道，而如今，我們的業務已擁有數十億美元的規模，並在全球擁有許多優秀的工作室和開發者。

While we've reached critical mass to drive first-party data -- the first-party data we need for our machine learning engine, we are continuing to make substantial investments in new content and are planning to release new first-party apps with evergreen potential over the next several quarters.

雖然我們已經達到驅動第一方數據的臨界點——我們的機器學習引擎所需的第一方數據，但我們仍在繼續對新內容進行大量投資，併計劃在未來幾個季度發布具有常青潛力的新的一方應用程式。

The midpoint of our outlook at $2.2 billion assumes modest revenue from new content, given the inherent difficulty in projecting new content performance. We project our new apps to have a more meaningful impact in the second half of '22.

鑑於預測新內容業績本身的難度，我們22億美元的預期中位數假設新內容帶來的收入不大。我們預計，我們的新應用將在2022年下半年產生更顯著的影響。

Note that if we do have a new hit title, we would ramp our user acquisition to drive revenue as we did with Project Makeover over the last year. As it relates to our EBITDA for '22, we're targeting an adjusted EBITDA margin in the high 20s percent range, up from 21 at 26%.

請注意，如果我們真的有一款新的熱門遊戲，我們會加大用戶獲取力度，以提升收入，就像我們去年對《Project Makeover》所做的那樣。就2022年的EBITDA而言，我們的目標是將調整後的EBITDA利潤率提高到20%左右，高於2021年的26%。

While Adam and I are focused on compounding free cash flow over the long term, we are and will invest for the long-term value creation rather than near-term profitability.

雖然亞當和我都專注於長期增加自由現金流，但我們現在和將來都會投資長期價值創造，而不是短期獲利。

Lastly, and importantly, Adam, Ryan and I would like to thank you for your support in 2021, and please know that we and our AppLovin team are working hard to deliver for all of you in '22 and beyond.

最後，也是最重要的一點，亞當、瑞安和我要感謝大家在 2021 年的支持，請知悉，我們和我們的 AppLovin 團隊正在努力在 2022 年及以後為大家提供服務。

With that, we'd be happy to take your questions. Thank you.

好了，我們很樂意回答大家的提問。謝謝。

(Operator Instructions) Our first question is from Stephen Ju with Credit Suisse.

（操作員指示）我們的第一個問題來自瑞士信貸的 Stephen Ju。

Okay. So Herald, just to kind of follow up on the migration costs of $200 million for 2022, and it looks like there was about $3 million or thereabouts for the fourth quarter. So can you discuss what this expense actually entails? And to your point, I guess, change in mediation platforms are pretty rare. So can you talk about what's actually going on underneath the hood in terms of the nuts and bolts of the migration?

好的。 《先驅報》，我想問2022年2億美元的遷移成本，看起來第四季大約300萬美元。能談談這筆費用的具體意義嗎？正如您所說，中介平台的變更非常罕見。能談談遷移過程中的具體細節嗎？

And Adam, undoubtedly, you've seen sort of the latest privacy-driven changes that Google was talking about for Android. So can you talk about how things might be the same or how things might be different versus what Apple did on iOS?

亞當，毫無疑問，您已經看到了谷歌在 Android 上談論的最新隱私驅動變化。那麼您能談談與蘋果在 iOS 上所做的相比，Android 系統可能有哪些相同之處，又有哪些不同之處嗎？

Stephen, thanks for the question. Historically, as we said, in the mediation space, there's been little volatility. And people, once they're on a platform, stay there because it is inherently integrated to all your applications. So very, very sticky. And historically, we've not paid these fees in the past. But with the acquisition of MoPub, and MoPub being shut down as a service at the end of this quarter, a lot of publishers needed to move quickly, and there's quite a bit of cost to them actually from a revenue standpoint where they potentially can lose revenue given the speed with which they need to migrate. And there's just cost to get the migration done. And so we negotiated these onetime fees to cover those costs.

Stephen，謝謝你的提問。正如我們所說，從歷史上看，在聚合領域，波動性很小。人們一旦使用一個平台，就會一直留在那裡，因為它本質上與你所有的應用程式整合在一起。所以使用者黏性非常強。從歷史上看，我們過去沒有支付過這些費用。但隨著對 MoPub 的收購，以及 MoPub 在本季末作為一項服務被關閉，許多出版商需要迅速採取行動，從收入的角度來看，這實際上給他們帶來了相當大的成本，考慮到他們需要快速遷移，他們可能會損失收入。而且遷移本身就是有成本的。所以我們協商了這些一次性費用來支付這些成本。

Stephen, it's Adam. On your second question around Google's announced privacy changes. A couple of things there. One is, we don't know exactly the direction these changes will take because it's a multiyear plan. And Google works with all the major companies in the sector to ensure that there's not a disruption of performance on advertising targeting and attribution, and then pair that with a better privacy environment for the consumer. So it's just going to take a while to develop.

史蒂芬，我是亞當。關於你關於谷歌宣布的隱私改革的第二個問題。我想說幾點。首先，我們不清楚這些改革的具體方向，因為這是一個多年計畫。谷歌與該領域的所有主要公司合作，以確保廣告定位和歸因的效果不會受到影響，並為消費者提供更好的隱私環境。所以這需要一段時間來發展。

That said, we obviously dealt with IDFA changes coming really quickly to the market in 2021. And you saw our business grow quite considerably on the software side in '21. Going into the changes, we talked about the privacy change on Apple being muted for our business. You saw us growing 30% quarter-over-quarter in Q2. You saw the same trend in Q3 and Q4. So our business is growing quite quickly. And these changes actually, in a way, do benefit us because the core of our data that drives our machine learning is all first-party data achieved and access from our own games. So we believe we're set up very well to perform regardless of any sort of privacy changes or regulatory changes in the ecosystem.

話雖如此，我們顯然已經應對了 2021 年市場上迅速變化的 IDFA 規則。大家也看到了，我們的軟體業務在 2021 年實現了大幅成長。談到這些變化，我們談到了蘋果隱私權保護政策的變化對我們業務的影響。大家看到我們在第二季較上季成長了 30%。在第三季和第四季也看到了同樣的趨勢。所以我們的業務成長相當快。這些變化實際上在某種程度上對我們有利，因為驅動我們機器學習的資料核心都是從我們自己的遊戲中獲取和存取的第一方資料。因此，我們相信，無論生態系統中出現什麼樣的隱私保護變化或監管變化，我們都已經做好了充分的準備，能夠保持良好的績效。

Our next question is from David Karnovsky with JPMorgan.

下一個問題來自摩根大通的 David Karnovsky。

Adam or Herald, can you maybe walk through what's embedded in your guide for expense growth in 2022? Your outlook, I think, has $750 million of software growth at the midpoint. And I know some of that is MoPub, which has a higher expense structure initially. But just trying to understand if your expectation for incremental margins on the underlying software business has changed at all.

Adam 或 Herald，你能否解釋一下你們在 2022 年費用成長指南中提到了什麼？我認為，你們的預期是軟體業務成長中位數為 7.5 億美元。我知道其中一部分是 MoPub，它最初的費用結構更高。但我想了解一下，您對基礎軟體業務利潤增量的預期是否會改變。

Thanks for the question, David. It's Herald. So yes, overall, we think the acquisition of MoPub with the $1 billion we paid in this incremental couple hundred million dollars to get publishers migrated is a highly accretive, as mentioned, and strategic deal for us. So we do feel very good about that. And we do think the flow-through on the software side margin structure is high. There is, as you note, more infrastructure to be built. And as we're scaling and doubling the size of our software business, we're having to keep up with that for sure. And we're bringing on a lot of new DSPs and a lot of new publishers simultaneously.

