Find here China Paradigm episode 109. Today we talk to Alex Li and Alex Duncan, co-founders of KAWO about the Chinese SaaS industry. KAWO is a leading social media management platform for international brands and agencies in China, providing them with data insights and finding social media solutions for their clients to help them grow in China’s rapidly changing market.
Full transcript below:
Hello everyone, I’m Matthieu David the founded of Daxue Consulting and its podcast China Paradigm and today our guests are Alex Duncan and Alex Li, co-founders of KAWO. KAWO is an enterprise social media software which helps companies to manage multiple accounts, plan and report social media on WeChat, Weibo and Douyin in China (Learn more about Daxue’s digital strategy services). The company has 10-15 employees and have served clients like Airbnb, Bacardi, Filo, VR1 and NFL, and many others – specifically strong in tourism in New York City and with a soccer club.
You have initially started the company to extract and localize content from the existing social media in the West and then publish in China. You provided an easy way to publish, report, plan content in China and have metrics on it, which is unique.
KAWO is six-year-old venture-backed at WeChat, Weibo and Douyin, a management platform headquarters in Shanghai, China. Originally a side project of The Mailman Group, like many other SaaS companies. It’s a bit like App Annie, a company with billion-dollar valuation, started from the urgency and then became independent as a software.
Alex Duncan: It’s the same founding story of Hootsuite and Basecamp as well from 37 Signals. they wanted to manage their project and they eventually made their software. So, I think it’s a great way to start a software company in the Chinese SaaS industry (Know more about how SaaS industry applied nudge marketing in China).
Matthieu David: It’s very difficult to move from service to software.
Alex Duncan: One thing we were quite lucky – sorry, to take you on that point is – the team that started building KAWO always built it with the vision to be a SaaS company, so we came at it from that perspective from the very early days.
Matthieu David: True, you have to have an independent team within this agency to be able to make it work.
Alex Duncan: Exactly, KAWO was a separate team right from the very beginning and it was built around – we sat just meters away from our early users but our vision was always how can we scale this thing to be a ubiquitous software platform.
Matthieu David: Let me continue reading your page – Back in 2013, many Western brands were just starting to develop a presence in China and social media was a very cost-effective way to begin( Know more about Daxue’s China Market Entry Solutions). Brands were understandably still nervous about a country with a very different language and culture. We thought we could use technology to make social media in China more transparent, simple and affordable. The first version of KAWO pulled in posts from a brands Facebook and Twitter accounts and made it easy to translate and publish them on Sina Weibo, Renren and Tencent Weibo.
We soon realised that translation alone wasn’t enough; brands needed localisation to take account of the cultural differences in China, so we pivoted KAWO to become a management platform helping brands and their global teams to work effectively “together”, which I think that’s a key element when we understand your platform, to collaborate.
As RenRen and Tencent Weibo become less popular, we stopped supporting them and focused our efforts on Weibo, WeChat – which replaced Tencent Weibo in some way and later Douyin.
KAWO has since grown to become the platform of choice for global brands building a social presence in China. Every week, thousands of posts are published to millions of followers for some of the top brands in the world.
When Kobe announced his retirement in New York City and New York City launched their ‘Love Campaign’, those posts were scheduled, approved and published through KAWO.
We’re fundamentally a tech company. Our team is passionate about code, design and user experience. We spend countless hours talking to users and analysing their usage data trying to understand what we can do to make their work easier and the result better. I think its key to understand your mindset: You are a tech company and you are communicating with your users often to implement new features.
Alex Duncan: There is this fantastic quote from Satya Nadella of Microsoft, and he talked about the mission of Microsoft is to meet the unmet and unanticipated needs of their users and I think that’s a beautiful way to put it because we try and solve problems that people themselves don’t necessarily know that they have, but through our research and our understanding of their roles, we observe that we can help with. I think that’s quite a common thing when we go to agencies we look at the way they’re working and we see that they’re doing things – we’re like wow, we could make that a lot more efficient because our expertise is in what can be done with software and code and I think it’s a beautiful way to look at the product.
Matthieu David: Totally, I saw how you interact with the users, implementing their suggestions into your software. To finalise the reading of your page– we’ve held developer meetups at our offices, open-sourced some of our code and tried to find ways to give back to the community that has helped us get where we are today.
Every week we push new updates to KAWO and have ambitious development plans for the future.
Look like very Silicon Valley thinking –open-sourcing part of the code, having meetups, change the world and the world community – so the features you added step by step you started with translation as we know – Weibo scheduling, so scale a post on Weibo, WeChat when you want to publish it, and then making it possible for people to collaborate on the same post because you may want your PR people or company to validate the post before publishing and this is not possible with current features of WeChat or Weibo. You’re making it possible to tag, to add an image gallery and so on, forecast as well as engagement. Very importantly is that to track results, you do it through the software while others do it manually.
Could you tell us about the size of the company and the number of clients?