謝謝你的提問，David。我是 Herald。所以，是的，總的來說，我們認為，我們用這筆增量資金中的10億美元收購 MoPub，用於幫助出版商遷移，正如我們之前提到的，這筆交易對我們來說是一項極具增值潛力且具有戰略意義的交易。所以我們對此感到非常滿意。我們確實認為，軟體端利潤結構的流動率很高。正如你所提到的，還有更多的基礎設施需要建造。隨著我們軟體業務規模的擴大和翻倍，我們必須跟上這一步伐。我們正在同時引入許多新的 DSP 和許多新的出版商。

I think for our guide for the year, maybe just taking a bigger-picture look at it in the high 20s EBITDA margin. As mentioned, we want to be investing for growth in the business, and we think our ability to continue to grow the business over the long term at scale requires that investment. So we're trying to include what we think are some new projects in that EBITDA number. We are investing quite a bit into our studios as well. I think you know that we acquired a numbers again at the end of last year. We're scaling the investment into those studios to ensure that they've got the right resources to deliver content on the back end.

我認為，對於我們今年的業績指引，或許應該從20%以上的EBITDA利潤率來更宏觀地看待。如前文所述，我們希望投資於業務成長，我們認為，我們長期持續大規模成長的能力離不開這筆投資。因此，我們正努力將一些我們認為是新項目的項目納入EBITDA數據中。我們也在工作室方面投入了不少資金。我想你們也知道，我們在去年年底又獲得了一些數據。我們正在擴大對這些工作室的投資，以確保他們有足夠的資源來提供後端內容。

So we can certainly drive the business in a different direction in terms of overall margin given the flow-through, but there's so many opportunities for us to invest, both on software and apps where we'd like to invest in both of them.

因此，考慮到流通量，我們當然可以在整體利潤率方面推動業務朝著不同的方向發展，但我們有很多投資機會，無論是在軟體還是應用程式方面，我們都願意投資。

Okay. And then Adam, in your prepared remarks, you mentioned some areas for investment like marketing solutions for OEM or blockchain. Just wondering if you could expand on this other investment costs against that in 2022. And if you can provide any specific details why are these the right opportunities to invest in?

好的。然後，Adam，在您的準備演講中，您提到了一些投資領域，例如OEM或區塊鏈的行銷解決方案。我想知道您是否可以詳細說明一下2022年其他投資成本。您能否提供一些具體細節，為什麼這些是適當的投資機會？

Yes. So when we look at anything that's new, we do that incremental to what our current opportunities are. But we do want them to be related to the expertise that we have, both in terms of technology platform, audience and data so that we have a higher likelihood of success if we do invest in something. And so the 3 areas I mentioned, the carrier and OEM space, marketing solutions, to really enable manufacturers to generate more from their device sales than they currently generate from the solutions they have available, is something that's a natural extension of our machine learning demand and software platform.

是的。所以，當我們審視任何新事物時，我們都會根據現有機會進行漸進式的探索。但我們確實希望這些新事物與我們現有的專業知識相關，包括技術平台、受眾和數據，這樣，如果我們進行投資，成功的可能性就會更高。我提到的三個領域，運營商和原始設備製造商領域，以及行銷解決方案，能夠真正幫助製造商從設備銷售中獲得比現有解決方案更高的收益，這是我們機器學習需求和軟體平台的自然延伸。

The NFT and blockchain marketplace and solutions there are just incremental ways to help monetize mobile gamers. And we've got 200 million of them playing our games. And then across our audience network, we access 2 billion of them every single month. And so because of our penetration in the mobile ecosystem, we think helping enable developers to expand their monetization opportunity with solutions there is beneficial to all parts of our business. So it makes sense to invest there.

NFT、區塊鏈市場和解決方案只是幫助行動遊戲玩家獲利的漸進式途徑。我們的遊戲玩家數量已達2億人。而在我們的受眾網路中，每月觸達的玩家數量更是高達20億。鑑於我們在行動生態系統的滲透率，我們認為幫助開發者利用解決方案拓展獲利機會，對我們業務的各個方面都大有裨益。因此，投資行動遊戲市場是明智之舉。

And then finally, the connected TV space. Very similar to what we do today. We're serving full-screen, immersive video advertisements on mobile devices, and extending that and the performance models to TV as a new and fast-growing digital access point for content feels like a natural extension of our platform.

最後是連網電視領域。這與我們目前所做的非常相似。我們在行動裝置上提供全螢幕沉浸式影片廣告，將這種模式及其效果擴展到電視，將其作為新的、快速增長的數位內容存取點，這感覺就像是我們平台的自然延伸。

All of these are covered in the guidance that we've given you in terms of costs. If any of them hit -- and frankly, hitting on a new business is always hard, but these are so related to what we do, but if any of them hit it dramatically widens and it expands our market opportunity and growth opportunities for years to come, and that's why we're investing there.

所有這些，都包含在我們為您提供的成本指導中。如果其中任何一個項目取得成功——坦白說，開拓新業務總是很困難，但這些項目與我們的業務息息相關——那麼，如果其中任何一個項目取得成功，都會極大地拓寬我們未來幾年的市場機會和成長機會，這就是我們投資於新業務的原因。

Our next question is from Tim Nollen with Macquarie.

下一個問題來自麥格理的蒂姆·諾倫。

I'd like to ask about the kind of the components of the software platform growth, if I could. So you had very, very, very strong growth, especially in Q2 and Q3, and really kind of building up from AXON rolling on to the -- to be a contributor to that service starting a year ago, which, as you mentioned, you've now lapped against. So not surprising that, that rate of growth has slowed from those super high rates in Q2 and Q3. I'm just wondering in Q4, if you can give us an idea of like what was an organic versus an AXON-related figure, if that's possible to do?

如果可以的話，我想問一下軟體平台成長的組成部分。你們的成長非常非常強勁，尤其是在第二季度和第三季度，這實際上是從AXON的推出開始的——一年前你們開始為該服務做出貢獻，正如你提到的，現在你們已經超越了AXON。所以，成長率從第二季和第三季的超高水準放緩也就不足為奇了。我只是想知道，在第四季度，如果可以的話，你能告訴我們有機成長和AXON相關成長的數據分別是多少嗎？

And then looking ahead to the '22 guidance, a nice big number there. Is it possible to give us an update on what the MoPub contribution would be to that versus any AXON versus any other elements?

展望22年的業績指引，這是一個不錯的數字。能否更新一下MoPub對此的貢獻，以及AXON和其他元素的貢獻？

Okay. On the first part -- do you want to -- I'll handle the first part. Herald, do you want to jump in on the second?

好的。關於第一部分——你想——我會處理第一部分。 《先驅報》，你想加入第二部嗎？

So on the first part, our software business is pretty much all organic. Really, Adjust was the only material acquisition we had before we announced closing MoPub in January. And in Q4, we saw a 27% quarter-over-quarter growth. Obviously, the numbers are getting bigger each quarter, and yet, the quarter-over-quarter growth is much higher than anything anyone sees in the industry on the software side and it has kept up.

首先，我們的軟體業務幾乎完全是有機成長的。實際上，Adjust 是我們在今年 1 月宣布關閉 MoPub 之前唯一一項實質收購。第四季度，我們的環比成長達到了 27%。顯然，這個數字每個季度都在增長，但環比增長遠高於軟體行業的任何水平，並且保持著成長勢頭。

And what we're seeing, even though AXON is maturing and we lapped the prior year's numbers with AXON, is that advertisers who are using our solution are investing more dollars every single quarter into the platform because as data builds and they can validate that the performance results are very strong, they want more of it. And that's why you saw a huge step-up in dollars per SPEC in the quarter. And we expect to continue to see AXON improve over time because the machine learning systems, as they get more data and they get more scale, benefit from that and become more and more accurate on behalf of the customers that are using it.

儘管 AXON 日趨成熟，並且我們的業績與上年持平，但我們發現，使用我們解決方案的廣告商每個季度都會在該平台上投入更多資金，因為隨著數據的積累，廣告商能夠驗證其效果非常出色，他們希望獲得更多資金。這就是為什麼您在本季度看到每 SPEC 的美元數量大幅增長。我們預計 AXON 會隨著時間的推移而持續改進，因為隨著機器學習系統獲得更多數據並擴大規模，它們會從中受益，並為其客戶帶來越來越精準的預測。

Going forward, MoPub presents an opportunity for continued expansion, more access to demand and then obviously, a large expansion of access to audience on our marketing platform, MAX. And MoPub, as of last year, had a very strong year. And then coming into January, it was doing very well, but it is going to be completely rolled into our solution. We're going to take those demand partners, the DSPs and advertisers that were buying through MoPub, to gain access to the MoPub audience and we're integrating them into our MAX solution to give them direct access to the unified offering, which is now going to be over twice as big as MoPub was.

展望未來，MoPub 為我們提供了持續擴張的機會，讓我們能夠更好地觸及需求，進而顯著拓展我們行銷平台 MAX 的受眾覆蓋範圍。截至去年，MoPub 的表現非常強勁。進入 1 月份，它的表現仍然出色，但我們將全面整合到我們的解決方案中。我們將幫助那些透過 MoPub 購買廣告的需求合作夥伴、DSP 和廣告主接觸 MoPub 的受眾，並將他們整合到我們的 MAX 解決方案中，讓他們能夠直接存取統一的產品/服務。現在，該產品/服務的規模將是 MoPub 的兩倍多。

So we think that's going to present an opportunity to create revenue growth from that access point for possibly years to come. And we're very optimistic about the acquisition because we've seen so little churn. As mentioned, 90% plus of all top publishers are migrating over to our platform.

因此，我們認為這將為未來幾年透過該接入點創造收入成長提供一個機會。我們對此次收購非常樂觀，因為我們發現客戶流失率很低。如前文所述，超過90%的頂級出版商正在遷移到我們的平台。

I'll hand it off to Herald for the second part.

我會將其第二部分交給《先驅報》。

Yes, Tim, I think a lot of the foundational growth drivers Adam mentioned there for '22, look, we have a great organic business. And quarter-over-quarter, from Q3 to Q4, right, that was 28% increase, which is all organic, right? So that's a pretty big move rolling into '22. Then we have the AppDiscovery continuing to improve with underlying AXON. We also have Adjust doing well and a good SaaS business growing at the rates that we'd hoped. As mentioned, it's difficult then to parse out the impact of MoPub and MAX and exactly how that lays out. I know previously, we gave a number that the run rate of MoPub would be in the mid-200s by the end of the year. We still remain very confident in that, and frankly, probably believe there's upside to that.

是的，提姆，我認為亞當提到的許多2022年的基礎成長動力，例如我們擁有出色的有機業務。從第三季度到第四季度，環比增長了28%，這完全是有機增長，對吧？所以，這對於2022年來說是一個相當大的進步。此外，我們的AppDiscovery在AXON的支撐下持續改進。我們的Adjust也表現良好，SaaS業務也以我們預期的速度成長。如同前面所提到的，現在很難分析MoPub和MAX的影響以及它們的具體佈局。我知道之前我們給的數字是，到今年年底，MoPub的運行率將達到250左右。我們仍然對此非常有信心，坦白說，我們可能相信它有上升空間。

One important note was we did give guidance on that revenue the way we did because it takes time to ramp, as mentioned before, migrate from one mediation platform to another one. One, to get the publishers to migrate; and then to get the demand side to migrate. And so we're going through that process now, as you know. But the impact of that in terms of the actual revenue booked in the year, therefore, will be less than the $250 million. It's more of a $250 million run rate was the target when we announced the deal. Now that we're well into it, we feel very good about that number as well.

需要注意的是，我們之前確實給出了關於收入的指引，因為正如之前提到的，從一個聚合平台遷移到另一個聚合平台需要時間。首先，需要讓發布商遷移；然後，需要讓需求方遷移。正如你所知，我們現在正在經歷這個過程。但就今年實際收入而言，這筆交易的影響將低於2.5億美元。我們宣布交易時設定的目標更接近2.5億美元的營運率。現在我們已經深入推進，我們對這個數字也感到非常滿意。

So that led us to, what is, I guess, not a small range in terms of software target, at the high end of 15 at $1.5 billion midpoint, just over $1.4 billion. And then where you can really see the impact overall is the guide to our target of trying to hit the $2 billion number in '23, just goes to the fact that we've got a lot of confidence in putting these things together. And then the ecosystem building around that should allow us to drive pretty significant growth for this year and next.

所以，就軟體目標而言，我想這應該是一個不小的區間，在15個軟體專案的高端，中間點是15億美元，略高於14億美元。然後，你真正能看到整體影響的是我們力爭在2023年達到20億美元的目標，這說明我們對實現這些目標充滿信心。圍繞這些目標所建構的生態系統應該能讓我們在今年和明年實現相當可觀的成長。

And if I could just follow up quickly. I think last quarter, you gave us some indication of the Adjust contribution to revenues. Is there a similar type of number you can give us now?

我能快速跟進一下嗎？我記得上個季度您已經透露了Adjust對營收的貢獻。現在您能提供類似的數字嗎？

Yes. Adjust, when we acquired it, it was around $100 million, and it's growing at a solid double-digit rate. We'll continue to do that, we think, for '22 and '23 as well.

是的。我們收購Adjust時，它的估值約為1億美元，目前正以兩位數的穩健速度成長。我們認為，在2022年和2023年，我們仍將保持這一成長動能。

Our next question is from Jason Bazinet with Citigroup.

我們的下一個問題來自花旗集團的 Jason Bazinet。

Just one quick question. You guys had done so much M&A and have been very successful doing it. I just wonder, as you sort of look at the landscape, do you feel like most of the sort of acquisitions are behind you and it's now more about investing organically in the assets that you have? Or do you think there are other sort of interesting assets that you could plug into your existing portfolio?

就問一個簡單的問題。你們做過很多併購，而且非常成功。我只是想知道，從目前的情況來看，您是否覺得大多數收購都已經過去，現在更專注於對現有資產進行有機投資？或者，您認為有其他值得關注的資產可以添加到現有的投資組合中？

Jason, it's Adam, and thanks for the question. So 3 years ago, when we got into gaming, we made it strategic and focused. And obviously, gaming is super fragmented, but it became strategic to us because we wanted to accumulate content across every single category of gaming so that we can have that data to feed into the software platform and execute on, obviously, what you're seeing flow out of our numbers today. We've checked the box on gaming. So we've got enough of the gaming business and enough audience and data that the software business can do its thing.

傑森，我是亞當，謝謝你的提問。三年前，當我們進入遊戲領域時，我們就制定了策略，並專注於此。顯然，遊戲產業非常分散，但這對我們來說具有戰略意義，因為我們希望累積每個遊戲類別的內容，以便將這些數據輸入軟體平台，並執行你今天從我們的數據中看到的結果。我們已經在遊戲領域打下了基礎。因此，我們擁有足夠的遊戲業務、足夠的受眾和數據，軟體業務可以正常運作。

And obviously, as you've seen this year, the exceptional growth of the software business proves that. And then in the last year, we announced 2 material software acquisitions, which we think strengthens our current offering. And obviously, we're in the middle of integrating those. So that will take us forward on an organic basis for quite some time on current initiatives.

顯然，正如大家今年所見，軟體業務的出色成長證明了這一點。去年，我們宣布了兩項重要的軟體收購，我們認為這增強了我們現有的產品線。目前，我們正在整合這些業務。這將使我們在相當長的一段時間內，在現有計劃的基礎上保持有機增長。

Then when we look at new initiatives, what we really think about is, is there a possible acquisition to go do to accelerate our ability to extend our platform, our audience and our data to something new in those categories that I talked about in my script. And then separately, when we look at those new categories, we wonder should we build it. And so that build-versus-buy decision is always going to be one that's top of mind.

當我們考慮新措施時，我們真正考慮的是，是否有可能進行收購，以加速我們擴展平台、受眾和數據的能力，使其在我在文稿中提到的那些領域中擁有新的發展。此外，當我們考慮這些新的領域時，我們也會思考是否應該自行發展。因此，自行開發或收購始終是我們最需要考慮的問題。

But I would say what's strategic for us now is a lot different than what's been strategic for us the last few years as we went through a whole bunch of acquisitions to get that data intact to power the software business to become what it has as a market-leading platform.

但我想說，我們現在的策略與過去幾年的策略大不相同，因為我們經歷了一系列收購，以獲得完整的數據，為軟體業務提供動力，使其成為市場領先的平台。

Okay. So enough 1P data, but maybe something on the NFT or blockchain side from an M&A standpoint. That's helpful.