Alex Duncan: I’ll let Alex handle the business side but in terms of where the products at – one of the great things about building software is that it’s kind of an interactive process where we gradually over the years build the tools to build the thing and so now we feel like we can work faster and faster because everything we’re implementing is sitting on top of six years of development work, so KAWO gets updated roughly twice every single week and we’re able to use all of these pieces we’ve built in the past, to very quickly implement new features. When we added the new network, Douyin instantly took advantage of all of the stacks to display and all of the structures that we’d already put in place for WeChat and Weibo and when we add other networks in the future, they will also sit on top of all this infrastructure we’ve built, and I think we’ve also built a culture as well of doing things to a very high standard and not trying to do too much, but trying to do what we do well. As an exam, in the early days when we were first publishing just to Weibo, many things could go wrong. When you press that red publish button, sometimes one of the images wouldn’t upload or sometimes the post wouldn’t get published and for a user that’s a frustrating experience, and they would not appreciate how many different things we were doing to try and make that happen successfully, so for us, robustness has always been a feature. We deal with very big enterprise clients who pay quite considerable fees and they just need it to work. They don’t need it to have the fanciest most cutting edge features necessarily, although a lot of people seem to think they need them, what they need it to do is to handle the everyday management of social media, the scheduling, the publishing, the commenting, the reporting and just work. So, that’s probably been an area where we’re a little bit different from some other local software companies in China’s SaaS industry.
We’ve not gotten too broad, we have a very clear area of focus and we try to be just strong. We’ve probably rebuilt almost every part of our product two or three times trying to make it better and better. We’re about to build version six of our Weibo editor – we’ve always been able to publish Weibo posts but we constantly see ways we can improve it, and so that maybe gives you an idea of our philosophy there.
When it comes to business, Alex can talk more about how we’re doing there.
Alex Li: Well, right now we work with over 500 global brands, I think it’s a mix of three different types of clients, some clients are completely based outside of China – we sign directly with them with a global contract. Some of them we might deal with both HQ as well as their China team, so a lot of brands already have the China presence, they have a team based in China so then we deal with both and other times we deal with completely local companies and that is one area that we’re trying to improve but like many people say, in China SaaS is still a new concept. I think a lot of enterprises are starting to realise the potential of the Chinese SaaS industry(Learn more about China’s cloud computing) – because I think again, it’s a lot of habits. I think this time with the coronavirus, a lot of companies are starting to realise- hey you know, this works. Yes, it’s a little bit different but it works. So, I think in some ways it has helped all software companies when it comes to enterprise sales. In terms of industries we are very diverse, we currently work with over 12 different types of industries, from industrial companies to chemical companies to governmental agencies, luxury, FMCG, travel destinations, sports, yes – can’t stop. We have a lot of sports brands, so in terms of industry work, we’re very diverse. Every brand needs to do social and it doesn’t matter what specific industry they’re in.
Today is the 21st of May 2020 and we’re still in a world with an epidemic, how has it impacted yourself and your clients?
Alex Duncan: I think we have been generally incredibly lucky. We’ve had a few of our clients particularly in tourism who have suffered and we feel like we’re in a fortunate position where we’re able to try and be as generous as we can. We can pause their accounts, give them extra months on their contract, give them a holiday on their billing – so we’ve tried to do as much of that as we can, but then there’s a lot of our other clients who are seeing that China is the most likely to rebound – 79% of Chinese consumers were surveyed said they were optimistic about the recovery of the company within the next 12 months compared to only 25% in Europe, so all of these brands are looking at this market and they’re going – we need to really focus here and particularly luxury brands who saw a lot of their sales outside of China when people traveled, they’re now looking at how can we strengthen ourselves domestically. So, we’ve been quite lucky in that respect. We also have been aware that a lot of teams have cut their staff by a quarter, a third or as much as half and so they’re starting to look at how can we deliver more efficiencies. The agency landscape is facing quite a tough time, we likely see a little bit of a shakeout, where some agencies probably won’t make it through and the ones that do survive will need to be leaner, meaner and more efficient and more competitive and KAWO undoubtedly gives you an edge. Clients want to be involved in the process and KAWO gives you that unique ace in your sales deck where you can say – not only are we going to deliver great results for you, but you’re going to see what’s happening, you’re going to be part of the process, we want to collaborate with you.
How did the company start and did you have to take external funding to start the company and to develop the product – was it already developed within the agency?
Alex Duncan: Yeah, I can probably talk about the early days. We did need funding early on but our goal was to as quickly as we could get to a product that people were willing to pay for.
Did you get funding?
Alex Duncan: Yes. So obviously like – the big challenge with a SaaS product in the Chinese SaaS industry compared to an agency – with an agency you can walk into that first pitch meeting and you don’t have to have done any of the work for that contract. When we walk into a meeting and promise people that they can sign up for our SaaS product next week – the product already has to exist and it was probably at least a couple of million US dollars of development before the product was ready to sell.
So, you raised a couple of million?
Alex Duncan: Yes it’s been around that amount. So there has been a huge amount of investment to get KAWO to the point where it was saleable. Some of that money came from early customers paying for an inferior product as well and sort of sticking with us.
Alex Li: I think we have had small amounts of investment in KAWO’s history, especially in the early history but as we’ve seen in recent times, we’ve been very fortunate that we are in a space where people are willing to pay for a good product and in the last year we’ve been bootstrapping pretty much all of our operating cost and we are profitable since the end of last year, so I think moving forward even without investment we can continue to grow at a very healthy rate and that is something very difficult to find in China’s SaaS industry. I think in terms of the entire investment landscape, investors are looking for companies that already have a working business model, meaning they have the capability of making a living on their own, without being the overhyped valuation and relying on capital to make sense of the business model. So KAWO has a very good business model, it’s a working business model, it’s a profitable business model and that is something that we strive to continue.
Is it two million USD investment?
Alex Duncan: Various rounds yeah, it’s hard to put an exact figure on it.
Do you include the time you spent and the funding you didn’t get or is it two million you got in the funding?
Alex Duncan: So, it was an initial seed investment of about 200,000 US dollars and then there was a small amount of debt and then there was an additional round –
Alex Li: like an angel round.
Alex Duncan: We’ve generally kept it very lean, so the first version of KAWO was built by two of our cofounders. Brian our CTO wrote the code for the servers, I wrote the code for the front end and did the design, meanwhile occasionally went to offices to pitch it to agencies who were very underwhelmed by our product at the time, and then it was probably after about 3 or 4 years that I started feeling like it was a good fit with what people were wanting.
Matthieu David: So profitable since the end of last year, so it’s about five years before getting profitable and now you are profitable. Your SaaS pricing to give a sense to people listening to us is about 350 USD per month for the first –
Alex Li: Standard product.
Matthieu David: Standard right, and 850 USD for the premium. You mentioned Core and Core Plus, but you can bill annually with a discount, basically two months for free roughly if we pay annually. It’s about 3,500 USD for one year and 8,670 for one year
How did you come up with those numbers for software pricing in China? Did you make a lot of projections?
Alex Duncan: Oh my god! I would say pricing in the Chinese SaaS industry is one of the hardest problems because you’ll be aware(Learn more about how to effectively apply the pricing strategy in China) – in your agency business there a very direct cost associated with everything you do for a client, but we have already built it – like when you pay for KAWO right now as a client, you’re paying for everything we’ve already done, so at some point, we had to make a decision – do we sell this to 10,000 companies at a 100$ a month or do we sell it to 500 or a 1,000 companies at 500 or 1000USD a month. And we looked at the Chinese SaaS industry, which is still incredibly immature in China, at that time all of our analysis told us that the enterprise and companies were the better way to go. The sales cycle for sure are longer, but you don’t have this churn on the distraction of free trials which we tried earlier on and we learnt they just didn’t work. Companies to use KAWO have to invest in it, they have to sign a contract, those contracts are generally multi-year contracts – they generally pay upfront, except in agencies where we’re a bit more generous, we do post-paid to support their cash flow, and it’s worked for us.
Alex Li: I think we just have to consider all the costs for software pricing in China, there are technology costs, there are development costs, there are training costs, sales costs – there are all these costs that are involved in the enterprise. I think a lot of times, especially in China – people think – oh you guys are software – it must be free, but that’s just not true. There is a lot of money – like real money going into developing, servicing and providing the service, that needs to be accounted for.
Did you change pricing from the start?
Alex Duncan: Yeah, we constantly sort of adjusted and changed pricing. It’s been around that point for about three years now, I think.
Matthieu David: I remember Bill Gates saying he was very happy to talk to Warren Buffet because he was the only one to ask the question – how did you set up your price? Specifically, in a software business, because marginal costs on having 1,000 to 100 people are not that high, you need to provide service but still, it’s a software so it’s purely scalable. So, people who are not in the software business don’t realise how the topic of software pricing in China is the key and requires a lot of thinking and research.
Alex Duncan: We still struggle with it, because in this way implementing new features – some of the things we have coming in the next couple of months they’re going to have very high data costs associated with them, so we’re going to have to rethink our software pricing in China and we’re going to come up with a new pricing model later this year because we’re going to be doing a product for core community management. Now if you are an incredibly big account, we’re going to be pulling in 100’s and 1000’s of comments through your account, but if your account only has a few thousand fans we’re going to pull in tens of comments, so again we’re going to have to adapt our pricing model in China’s SaaS industry.
Matthieu David: I think your pricing is very transparent on your website-350 USD per month and I feel that it could be someone is spanning by writing down all the resolves, the reporting on the WeChat results.
Is it a way you were thinking about the costs of doing it manually on excel and then to say – because we saved you this amount of time, we charge this amount of money?
Alex Duncan: In the past, it’s been a little bit hard to draw that comparison, and you’ll know this probably quite well, there was this perception in China that labour was very cheap and so there is a fantastic quote by an American software entrepreneur and he said the business model for SaaS only becomes viable when GDP per capital passes 10,000USD. Now, do you know when GDP per capital passed 10,000$ in China? In December 2019.
Matthieu David: Interesting and you’re profitable since the end of last year, you have very good storytelling.
Alex Duncan: So we believe the age for the Chinese SaaS industry is starting to come, one of the huge things we’ve seen during the COVID period when there’s a lot more remote working, we’ve seen all of these companies – Ding Ding, WeChat Work, Feishu, Pinduoduo just launched their internal platform called Knock. There’s a lot of companies coming out with enterprise software packages. So we think that is going to be a big opportunity. I don’t know if you’ve ever read Marc Andreessen’s essay on SaaS about how software is eating the world, It’s a very famous essay and it’s basically how the SaaS business model is the closest thing we’ve all had to grow a money tree.
Because it’s very difficult to change the price for existing clients, how have you managed the software pricing in China in the past and what’s your plan in the future? Are you going to add options and people can opt in for those options?