好的。 1P 數據已經足夠了，但從併購的角度來看，或許可以談談 NFT 或區塊鏈方面的內容。這很有幫助。

Our next question is from Clark Lampen with BTIG.

下一個問題來自 BTIG 的 Clark Lampen。

I've got one on the software business. As we were thinking about, I guess, software segment performance, not only in '22 but years past that, in '23 and beyond, can you remind us -- you may have just answered this, Adam, but how does the gaming portfolio sort of affect software growth in concert with one another? Does the incremental swing, whether it's positive or negative with the gaming business, meaningfully impact your ability to continue growing or achieving those out-year targets?

我有一個關於軟體業務的問題。我們之前一直在思考軟體部門的表現，不僅是在2022年，還包括之後幾年、2023年及以後的表現。亞當，你可能剛剛回答了這個問題，但遊戲業務組合是如何協同影響軟體成長的呢？遊戲業務的增量波動，無論對遊戲業務是正面還是負面，是否會對你們繼續成長或實現未來幾年目標的能力產生重大影響？

And then separately on the margin outlook for '22, excuse me for asking you a dense question, but I just want to confirm that the $200 million of migration costs that's associated with MoPub is embedded within that high-20s margin outlook? Is that correct or incorrect?

然後，關於2022年的利潤率前景，請原諒我問了一個比較深奧的問題，我只是想確認一下，與MoPub相關的2億美元遷移成本是否包含在20%以上的利潤率前景中？這是正確的還是錯誤的？

I'll let Herald answer part 2 maybe first because I think it's quick, and then I'll answer part 1.

我會讓 Herald 先回答第二部分，因為我覺得這樣很快，然後我再回答第一部分。

We assume we're going to add back a substantial part of that $200 million to adjusted EBITDA, and therefore, that is excluded in the guidance at the high-20s margin.

我們假設我們將把這 2 億美元中的很大一部分加回到調整後的 EBITDA 中，因此，這部分金額被排除在 20% 以上的利潤率指引之外。

Yes. On the first part, really, what you're asking about is like how important is gaming and gaming growth to the software growth. And what -- really, what we knew going in, as we built out this machine learning system is, these types of software solutions are built on having really good data and then being able to extend that data into predictive lookalikes across the whole audience. And we've got 200 million monthly active users and 2 billion that we see across the platform, and that ratio is working really well for us today. Longer term, and we've said this repeatedly over the last year, we're not a games company. We're a software company really helping our partners grow. And as you look at our software revenue, it's expanded dramatically.

是的。關於第一部分，你實際上想問的是遊戲和遊戲成長對軟體成長有多重要。我們在建立這個機器學習系統時就知道，這類軟體解決方案建立在擁有真正優質數據的基礎上，然後能夠將這些數據擴展到整個受眾群體的預測性相似用戶。我們擁有2億月活躍用戶，整個平台擁有20億用戶，這個比例對我們來說非常有效。從長遠來看，我們在過去一年反覆強調，我們不是一家遊戲公司。我們是一家致力於幫助合作夥伴成長的軟體公司。從我們的軟體收入來看，它成長迅猛。

On the game side, the vast majority of our user acquisition spend, which is the core driver of gaming growth, is on our own platform. But as we continue to expand our relationships on the software side and really get more dollars from third parties, we're going to take the third-party reported revenue number and converge it with that TSTV number because we expect our software platform to be able to grow much faster than our desire to spend on user acquisition on the Games portfolio because we already have the data that we need to power the software platform.

在遊戲方面，我們絕大部分的用戶獲取支出（遊戲成長的核心驅動力）都投入了我們自己的平台。但隨著我們繼續拓展軟體方面的關係，並從第三方獲得更多收入，我們將把第三方報告的收入數字與 TSTV 數據進行合併，因為我們預計軟體平台的成長速度將遠遠超過我們在遊戲產品組合上用戶獲取支出的預期，因為我們已經擁有了驅動軟體平台所需的數據。

So you'll see our focus is going to be to continue to expand the dollars that third parties are paying us on the software side, whether it's through MAX, whether it's through the demand side relationships that we're going to inherit through MoPub, or the vast majority of the revenue, the direct relationships with advertisers buying on a performance basis on our platform.

因此，你會看到我們的重點是繼續擴大第三方在軟體方面向我們支付的資金，無論是透過 MAX，還是透過我們將透過 MoPub 繼承的需求方關係，還是絕大多數收入，即與在我們的平台上按績效購買的廣告商的直接關係。

Got it. And Adam, is there a vertical sort of adjacent to gaming that you guys would sort of like to attack next? I guess, do you feel comfortable that you have a solid business and a solid assortment of first-party data there? Is there something that, I guess, just sort of is at the top of your focus?

明白了。亞當，你們接下來想進軍的遊戲垂直領域有哪些？你們對遊戲業務和豐富的第一方數據感到滿意嗎？你們目前最關注的領域是什麼？

We've talked about this before. The nice thing about gaming data is that it expands to other markets. And we're seeing nongaming as a fast growth vector in our business, and we've seen it over the last year. And if you think about the data -- what data do we collect today, what is the first-party data? Well, if someone pays $50 in the last week in a game like Project Makeover, that tells you a lot about that person, but they're not only playing Project Makeover and they're not only playing games, they're doing a lot more than that.

我們之前討論過這個問題。遊戲數據的好處是它可以擴展到其他市場。而且，我們看到非遊戲領域是我們業務的快速成長方向，這一點在過去一年中已經有所體現。想想數據——我們今天收集了哪些數據？哪些是第一方資料？例如，如果有人在上週為像“改造計劃”這樣的遊戲支付了 50 美元，這就能告訴我們很多關於這個人的信息，但他們不僅僅是在玩“改造計劃”，也不僅僅是在玩遊戲，他們做的遠不止這些。

They might be in the shopping apps. They might be in the fashion design app. They might be in the home design app, food delivery, et cetera. And our systems can figure that out with the data sets that we have. So expansion across vertical to other app categories is something that we've been excited about, we're seeing traction with and we continue to be excited about going forward.

它們可能存在於購物應用中，也可能存在於時裝設計應用中，也可能存在於家居設計應用、外帶應用程式等等。我們的系統可以利用現有的資料集來識別這些應用。因此，我們一直對向其他應用程式類別的垂直擴展感到興奮，我們也看到了成長的勢頭，並且對未來的發展充滿期待。

Our next question is from Vasily Karasyov with Cannonball Research.

我們的下一個問題來自 Cannonball Research 的 Vasily Karasyov。

I had a question about MAX and all the new publishers that's coming on board. So would it be fair to say that there are some publishers that only use mediation and there are some that use mediation and real-time bidding to monetize their inventory? So my question is, number one, do you try to steer publishes towards one or the other option? And if yes or no, why?

我有一個關於 MAX 以及所有新加入的出版商的問題。那麼，是否可以這樣說，有些出版商只使用聚合，而有些則同時使用聚合和即時競價來變現他們的庫存？所以我的問題是，第一，您是否會嘗試引導出版商選擇其中一種方案？如果是或否，為什麼？

And then my second question is about the take rate. I understand that your take -- you only have the take rate for the real-time bidding auctions, but how do you see that going forward? Do you expect it to be stable? Do you think it can go up? Are you prepared to seek a contraction in there? So I would appreciate your thoughts on this.

我的第二個問題是關於佣金率。我理解您目前只掌握了即時競價拍賣的佣金率，但您如何看待未來的發展？您預計它會保持穩定嗎？您認為它會上升嗎？您是否準備尋求佣金率的收縮？所以，我很想聽聽您對此的看法。

Yes. So a couple of things embedded in there. When we launched MAX, we did it because we wanted to facilitate real-time bidding to the marketplace quicker. And we -- fundamentally, we believe that the ad ecosystem should be traded in real time. It's no different than a stock trading floor that's gone programmatic over the last 20 years. And when it becomes real time, it enables companies with a data advantage and a technology advantage to win their share of inventory and make whatever margins they deserve to make, but it maximizes truly the dollars that the publisher earns.