Alex Duncan: We’re still debating that at the moment, whether or not we – cause we’ve seen a lot of other companies doing this, so, for example, GitHub who is a code hosting company, what they did was they kept all their old plans and if you run all their old plans you could continue, but nobody could sign up for those plans and then they brought out a load of new plans with new names that you could sign up for – so we may do something like that to reduce confusion when we switch the pricing models, so old customers are not suddenly getting a big increase but then new customers are paying a new model.
Alex Li: I think it’s going to be a – what we plan on doing is put more of a mix, so it’s kind of being – some of the continuity of the old model plus a little bit more fluidity with the new upgradeable functions with additional costs obviously because when they were working with the old contract or the old pricing they didn’t have these new functions and these new functions will come up with a different price. And, additionally, we, later on, have new product ideas, it could perhaps be added as a new product but not on top of the existing price structure of the existing product. We see that quite often with other software companies as well, sometimes social listing is sold in one product and then half the social management is sold in a separate product and they’re priced differently, and the way that they’re priced is also different. But to the same clients, so that’s probably a better way to kind of make it into different products and not all in one big product.
Alex Duncan: We ultimately, we want people to feel like KAWO’s great value, so we don’t want that to be in question.
Alex Li: I think one really good point about that is a lot of software companies charge people by seat, say how many people are using the software – the more people that use it, the more we’re going to charge you. But KAWO’s whole philosophy is about transparency and collaboration, so we just don’t do that and we know that every single other social media platform in the West does this and we don’t. We want you to add as many of your team on board as possible, because through us – it doesn’t add that much extra cost, but it accomplishes our mission.
Alex Duncan: Yeah, so if you have somebody who only logs in once a month – are you going to pay for a seat for them? Probably not, but we want them to be part of the plan.
Can you describe multiple accounts? Is it because you can plug different platforms or you have different users? how does it work?
Alex Li: Multiple accounts just means that you have multiple brands. So, every brand you can have multiple platforms, platforms as in WeChat, Weibo, Douyin but what we see is like – when we’re working with a client like the NFL – let’s say they have 34 brands on KAWO, so that’s called multiple accounts. So, they have 34 accounts inside KAWO.
Alex Duncan: 32 NFL teams, the main NFL account and then they have an NFL player as well.
Does this mean that they have 30+ WeChat accounts?
Alex Duncan: 34 Weibo accounts, I think two WeChat accounts, the Patriots and the NFL have WeChat accounts and then they just added Douyin as well so –
Do you charge the same price whether they have 10 or 30 or 50?
Alex Li: No, we charge based on how many brands they have, so they have 34 times let’s say 350.
Alex Duncan: You’ve noticed our pricing page, there are four columns, there is standard pricing, so if you just have one or two accounts you pay standard pricing. As soon as your needs get more complex like the NFL – we sit down, we understand – how big is your team, how are you using it and then we give you a discounted rate on the main packages.
Matthieu David: Enterprise package.
Alex Duncan: Yes, and then the big difference with the agency is agencies obviously add and remove clients regularly, so you can’t pay upfront because you don’t know what’s going to happen. So, we bill you quarterly at the end of your quarter for what you’ve used during that time. So, if a client leaves – you just turn off their account.
Alex Li: Imagine, Nike has many accounts. Many WeChat, many Weibo accounts, because they’re separated into categories internally. So, they are all making their decisions.
Alex Duncan: Managed by different agencies.
What you are providing is a one-stop software to manage all these multiple accounts simultaneously and easily. That’s the value provided by the feature multiple accounts – is that correct?
Alex Duncan: Yes, it is your right from the perspective of users as well. So, I don’t know whether or not you have any social retainer clients but probably in your position, you don’t have the Weibo username password or the WeChat login for that client. Now with KAWO, you would be able to just login, because your account would be connected to KAWO, so everybody in your team would be able to see what’s going on. You can also invite the clients in. so it’s multiple users and it’s multiple social channels.
Matthieu David: I thought this covered another feature in your software which is a bit like Last Bass – being able to share your password.
Alex Duncan: We don’t share the password.
Matthieu David: You don’t share the password but you make it possible for all the people to log in and manage the account without sharing the password, which is exactly what Last Bass is doing.
Alex Duncan: Well we connect to the Weibo API, so one person will do the connection, but then all of the users who you invite – either the client, your account executive, your account manager, your media manager, they can all see what’s going on. So, this is where collaboration and transparency come in.
Alex Li: Yeah and also it provides massive security guarantee because you know – you don’t want to be sharing your main platforms actual username and passwords around. With KAWO you can just share your KAWO account and we all know in China people move around quite often. I think the average is 18 months before they change jobs. So, when you have an intern just open your WeChat account for a massive corporation and that intern leaves and you can’t find this person to help you log in to the back end – what do you do? So, with KAWO it protects the brand to be able to say – you know, I’m not giving this to everybody on my team.
Matthieu David: It’s very interesting, you don’t emphasise on your website, but I think that’s very, very interesting. The Facebook page as well – all the software, social media software is leading to individual profiles, but when its managed by companies, it’s very hard and here you make it easier to manage enterprise owned content on social media.
I’d like to go to the second step which is planning and reporting. For planning, people can plan when to push a post on Weibo, WeChat and Douyin – is there something I’m missing here?
Alex Duncan: So, the way I would describe it is, if you’re coming in on Tuesday morning and creating content for Tuesday afternoon- that’s probably not going to be the best content you can publish. It’s much better to have a strategic plan for month or quarter than to have a content plan for a month ahead and then to refine that content down over that period. If you have a client who is based overseas as well – to have them come in and make comments and then to gradually iterate on that content until its ready to be published. So, this is the goal of KAWO’s content planning solution. Instead of you sending an excel sheet or a ppt to your client once a month saying this is what we’re planning for next month and they can leave comments on it – you have this live calendar that can be constantly commented on, iterated on, you can see all the changes and transparency and one the best case studies we have is with new York city – so we had Daisy who worked here in Shanghai for an agency, we had Tyser who worked for New York City in New York, born in New York, loves the city, knows it very well – and they worked together incredibly effectively. Daisy would suggest – oh let’s promote this hotel and then Tyser would comment on the post and say, actually I think there’s another thing we can promote and they collaborated to make the content for New York City some of the best, organic content of any US tourism destination.
Going back to planning, how did you plan? What things you can do with your software?
Matthieu David: I understand planning is to schedule a few posts on it, an entire quarter of posts. For example, an ad agency who set up a newsletter sending one training every day for five days, so we would know everything about posts on WeChat.
Alex Duncan: Yes, without going into all the features, we also help you be more strategic, so almost every – well certainly all the good agencies have a content plan in terms of the different pillars of their content, so maybe part of the content is education, part of the content is commercial, maybe part of the content in the case of tourism is promoting partners who fund the tourism organisation, maybe part of it is about food, maybe part of it is about activities. Also in the case of China we’ve got to remember, we’re dealing with a very diverse set of customers, who is a user who is thinking of travelling to New York City for the very first time but then leaving China is very different to somebody who studied in New York and then is thinking of going back and so being able to look at your content from a strategic perspective and saying – we need to be making 15% of our posts about this, 20% about that – so we’re helping you to plan strategically in KAWO and then we also close that loop by helping them to report strategically as well, and this is kind of going on to I think where you’re going next – the third part of our solution is reporting. One of the things that I’ve – we’ve spoken to dozens of teams over the years – when it comes to reporting and one of the things that makes me saddest is seeing what I call the reporting merry go around, where it’s just a constant cycle of – the client needs a report, let’s just make a report – we send it to the client, the client possibly doesn’t even read it, let’s get on with doing next month. There is not this closing the loop – of looking at okay – what worked in that report. Let’s segment our posts and let’s see – oh this type of content is working well, how can we repeat the success of that content – or this content is really important but it’s not performing very well, how can we improve it? So, this is where we try to build the reporting inside KAWO in line with what we think the best practices are in marketing.
So, you asked this question earlier, before we were starting you said – why do you not provide a freemium model – because although KAWO is a software company, we’re also providing consulting on how to build an effective marketing team. Our product is built with an opinion as to how great marketing works. We think great marketing works when you plan – when you don’t do things at the last minute. We think great marketing works when you look at the data and then use that to improve and inform the way that you do things in the future. We think great marketing works better when you involve your client in a deep way in what’s happening and are more transparent in the way you work. So KAWO was a software company, has an opinion about how great marketing gets done.
How did you react to the comments of people on your channel? Does the software react to comments on WeChat, Weibo or on Douyin or just to push the main messages?
Alex Duncan: That’s a great question, we have a huge software release coming in the next six weeks which will add community management to KAWO, so it’s been a long time coming and it’s been very well engineered, but we are going to give you a social inbox where you can see all of your comments, reposts, mentions, direct messages across WeChat, Weibo, Douyin all in a social inbox, you’re going to be able to search for a keyword, you’re going to be able to assign messages to different teammates and so yes, community management is coming to KAWO in the next couple of months.
How easy it is to build a product on the environment of WeChat, Weibo and Douyin with AAPI Do you have to build a login by yourself and or in the login of WeChat to go through the account or API?
Alex Duncan: We have very different relationships with each of those platforms. With Weibo, we pay quite a lot of money to Weibo to get access to that data –
What do you mean by data?
Alex Duncan: We have to pay to access the data and – so every single time we go and get a piece of data from Weibo or we publish a piece of content, we’re paying a few tenths of a yuan. We pay Weibo upfront for an entire year, a very large amount of money and that gives us 20 – 40 million credits and over the year we burn through those credits every time we connect.
Matthieu David: It’s a very common model, to pay in advance and then you use the credits. Cash is king, the platform gets the cash and then you use credit from that.
Alex Duncan: That’s the way it works with Weibo. With Douyin we were incredibly lucky to work very closely with the team who were building the API, so we reached out to Douyin in mid last year and we said – look, all of our customers are wanting you guys to have an API so they can integrate into our dashboard, we can show you what we want to do. Then a couple of months later they got back to us and they said – we’re about to build that API, we would love to know what it is you need, and so we got to work very closely with them, by the testing, providing feature suggestions – so that’s been one of the best interactions we’ve had.