是的。這裡面嵌入了一些東西。我們推出MAX，是因為我們希望能夠更快地將即時競價引入市場。從根本上來說，我們相信廣告生態系統應該是即時交易。這與過去20年程式化的股票交易大廳沒什麼不同。當它變成即時交易時，它能讓擁有數據優勢和技術優勢的公司贏得他們的庫存份額，並獲得他們應得的利潤，但這實際上最大化了出版商的收入。

So today, MAX is the furthest ahead in the marketplace, delivering that real-time bidding to the publisher and to the end consumer to get that best ad out there on the device. It is a hybrid solution. It's bidding plus mediation. We get paid when companies bid on MAX. They pay for a seat on our marketplace. We don't with mediation. So over time, MAX' economics continue to improve.

如今，MAX 已在市場上遙遙領先，它為發布商和終端消費者提供即時競價，從而在設備上投放最佳廣告。這是一種混合解決方案，即競價加中介。當公司在 MAX 上競價時，我們會獲得報酬。他們為在我們的市場上佔有一席之地而付費。我們無需支付中介費用。因此，隨著時間的推移，MAX 的經濟效益將持續改善。

Now we wanted to facilitate bidding. We are a bidder on MAX. We've been a bidder in the marketplace for a couple of years. We're very good at it. Obviously, you've seen the growth of our software business expand dramatically, but we don't target a specific take rate from any of the ads we serve. If we can make $0.01 on the dollar, we're making an extra dollar -- an extra $0.01 of revenue, and we're reporting to you on a net revenue basis.

現在我們想促進競價。我們是MAX的競價者。我們已經在這個市場競價好幾年，非常擅長競價了。顯然，您已經看到了我們軟體業務的大幅成長，但我們對任何投放的廣告都沒有特定的佣金率目標。如果我們能從每一美元中賺取0.01美元，我們就能額外賺取一美元——額外的0.01美元收入，而且我們以淨收入為基礎向您報告。

This is one thing that's important to understand. When we talk about our revenue this past quarter or even when we're talking about $2 billion of software revenue in '23, that's net revenue reported software revenue. So it's exceptionally high margin. The costs to the publisher have already been taken out of the dollars that the advertiser pay us. So really, any sort of constriction on take rate doesn't impact at all what we're going to report to you because we're not operating afraid of that today. And we're seeing continued growth in that software business, and we're very confident about the continued growth of it going into '22 and '23.

這一點很重要，需要理解。當我們談論上個季度的收入，或甚至當我們談論2023年20億美元的軟體收入時，這都是報告的淨收入。因此，它的利潤率非常高。出版商的成本已經從廣告商支付給我們的金額中扣除了。所以，任何形式的佣金率限制都不會影響我們向你們報告的業績，因為我們現在營運時並不擔心這一點。我們看到軟體業務持續成長，我們對它在2022年和2023年的持續成長非常有信心。

Our next question is from Ralph Schackart with William Blair.

我們的下一個問題來自威廉布萊爾的拉爾夫沙卡特。

On the App side of the business, as it relates to the 2022 guide of, I think, $2.2 billion, $2.35 billion in revs you talked about modest revenue contribution from new content, and I think more so contribution in second half of the year. But any color you could add or help us think about how the relative growth rate of the consumer apps might compare different versus the business apps? Was the conservatism more on the consumer side or is it more broad-based?

關於App業務，就2022年的營收預期（我認為是22億美元到23.5億美元）而言，您提到了新內容帶來的適度收入貢獻，我認為下半年的貢獻會更大。您能否補充一些細節，或是幫助我們思考消費者App的相對成長率與商業App的相對成長率有何不同？這種保守主義是更多地體現在消費者方面，還是更廣泛的層面？

Yes. So we -- go ahead, Herald. Do you want to jump in?

是的。那麼，《先驅報》請繼續。你想插話嗎？

Go ahead, Adam. Go ahead.

來吧，亞當。來吧。

I was going to say, like, we don't operate the games as business apps or consumer apps. They're just games and apps, and they monetize with IP plus ads. So a lot of these products are hybrid. They're making money from advertisement to users and then some users will pay within the games for content as well.

我本來想說，我們不會把遊戲當作商業應用或消費者應用來運作。它們只是遊戲和應用，透過IP加廣告獲利。所以很多產品都是混合型的。他們透過向用戶投放廣告賺錢，然後有些用戶也會在遊戲內為內容付費。

Fundamentally, though, like -- again, we've said this a lot of times, we're not a games company. The games serve a purpose for us in our business. We've gotten them to scale $2 billion-plus business, which is great. More importantly, we're getting data first party on 200 million monthly active users. And the games and growth in the games are driven strictly by what we're spending on user acquisition. That's what drives mobile gaming businesses outside of acquisitions and M&A, which we haven't done recently in the games business, and are less inclined to do so now that we have the data that we needed.

但從根本上來說，就像──我們再說一遍，我們不是遊戲公司。遊戲在我們的業務中發揮作用。我們讓它們規模化，業務規模超過20億美元，這很棒。更重要的是，我們獲得了2億月活躍用戶的第一手數據。遊戲和遊戲的成長完全取決於我們在用戶獲取方面的支出。這就是推動行動遊戲業務發展（除了收購和併購）的動力。我們最近在遊戲業務中沒有進行收購和併購，現在我們擁有了所需的數據，也不太願意這樣做了。

So as you think about what matters to us in our business, we're talking about software, software growth. We're one of the fastest-growing software businesses on the face of the planet today, scaling this business to a couple of billion dollars next year. And that's important because, today, we report to you that TSTV number, which is a lot bigger than the software reported revenue we give you. That delta is the spend for own games on our own platform. Well, as our software business goes from where it is -- and we talked coming into this year $1 billion projected for this year, we just raised that up 40% and are now talking about $2 billion next year, as that continues to grow, that's all going to be dollars from third parties.

所以，說到我們業務中什麼對我們很重要，我們談論的是軟體，軟體成長。我們是當今全球成長最快的軟體公司之一，明年的業務規模將達到數十億美元。這很重要，因為今天我們向大家報告的TSTV數字比我們公佈的軟體收入大得多。這個增量是我們平台自有遊戲的支出。我們的軟體業務正在不斷發展——我們今年的預期收入是10億美元，我們剛剛將這個數字提高了40%，現在我們談論的是明年20億美元，隨著這個數字的持續增長，這些收入都將來自第三方。

We're not going to be able to double the amount that we're spending on our current games because they just don't justify that kind of expansion. And so you'll see the third-party number can merge with that TSTV number. And that gives us a lot of room to grow, the dollars we're collecting and the margins that we're able to make from the business, while focusing on what we care about, which is software, not necessarily the games part of our business.

我們無法將現有遊戲的支出翻倍，因為它們根本不值得進行這種擴張。所以你會看到第三方收入可以與 TSTV 收入合併。這給了我們很大的成長空間，讓我們能夠增加收入，並從業務中獲得利潤，同時讓我們專注於我們真正關心的事情，也就是軟體，而不一定是遊戲業務。

Our next question is from David Pang with Stifel.

下一個問題來自 Stifel 的 David Pang。

Just wanted to follow up on the in-app bidding question. Could you talk about what portion of the ad transactions on MAX was in-app bidding? And how should we think about that mix changing over the next few years with the addition of MoPub?

我想跟進一下應用內競價的問題。您能談談 MAX 上的廣告交易中有多少部分是應用程式內競價嗎？隨著 MoPub 的加入，我們該如何看待未來幾年這種結構的變化？

Yes. So in the past, obviously, a couple of years ago, when we launched MAX, the market had no bidding. We've gotten MAX to probably half bidding at this point, and we expect by the end of the year the vast majority will be bid realtime.

是的。幾年前，我們推出 MAX 的時候，市場上還沒有競價。現在，我們讓 MAX 的競價率大概只有一半，我們預計到今年年底，絕大多數競價都將實現即時競價。

MoPub allows us to access MoPub Demand, which is more of exchange-based demand. Real-time bidding is -- takes 2 forms. One is from header bidding networks. These are companies such as ourselves. Facebook was one of the first adopters of bidding. So think, the platforms with SDKs. And then there's another aspect of the market, which is demand-side platforms, or DSPs. They plug into exchanges to use someone else's SDK to access the audience and bid in real time as well.