So, we launched Douyin in January 2020 with a great set of features and that’s been a great relationship
Alex Duncan: Yes, Douyin came out. The WeChat I would say we’re still trying to work a bit more on that relationship. WeChat and Douyin don’t charge for their API at this point. Whether or not they will in the future, all of the API’s we work with are a challenge. I would say that the development of KAWO is like an iceberg, what are users seeing is just a 15% that sticks out above the surface. Behind the scenes, we do all sorts of different things to try and make sure that Weibo data is accurate.
One of the most popular dashboards inside KAWO is the overview of an account and it just shows simple numbers like – in the last month how many reads did we get? How many followers did we gain? How many likes, clicks etc. that’s probably one of the parts of our system we have fixed so many bugs with. Sometimes the Weibo API just returns to us that you have zero followers. Now if you have a hundred thousand followers the last time, we asked it and you have zero this time – we started to learn over time to sometimes ignore some of the responses we get from the API cause they’re just occasional blips. We also sometimes get back numbers saying that a post had negative 17 comments? Well, how do you have negative 17 comments on a post? So, huge problems are working with the API’s in China and one of the services being the most effective for us is the KAWO API. So, if you are a foreign company, if you’re a big group of brands, if you’re LVMH or if you’re the NFL for example, you want to get your Chinese social data into your global dashboard. You want to see your Facebook, your Instagram, you want to see China data there too. Now you could go and connect directly to WeChat, to Weibo to Douyin, but we’ve solved all those problems for you. You don’t have to connect up your 30 – 40 accounts, you connect once to KAWO, your teams in China connect their account and then we provide you with one very stable smooth feed of data with documentation and support in English.
How does it plug in your KAWO API? to which software do your customers plug in to – Microsoft Power BI?
Alex Duncan: Yes exactly, so Tableau, Power BI, Google data studio, you know BI tools have become incredibly popular these days. They want to see their eCommerce data, they want to see their social data and so KAWO has the documentation is online, developer.KAWO.com you can see the data, the documentation there. It’s all in English, it’s very clear, it has code examples. If you want to connect to Weibo API, good luck – the documentation is sometimes in English, it’s about 3 years out of date. Like you have to have a Chinese business license, you have to pay an upfront fee. With KAWO API we just make their lives a lot easier.
So, one of our greatest case studies is the NFL. Previously the team here in China were manually pulling their data together and every month sending it across to the New York office to include in their report. Now, every single day they pull data directly out of KAWO for 34 different Weibo accounts, two WeChat accounts and a Douyin account, directly into Tableau and get that live data in New York. So, that’s been a great service for us too.
One of the key feature of your post is the timing. What other learnings do you have about good management of WeChat, Weibo and Douyin? For Douyin, maybe you don’t have enough data about it.
Alex Duncan: You’re right on Douyin, we don’t yet have enough data. For WeChat and Weibo, we do. We’ve published I guess it probably must be getting close to a million posts onto Weibo and we’ve published probably about 60 -70,000 WeChat articles. So, we have quite a good lot of data there. Last October we released an incredibly in-depth report where we took all of this data, we anonymised it, so we didn’t even know who it was for – so we just anonymise all the data and then we laid it out in various graphs and we tried to understand it. We tried to understand what success looks like on WeChat.
For me, one of the most powerful graphs was when we looked at – we tried to follow a growth to content and we realised that on a day that a WeChat account stays silent, is likely to gain followers than on days that it publishes its worst 50% of the content. 50% of content on the average WeChat account is worse than staying silent.
How do you explain it? Is there a kind of retention format and people continue to share afterwards?
Alex Duncan: So, for WeChat, I think it doesn’t make sense to post something unless it’s delivering value for your audience. WeChat as a product-led by Allen Zhang, this real software visionary, who created WeChat and QQ Mail before that, he has said that he has this very clear vision for WeChat which prioritizes the user and so it’s also interesting to see that Tencent gets very little of its revenue from advertising(Know more about the Chinese internet users). It gets more revenue from its payments business than it does from advertising. So, if you compare that to Facebook where its 99% of its revenue, Weibo 86% of the revenue comes from advertising, those businesses are both a lot more focused on how they can support brands, whereas WeChat they have kept brands much more in a walled garden. You can only post a certain time and amount of times to your audience in a service account. That service account is sitting in your main feed and amongst your friends. That’s quite valuable real estate.
To get rid of that service account, you just swipe to the left and you click unfollow. They’ve not made it hard for you to unfollow. 76% of WeChat users follow less than 20 official accounts. So it’s a challenging place to do concept marketing and my argument and I think the data supports this argument is – there is room for crappy content on WeChat, and the way I see it is that so many contracts and so many KPI’s given to teams and set between agencies is – you have to publish a certain number of posts per month – but why try and focus on a KPI like that when the goal is to get your content to a certain level of quality where it’s reaching out beyond your audience. There is no such thing as going viral on WeChat, but 41% of users follow WeChat accounts based on recommendations from friends, either that content being shared directly to them in a chat or it being shared on moments.
When you say WeChat account, do you mention the service account, the subscription account?
Alex Duncan: In both cases, subscription accounts– and in our report last year we showed that subscription accounts are a little bit safer these days. Once WeChat switched to the feed where it’s now easier to just scroll past accounts that you don’t necessarily – are not necessarily interested in, we saw a decrease in the unfollow rate of subscription accounts. Before when it was folders and you had to go into every single folder. If you go into a folder and you didn’t see anything you like – you go back and you just swipe and unfollow it – now, you just scroll past that post. Although WeChat just announced last week – they’re switching the subscription fee to be algorithmic, so that is another potentially very challenging thing for marketers because now whether or not your users even get to see your content is going to be based on some algorithm designed by WeChat which will – we don’t know which factors it’s going to use, but it will be more like the Facebook feed and LinkedIn feed.
How do you understand by the algorithm? Will it push up some subscription accounts you have not subscribed to? Or you will receive content which is not the newest ones from subscribed accounts?
Alex Duncan: Yes absolutely. Twitter was received hell for when they switched their timeline from being this chronological to algorithmic. Previously the WeChat subscription feed had just been – these are all the accounts you follow and these are the posts in order of when they were published. They have recently said that that is going to change and it’s going to be – trying to push you the most interesting content to the top. We don’t yet know how that algorithm is going to work, one thing we can guarantee is that there will be dozens of people out there trying to gamify it, just like there have been for LinkedIn and Facebook etc., all of these tips you get – where you have to get one comment within the first certain number of minutes and somebody has to reply to that comment, you know – this is the side of marketing that I think – it frustrates me a bit. It’s when we’re to gamify success of content based on little tricks like that rather than focusing on delivering value to our followers, understanding our audience – who are they and what can we do to help them and so – yeah, KAWO we still focus our product on helping you understand your audience and create the best content because that’s what we believe is best for your brand.
Matthieu David: On your report, you shared that on WeChat, 68% of the reads coming from pushing content to existing followers. So, the followers are key when you want to push content. So my conclusion is that it’s hard to be viral on WeChat (Learn more about WeChat mini-program 2020), correct me if I’m wrong.
Alex Duncan: Very hard to be viral on WeChat yeah. I mean, I think it’s much easier for foreign marketers to think of WeChat more like an email newsletter – like, you’re very much looking more at an open rate as a metric, whereas on Weibo – I mean Douyin is probably the other end where virality is almost the only thing that matters – like having followers on Douyin doesn’t matter as much as just creating something viral and Weibo there is the potential to go viral like if you hit on a certain hot topic or you create something interesting and cool – again in this report we published last year on WeChat, Weibo data insights, you can download it on our website, we showed the content distribution and this incredibly long tale for Weibo were there this spikes at the end of all this content, just delivered outsized results.
Matthieu David: And in the same article report you wrote you say that 10% of the reads come from sharing on Moments, 3% from chats – basically sending to chat posts if I understand correctly, 4% history – I didn’t get exactly what does this mean-
Alex Duncan: So, when you look in on a WeChat account you can see the previous articles they’ve posted and so you get quite a lot of reads there and one of the disadvantages of segmentation is that those posts don’t show up in your article history.
Matthieu David: I see, and you mention as well that the content should be shared on Weibo 2 or 3 times per day to gain visibility, do you confirm –
Alex Duncan: Well we did an analysis which showed that you get diminishing returns beyond that point. It’s kind of a curve – I think we have a graph in there that shows if you publish once a day – if you publish twice a day you get twice as many reads, you publish three times you get another third on top of that, but then when you publish four times you only get a small increase, five times – we did the same for a look at how many WeChat articles you publish as well.
Matthieu David: Weibo has more people around lunchtime while WeChat reads –you said early evening, I know that you would say that even though you publish early evening if the content is not good, people will not read anyway.
Alex Duncan: I think if you publish something of incredible value at 2 in the morning it’s going to do well.
Matthieu David: Yeah. That’s common to a strategy which is very clear to me now that on WeChat you need to be niche, you need to target your audience. You need to know it. You need to curate the content so that it resonates to your audience and it’s very hard to build an audience, so maybe at the beginning 1,000 people – 2000, 3000 – but you need something which is very niche to them.
Alex Duncan: Yes absolutely, and WeChat – the average WeChat account loses one follower for every two or three they gain, so that makes like – I mean they say the average cost of follower acquisition on WeChat is about 35 yuan when you advertise. So, you’ve got to increase that – you’ve got to double that because a lot of those followers will unfollow you shortly afterwards.
What books have most inspired you during your entrepreneur journey? Could be a book that any entrepreneur would use, what would you say?
Alex Duncan: One of my favourite books ever is David Attenborough’s autobiography. So, for people that don’t know him, he’s a British naturalist. He invented nature documentaries. He started his career very early when television was just getting going in the UK and it effectively defined the genre and for me, it was just incredibly inspiring book because you just get to see the career of somebody with a singular passion and he just became excellent at something and advanced the field and I think from a very personal perspective, my goal is to be an incredible product manager, whether or not I will achieve it in my role at KAWO but certainly, as I look to the future that’s what I want to do – I want to build incredible products that inspire people and I want to share that knowledge that we’ve learned, and so that’s why we give talks, that’s why we try and share – because we learn from other people that shared with us and so – that’s one book.
There’s another – there’s a couple of other films that I would recommend, I don’t know if anybody has ever seen Jamie Oliver’s – School Dinners, it was a TV show in the UK where the chef Jamie Oliver tried to change the way children eat in school and it was incredibly inspiring to see somebody with a singular passion just try and walk through walls when it came to the challenges he met.
When was it?