MoPub 讓我們能夠存取 MoPub Demand，這更像是基於交易平台的需求。即時競價有兩種形式。一種是來自頭部競價網。這些公司就像我們一樣。 Facebook 是最早採用競價技術的公司之一。想想那些有 SDK 的平台。此外，市場還有另一個面向，即需求方平台（DSP）。它們連接到交易平台，使用其他公司的 SDK 來接觸受眾，並進行即時競價。

With MoPub, we're unifying the demand-side bidders that are -- ad networks traditionally thought of as well as the DSPs into one unified marketplace, one unified real-time auction. And so bringing those DSPs live through our access point to this massive audience on MAX plus MoPub is what we're excited about. That drove MoPub's revenue. The platform itself unified is over twice as big as MoPub ever was. And so that gives us a lot of access to growth in these demand-side partnerships as they come online.

透過 MoPub，我們將需求方競價者（傳統意義上的廣告網路）和 DSP 整合到一個統一的市場，一個統一的即時拍賣平台。因此，我們很高興能夠透過我們的存取點將這些 DSP 即時推送到 MAX 和 MoPub 上的海量受眾。這推動了 MoPub 的營收成長。統一後的平台規模是 MoPub 的兩倍以上。因此，隨著這些需求方合作夥伴的上線，這讓我們能夠更好地利用這些合作夥伴的成長機會。

Our next question is from Martin Yang with Oppenheimer & Company.

我們的下一個問題來自奧本海默公司的馬丁楊。

First question is on your natural extension on the new initiatives. Can you clarify, are you providing software solutions to the nongaming apps? Or are you thinking about publishing nongaming apps yourself? Or both?

第一個問題是關於您對新計劃的自然延伸。能否解釋一下，您是為非遊戲應用提供軟體解決方案嗎？還是您正在考慮自己發布非遊戲應用程式？或兩者兼而有之？

I don't think we spoke about nongaming apps. We're focused on building software platform technologies for these bigger markets. Connected TV is obviously one. We think performance marketing and an extension of our current demand and data into the connected TV world can be very impactful. When it comes to blockchain and NFTs, both for gaming and nongaming, if we can help facilitate the ability for users to engage with content and really take ownership over that content through our software platform and our relationships, we think we can expand the modernization opportunity in mobile apps, which is very impactful for all of our partners and ourselves.

我認為我們討論的不是非遊戲應用。我們專注於為這些更大的市場建立軟體平台技術。連網電視顯然是其中之一。我們認為，績效行銷以及將我們當前的需求和數據延伸到連網電視領域將會產生巨大的影響。說到區塊鏈和NFT，無論是遊戲還是非遊戲領域，如果我們能夠透過我們的軟體平台和關係，幫助用戶更好地參與內容互動，並真正擁有對內容的所有權，我們認為就能拓展行動應用的現代化機會，這對我們所有的合作夥伴以及我們自己都意義重大。

And then lastly, we talked about taking our marketing solutions, machine learning and demand and enabling carriers and device manufacturers to generate more from the handsets that they sell to the consumer by doing partnerships with our platform. And so those are the 3 big areas that we're really focused on, all software-centric.

最後，我們談到如何利用我們的行銷解決方案、機器學習和需求，並透過與我們的平台合作，幫助營運商和設備製造商從銷售給消費者的手機中獲得更多收益。這就是我們真正關注的三大領域，它們都以軟體為中心。

So my question is really referring back to your comment on checking the box on gaming, but there are still pretty substantial in-app marketing and advertising opportunities in nongaming apps. So would you do something similar for -- to obtain the first-party data from nongaming apps if you orient your software solution to other publishers that are not game publishers?

所以我的問題實際上是針對您關於勾選遊戲應用程式選項的評論，但非遊戲類應用程式中仍然存在相當多的應用程式內行銷和廣告機會。那麼，如果您將軟體解決方案定位於非遊戲類發行商，您會採取類似的措施來獲取非遊戲類應用程式中的第一方資料嗎？

I got you. I think the answer is no for the reason I gave you before, which is, we've got so much first-party data now. When you know someone paid the $50 in Project Makeover, you know a lot of things about them. And one of those things you know for certain, just conceptually, we know this human being, that person doesn't only play Project Makeover 24 hours a day. They do a lot of other things. If they've got $200 a month to spend in a fashion design game, they're probably spending a lot of money on fashion, e-commerce apps, food delivery apps, rideshare apps, a whole bunch of other things that heads of households tend to invest in.

我懂你的意思。我認為答案是否定的，原因我之前也說過，我們現在掌握了海量的第一方數據。當你知道有人在「改造計畫」上花了50美元時，你就能了解很多關於他們的資訊。其中之一，從概念上來說，我們肯定知道這個人，這個人不會一天24小時都玩「改造計畫」。他們還會做很多其他的事情。如果他們每個月有200美元花在時裝設計遊戲上，那麼他們很可能還會在時尚、電商應用、外帶應用、共乘應用以及家長們傾向於投資的其他很多東西上投入大量資金。

And the machine learning can figure that out and enable nongaming companies to market through our platform, too. And we've talked about the category as being a fast-growing category. It's still not a big part of our business, so we don't talk about it broken out. But it's an area that we're excited about because we've already seen success in.

機器學習可以解決這個問題，並讓非遊戲公司也能透過我們的平台進行行銷。我們之前說過，這個類別是快速成長的類別。它在我們的業務中仍然佔比不大，所以我們不會單獨討論它。但我們對這個領域感到興奮，因為我們已經看到了成功。

The other part of that, that's important is a lot of the nongaming spend that happens in mobile and then just on digital, in general, comes from demand side platforms. The big demand side platforms, omnichannel, where agencies will spend dollars through, are not an area that we had any exposure to in our business prior to MoPub. As we integrate these MoPub partnerships into our unified offering and give them access to over twice as much of the eyeballs, we expect to see a lot of dollars flow through our platform direct to the consumer from those types of advertisers as well, which would qualify as nongaming.

另一點很重要的一點是，行動領域以及數位領域非遊戲類支出的很大一部分來自需求方平台。大型需求方平台，即全通路平台，代理商會透過這些平台進行支出，但在 MoPub 之前，我們的業務從未涉足這些領域。隨著我們將這些 MoPub 合作夥伴整合到我們的統一產品中，並使其能夠接觸到兩倍以上的受眾，我們預計，這類廣告商的大量資金也將透過我們的平台直接流向消費者，這也符合非遊戲類廣告的範疇。

Got it. I have another logistics question, if I may. Is there any migration fees you expect after 1Q '22?

明白了。如果可以的話，我還有一個關於物流的問題。 2022年第一季之後預計會收取遷移費用嗎？

No, really, like -- look, our business long term may have some sort of nominal migration fees. We've had them in a nominal way in the past. The big amount in the quarter is specifically for MoPub publishers switching off of a platform that's just going to go off-line.

不，真的，就像——你看，我們的業務長期來看可能會有一些象徵性的遷移費用。我們過去也曾以像徵性的方式收取過這些費用。本季的大筆費用是專門用於 MoPub 出版商關閉即將下線的平台。

One of the big issues of the platform that goes off-line, which really doesn't happen, this is a decade-old software platform, is that in the mobile app ecosystem, you can either update an app or not update an app. And a lot of old devices are set not to update. And so a lot of these publishers stand to lose an annuity value of revenue on users that choose not to update, and then MoPub is going to go off-line. So they're no longer going to be able to advertise to those customers and you just cross off a stream of revenue.

這個平台下線的一大問題是，在行動應用生態系統中，你要么更新應用，要么不更新，畢竟這是一個有十年曆史的軟體平台，這種情況實際上很少發生。而很多舊設備都設定為不更新。因此，許多出版商可能會因為用戶選擇不更新而損失一筆年金收入，而 MoPub 也會下線。這樣一來，他們就無法再向這些用戶投放廣告，收入來源也就斷送了。

And so we're taking this fund to make that not a problem for these publishers because ultimately, we looked at the MoPub deal as access to publishers and the demand side. It was a very cost-effective deal for us, but we didn't want the publishers to lose in the middle of that transaction. So we're covering the cost of that lost annuity due to the quick shutdown of this platform. And that ends up being a very unique case related to this acquisition, which is something you just don't tend to see in technology where a huge platform goes off-line.