Alex Duncan: It was quite a few years ago but he realised that kids in the UK were eating crap and he decided that something had to change and so he went into a school and he worked as a cook in the school and then he got permission from the local government to take over the running of that school kitchen. He eventually took over the running of an entire school district and he eventually went to meet the prime minister and got him to commit 300 million pounds to improve the way that children eat in schools and he eventually went to the USA and tried to do a similar thing but with much more mixed results. So, I find that kind of content incredibly inspiring as well – people that just wanted to change things, and in terms of audiences, maybe more interested in marketing, if anybody out there who is in digital marketing in China and has not seen the documentary “the Peoples republic of desire” you have to watch it. It’s a really good one. It’s about live streaming in China and I think it gives you a great insight into the diversity in a cross-section of Chinese society and it follows people who are streamers and people who are consumers of streaming content on the platform Waiwai.
What do you read to stay up to date on China?
Alex Duncan: Oh goodness, that’s a good question. I don’t think there’s any specific thing, I mealy we follow your WeChat account. Just a mixture of different sources. We have an internal chatbot within Slack, that pushes our topics from a wide range of different sources every day. It was written by one of my colleagues – I follow quite a few WeChat groups and occasionally check my LinkedIn feed.
Matthieu David: I feel that now with the presence of WeChat in our daily life and social media – we rely more and more on what friends are sharing, including for marketing content. Even for business information where I think 10 or 20 years ago, we were relying on Financial Times, Wall Street Journalor in China – CAIXIN MEDIA, Southern Weekly to learn about the country and the economy.
Do you feel the same?
Alex Duncan: Yeah definitely. Absolutely. When it comes to China, I don’t rely on mainstream media publications as much – I do rely on another area’s some good ones here, there’s something like TechNode that creates some great content. I rely quite a lot on my colleagues as well, I have colleagues who have e particular interest in what are the latest updates and changes to WeChat and the platforms and so they often inform me.
You have been in China for 12 – 15 years?
Alex Duncan: I think this is my 14th year. You kind of lose count after a point, don’t you?
What unexpected success or failure you witnessed in China’s society or business in the last 15 years?
Alex Duncan: Wow that’s a really good question but I don’t think I was that well prepared for. I think the failures are obvious – we’ve seen this cycle of there just being too much capital in the investment space, after there was 1, or 2 or 3 bike-sharing companies, we didn’t need another 37 – like whoever started the 4th, 5th, 6th, 7th – 20th bike-sharing company, I have no idea what the fact of the idea was in their mind. It certainly cannot have been to create more value in the world and to improve sustainability cause they were doing the exact opposite and wasting investors’ money and we’ve seen that happen with the group buying space, where that got shaken out, we saw that with the car-sharing, with Deedee and the merger there and Uber being kicked out and paying to acquire drivers. We’ve seen it with so many different industries and so that’s one of the things that just keeps happening again and again and we’re kind of watching the demise of – we’re watching Luckin hitting a tough period which I think was obvious to see. Like that product didn’t make sense the way they were building that building, the founders had a track record of doing the same thing before, and that kind of thing makes me very sad. I much prefer – we exist to create value at KAWO, we’re not aiming to profit personally anywhere near as much as we want to build something of an impact. The businesses that inspired me when I was growing up are the likes of people like 37 signals at Basecamp and they’re very open about how they built their business and they talk about how they try and have a lifestyle balance and they try and innovate and they open-source things and they try and create value. We as a company would not exist without the other companies that have done that and the software that we depend on, and so I hope as part of our journey we get to give back and deliver value to other people, deliver value to our team, that’s incredibly important. I want people who work in our team to feel like they get the freedom and the opportunity and the support to do the best work they ever got to do and that when they look back at their career in the years to come, KAWO will be a real highlight of their time here, because we try to be excellent and in terms of what has succeeded in China –
Which is surprising for you?
Alex Duncan: I guess I’ve been a little bit surprised by the rapid growth of things like Starbucks. Cause when I first came here I felt like it was just a Western luxury and I felt like China was a tea-drinking culture, but obviously, China has very rapidly switched to a coffee-drinking culture and I’m incredibly happy to say when I go on bike trips in very remote parts of China, there’s often a Starbucks there.
What’s the underlying wave you are seeing which has been supporting the success of Starbucks in China? How do you explain it?
Alex Duncan: I think it did the same thing that a lot of other companies did when they came to China which was they positioned themselves as a luxury status, desirable product and they obviously – coffee has a certain addictive nature to it which has probably helped them with their growth as well and yeah it’s become – for the colleagues, my Chinese colleagues are just as into drinking coffee as I am and if you live in a French concession like I do, you’ll know that coffee shops are – every second building seems to be a coffee shop these days.
Matthieu David: Yeah, non-big branded coffee shops are also very present, and I want to add up one thing that Starbucks is the case we use in our company to brainstorm about how you can redefine yourself when you enter China –Starbucks is not a takeaway business as it could be in the US and it’s much more like a place where you sit and you stay and as you said, very well decorated, much more premium and nice to stay in.
Matthieu David: Thank you so much Alex Duncan and Alex Li, it was very interesting. Also, thank Sofia, Etienne and Michelle who helped prepare this interview and I hope everyone enjoyed it, thanks to everyone for listening.