因此，我們利用這筆資金來避免出版商遇到這個問題，因為歸根結底，我們將 MoPub 交易視為接觸出版商和需求方的途徑。這對我們來說是一筆非常划算的交易，但我們不希望出版商在交易過程中蒙受損失。因此，我們承擔了由於該平台快速關閉而造成的年金損失。這最終成為了此次收購的一個非常獨特的案例，在科技領域，大型平台突然下線的情況並不常見。

Our next question is from Matthew Cost with Morgan Stanley.

我們的下一個問題來自摩根士丹利的馬修·科斯特。

Maybe the first one is for you, Adam. You've mentioned a couple of times, basically, if I'm interpreting it correctly, this idea of once you have the data in place and you feed it into AXON, there's a lot that you can do with it. And so I guess I just want to understand, does that mean that when you reach a certain audience side in terms of your portfolio -- audience size in terms of your portfolio of apps that, that yields enough data on an ongoing basis that you can do what you need to do with it? You don't need to keep scaling the app portfolio beyond a certain point because you maintain 200 million MAUs and have all the data that you need? I guess, what do you mean by that? And then I have a follow-up.

亞當，第一個問題可能適合你。如果我理解正確的話，你提到過幾次，基本上就是說，一旦你有了數據，並把它輸入到AXON，你就可以用它做很多事情。所以我想了解一下，這是否意味著，當你的應用程式組合達到一定的受眾規模時，你就能持續產生足夠的數據，讓你去做你想做的事情？你不需要繼續擴展應用程式組合，因為你保持著2億的月活躍用戶，並且擁有所有你需要的數據？你這是什麼意思？然後我還有一個後續問題。

Yes, that's 100% right. That's why we've been talking about for the last year. Like the games business may grow, it may not grow. It doesn't matter because it's scaled enough where it's fueling the massive growth in the software business. And so you sort of have today, 10% actual data, what we have perfect data on our own apps, and then 90% is predicted lookalikes.

是的，完全正確。這就是我們去年一直在討論的原因。例如遊戲業務可能會成長，也可能不會成長。這無關緊要，因為它的規模足夠大，能夠推動軟體業務的大幅成長。所以，今天我們掌握的數據大概是10%的實際數據，也就是我們自己應用的完美數據，其餘90%是預測的類似數據。

Being at a scale that we're at today with the technologies and data platform and machine learning and algorithms that we've built, you're obviously seeing a software platform that's growing much, much faster than any other platform in our industry. And so you can tell that 10x ratio of lookalikes to actual data works.

就我們如今的規模而言，憑藉我們建立的技術、數據平台、機器學習和演算法，你顯然會看到一個軟體平台的成長速度遠遠超過我們行業中的任何其他平台。因此，你可以看出，將相似數據與實際數據的比例提高10倍是有效的。

Now if we can get data on 2 billion mobile devices, we would take it. Actual data is always better than lookalikes, but we know that it's exceptionally hard to go scale a platform to get that much direct first-party data. There's maybe only one company in the history of the world that's gotten that much scale to consumers, right? But we've proven that at 200 million, we're very, very good on the software side, so we're excited with where we're at on the data side of this business.

現在，如果我們能獲得20億台行動裝置的數據，我們就會毫不猶豫地去做。真實的數據總是比看似的數據好，但我們知道，要擴大一個平台的規模，要獲得如此大量的直接第一方數據極其困難。世界上也許只有一家公司能將消費者規模做到如此之大，對吧？但我們已經證明，在2億用戶規模下，我們在軟體方面非常非常出色，所以我們對目前在數據業務方面所取得的進展感到非常興奮。

Great. And then maybe just one for Herald. Just thinking about like the growth in SPECs over the course of this year, I think excluding Adjust, you guys added around 40-plus -- a little over $40 million -- or sorry, not $40 million. 40 SPECs in 2Q and 3Q. And I think that, that number, if I have it right, was about 7 in 4Q. So I guess, again, excluding Adjust, what are kind of the dynamics behind SPEC adds over the course of this year?

太好了。然後也許只說一個關於 Herald 的例子。想想今年 SPEC 的成長情況，我認為不包括 Adjust，你們增加了大約 40 多個 SPEC，略高於 4000 萬美元，抱歉，不是 4000 萬美元。第二季和第三季各增加了 40 個 SPEC。如果我沒記錯的話，第四季大約增加了 7 個。所以，我想，除了 Adjust 之外，今年 SPEC 成長背後的動態是什麼樣的呢？

Yes. Thanks, Matt, for the questions. Yes, again, the software business, as you know, has grown tremendously over the course of the year, and we're projecting it to continue to do that. And we gave a SPEC number to try to give a little more granularity into the build price volume-wise. One aspect -- and of course, it's grown considerably, price volume more in the fourth quarter in terms of people scaling quite rapidly, particularly ex Adjust, to $757,000 in the quarter, so a pretty significant amount of money of increase.

是的。謝謝 Matt 的提問。是的，再次強調，如您所知，軟體業務在過去一年中取得了巨大的成長，我們預計它將繼續保持這種成長勢頭。我們給出了一個 SPEC 數字，以便更詳細地了解建立價格量。一方面——當然，第四季度的價格增長顯著，這得益於人員擴張速度相當快，尤其是在 Adjust 之外，本季的價格量達到了 75.7 萬美元，因此增長幅度相當可觀。

What was slower in that period was the actual SPEC count. And that's really due to the fact that in the third quarter, we added a lot of SPECs that we're testing and whatnot and didn't add as many of those in the fourth quarter. So we still feel very good about the customer growth, obviously, with all the migrating customers from MoPub. Again, it will be hard to distinguish who's a MAX customer, who's a MoPub customer because it's all going to be MAX customers. We're highly confident in the software line growing and, therefore, price quantity has to follow that.

那段時期增長較慢的是實際的SPEC數量。這主要是因為我們在第三季增加了許多正在測試的SPEC等等，而第四季沒有增加那麼多。因此，我們仍然對客戶成長感到非常樂觀，顯然，這要歸功於所有從MoPub遷移過來的客戶。同樣，很難區分誰是MAX客戶，誰是MoPub客戶，因為所有客戶都會是MAX客戶。我們對軟體產品線的成長充滿信心，因此價格和數量也必須隨之成長。

One footnote to that, and it is in our letter, but didn't want to highlight it. Just I wanted to keep things clear for this. In our letter, on the upfront side, we kept the SPEC numbers the same. One very detailed point is our SPEC definition was based on a run rate per quarter. They had to get to $125,000 of run rate in the quarter. Going forward, we'll continue to report that metric for a little bit, but we're really going to go focus more on an LTM basis to have $125,000 of revenue.

對此有一個腳註，它在我們的信中，但我不想強調它。我只是想把事情說清楚。在我們的信中，在預付款方面，我們保留了相同的SPEC數字。一個非常詳細的要點是，我們的SPEC定義是基於每季的運行率。他們必須在當季達到12.5萬美元的運行率。展望未來，我們將繼續報告這項指標一段時間，但我們實際上將更專注於LTM（長期銷售）的計算，即12.5萬美元的收入。

The reason I point that out is that we'll get rid of some of the volatility that is inherent if you use just a quarterly run rate number of SPECs coming and going. And given the scale of the customers we're talking about, we think that will be a much better metric going forward.

我指出這一點的原因是，如果只使用季度運行率來計算SPEC的變動，我們就能消除一些固有的波動。考慮到我們討論的客戶規模，我們認為這將是一個更好的未來指標。

Our next question is from Omar Dessouky with Bank of America.

我們的下一個問題來自美國銀行的 Omar Dessouky。

Right. How would the Open App Markets Act, as currently written, affect your business strategically? And are you playing any role, direct or indirect, in how the law is being written?

對。 《開放應用市場法案》目前的版本會對你們的業務策略產生什麼影響？你們在法案的製定過程中是否發揮了直接或間接的作用？

I will be honest, I'm reading up on what the law is right now. Can you give us a little insight into specifically what you're talking about, and then I'll answer it.

說實話，我正在研究相關法律。您能否具體解釋一下您說的是什麼，然後我再回答。

Yes. So I think it's pretty well known that major players that own platforms are able to advantage themselves in certain ways, or at least that's the view of some market participants. And it appears that legal interests are potentially going to try to make it a more competitive industry. And I'm just wondering what you guys think about it and how it might reshuffle the -- potentially the attribution ecosystem or the actual targeting ecosystem itself in your favor or not?

是的。所以我認為，擁有平台的大型企業能夠在某些方面獲得優勢，這是眾所周知的，或者至少一些市場參與者是這麼認為的。而且，法律利益似乎可能會試圖讓這個產業更具競爭力。我只是想知道你們對此有何看法，以及它是否會對歸因生態系統或實際的定位生態系統本身進行重組，使其對你們有利？

Got it. And this is -- you're talking about like the actual operating systems, the App Store tax and any other rule changes they make that could be seen as benefiting themselves? Is that what you're talking about?

明白了。您指的是實際的作業系統、應用程式商店稅以及他們做出的其他任何可能被視為有利於自己的規則變更嗎？您指的是這個嗎？

Yes. Yes.

是的。是的。

So the app store fees, obviously, they're not going up. And I think a lot of people are hopeful that alternatives can be introduced or the actual tax from the outsource goes down. Overall, we're in a good position because, one, our apps business is paying a hefty amount in the 30% every single year. But more importantly, these types of added competition to a marketplace make the pricing that goes to the end developer stronger.

所以應用程式商店的費用顯然不會上漲。我認為很多人都希望能夠推出替代方案，或是降低外包的實際稅率。總的來說，我們處於有利地位，因為首先，我們的應用程式業務每年都要支付30%的巨額稅金。更重要的是，這些市場競爭的增加使得最終開發者的定價更加合理。

And so if that happens, you sort of take -- if $1 becomes $0.70 today and say -- and tomorrow it becomes $0.85, that's 20% more LTV that advertisers can go spend on a platform like ours. So that would just immediately mean, in a competitive marketplace like ours, that you see a step function up in terms of the dollars that companies are willing to spend to go acquire new users because they have those kinds of economics that they get in their favor.

如果這種情況真的發生，比如說，如果今天1美元變成0.70美元，明天變成0.85美元，那麼廣告商在我們這樣的平台上投入的生命週期價值（LTV）就增加了20%。所以，這意味著，在我們這樣一個競爭激烈的市場中，你會看到公司願意投入的資金在獲取新用戶方面呈現階梯式成長，因為他們擁有這種經濟優勢。

When it comes to other rule changes, the platforms do provide an immense amount of value. So we operate within the rules of the platforms. And we can't predict where they go. The competition is always good for these ecosystems. And I think when we look at this, we know that we're one of the leaders in the market on the app side. And we're undoubtedly the largest advertising platform outside of the Facebooks and the Googles on mobile today that exists independent. And so we're set up really well to benefit from some of this disintermediation that might happen, added competition and better economics for the content creators.

就其他規則變化而言，平台確實提供了巨大的價值。因此，我們遵循平台的規則運作。我們無法預測這些規則的走向。競爭對這些生態系統來說總是有益的。我認為，當我們審視這一點時，我們知道我們是應用市場的領導者之一。毫無疑問，我們是當今行動領域除Facebook和Google之外最大的獨立廣告平台。因此，我們擁有良好的條件，可以從可能出現的去中介化、競爭加劇以及內容創作者獲得更好的經濟效益中獲益。

Our final question is from Franco Granda with D.A. Davidson.

我們的最後一個問題來自 D.A. Davidson 的 Franco Granda。

Just following up on an earlier question, I was hoping you could speak about your progress in nongaming revenues on your software platform. What was the growth rate of nongaming revenues in '21? And then now that MoPub has closed and you're integrating it, what does that mix look like?

繼續之前的問題，希望你們能談談你們軟體平台非博彩收入的進展。 2021年非博彩收入的成長率是多少？現在MoPub已經關閉，你們正在整合它，你們的收入結構是什麼樣的？

Yes. That's not something that we break out yet. But I'll say really the area that's a big opportunity is both the direct app market on our platform, getting access to more of them through the Adjust sales force, bringing more nongaming customers into our ecosystem, as we've touched on. But more importantly, something we haven't talked on that I referenced earlier is, you end up with these omnichannel DSPs. Trade Desk, obviously, being a famous one, but there's very big companies here, that route nongaming demand, specifically agency and brand dollars, into mobile applications.

是的。這方面我們目前還沒有突破。但我想說，真正蘊藏巨大機會的領域是我們平台上的直接應用市場，透過Adjust的銷售團隊接觸更多應用，並將更多非遊戲客戶引入我們的生態系統，正如我們之前提到的。但更重要的是，我之前提到過，我們還沒有談到，最終你會得到這些全通路DSP。 Trade Desk顯然是其中一家很有名的公司，但這裡也有一些非常大的公司，它們將非遊戲需求，尤其是代理商和品牌的資金，引導到行動應用中。

With the MAXs plus MoPub unified platform, we've got the biggest supply side platform that the mobile app ecosystem has ever seen in one place. We're the direct access to the consumer. And these agencies and those omnichannel DSPs are really going through this shift in mindset to try to optimize the supply path so they get direct to the consumer. Well, they're going to be able to do it through our platform at massive amounts of scale now as they go and integrate into the MAX marketplace directly through us. And we think that's going to facilitate a lot of dollars in nongaming to move into our ecosystem, which will be to the benefit of our publishing partners, but also to the benefit of us as the trading floor.

憑藉 MAXs 和 MoPub 的統一平台，我們擁有了行動應用生態系統迄今為止最大的供應方平台。我們是直接接觸消費者的管道。這些代理商和全通路 DSP 正在經歷思維轉變，試圖優化供應路徑，以便直接面向消費者。現在，他們將能夠透過我們的平台大規模地實現這一目標，並透過我們直接融入 MAX 市場。我們認為，這將促進大量非遊戲領域的資金進入我們的生態系統，這不僅有利於我們的出版合作夥伴，也有利於我們作為交易平台。

All right. And then as it relates to some of the new opportunities in the software side that you talked about, such as NFTs and CTV, I was curious about your approach there. Are you approaching talent from their respective industries? Or are you moving resources from within the company to help you build out those strategies?

好的。然後，關於您提到的軟體領域的一些新機遇，例如 NFT 和 CTV，我很好奇您在這方面的做法。您是否會接觸各自產業的人才？或者您會從公司內部調動資源來幫助您制定這些策略？

Yes. I mean, look, when we look at a new business, we're actually -- all of our core team has been here for -- since we started the business a decade ago. And so we're very entrepreneurial. And the best new businesses always start with a few people that put together an idea. And if it sticks and it hits, it can really expand. Now that's a traditional startup. We have advantages with our scale data platform, our algorithms and the demand access and data access that we have. And so we picked categories where our relationships and our technologies will give us an advantaged head start.

是的。我的意思是，當我們考慮一項新業務時，我們的核心團隊實際上已經在這裡待了十年了，從我們十年前創業開始就一直在這裡。所以我們非常有創業精神。最好的新業務總是由幾個人共同構思一個想法開始。如果這個想法能夠堅持下去並取得成功，它就能真正發展壯大。這是一家傳統的新創企業。我們的優勢在於我們的規模資料平台、我們的演算法以及我們擁有的需求和資料存取能力。因此，我們選擇了一些能夠利用我們的關係和技術優勢的領域，這些領域能夠讓我們佔據先機。

We're entrepreneurial, so we're building a lot of the core technologies ourselves, but we will look to buy potentially if it makes sense. We'll also look to staff from the outside-in. And in particular, with a couple of these businesses, we brought people from the outside to come in and run these businesses for us to give us even more confidence in executing them over the coming years.

我們具有創業精神，因此許多核心技術都是我們自己開發的，但如果條件允許，我們也會考慮收購。我們也會從外部招募人才。特別是對於其中幾項業務，我們引入了外部人才來運營，這讓我們對未來幾年的業務發展更有信心。

This does conclude our conference. Thank you for your participation. You may disconnect your lines at this time.

我們的會議到此結束。感謝您的參與。現在您可以斷開線路了。

Thanks, everyone.

謝謝大家。