TechTalk ep 21: What's your brand's digital IQ?

Paramita: Hello welcome to PwC Luxembourg TechTalk. How digitally intelligent is your company? How to understand your digital presence and know how to leverage that? Today's episode addresses these questions with Benedikt Jonas and Christine Lugrine.

Paramita: Hello Benedikt and Christine, I'm very happy to have you. And Christine, it's going to be just fine because you look like...

Christine: You see the panic in my eyes...

Paramita: Yes with the microphones and everything. OK so to business now, we will talk about digital intelligence today. What is Digital Intelligence, Benedikt?

Benedikt: So digital intelligence actually was created based on a need that our customers have which is like cutting through the data. So there's a lot of data actually around. The amount of data is doubling every two years so and using that data in an intelligent way was kind of like our objective. So what we are doing is really looking at the amount of freely available data that's around there on the web, on websites, on blogs, on forums, on social media and then really analysing that. So filtering out the noise. So that's really what we bring in: filtering out the noise. It's very relevant like in your podcasts. So like our audio guy here is filtering out the noise... So to really have relevant information for our clients to do different things like to better understand the customers, to better understand the market they're in, to better understand their competitors, to better understand certain societal topics that have an impact such as sustainability. You had the sustainability podcast a couple of weeks ago. So it's also something that we can cover. So all relevant topics that's driving and impacting our clients... So that's something that we use and analyse the data to really help then our clients to take better informed decisions based on what the market and what the customers are driving.

Paramita: OK. So we will go into in detail into how you do it, in which sectors you can do it and everything but Christine why did we start thinking about this digital intelligence in the first place?

Christine: Well I guess I mean for me it's maybe... So there's this big data explosion as Benedikt said. More and more data is being created and more and more data is being consumed. So digital intelligence became kind of a need or something that's important as a holistic data... for businesses... to have a holistic picture of the market or what the consumers are saying because today I mean the Web and the Internet is so pervasive in our society, we're on it all the time,  we're creating content on it consuming content on it. So it's data that's out there. And if you're not analysing it and if you're not using it it's a missed opportunity. And there's a lot of insight that can be gained from just what people are publicly and what the media are publicly putting out every day.

Benedikt: And the starting point was actually an internal need. So it started with PwC's needs. Kind of like of better understanding this. So and based on this internal needs then we spoke to our clients and actually they had like similar issues. So that's how it started based on our own internal needs and then actually discussing with clients that they were facing like similar challenges.

Paramita: Yeah. To know where they stand as a brand, yeah?

Benedikt: Not only that but also like really having kind of like the real time feedback of their customers, of stakeholders in the markets, having this really pertinent information on what's driving the market, what are your customers saying rather than having that information from survey data. But this real time connection, looking into real time insights, that was something that people were lacking.

Paramita: OK. Let's go into the how. What are the methodologies? How do you do it?

Christine: OK I can take that question. So for the how... So the first thing you need is the data. So we get that through a third party provider which scrapes all of the data from the Internet. So anything that's publicly available. There is millions of little bots that go through and collect and archive all of the data in real time. So it usually takes like maybe a few minutes, if you were to tweet now, for us to get that data into our platform. So then through that then like Benedikt said we need to filter out the noise because if you look up anything on the Internet today there's millions and millions and millions of data points and web pages and tweets. So then we have to filter out what our clients want and what the question is and what we want the answer to. So then that's through things such as Boolean search queries so saying I want this word not this word. And then in addition some analytics like AI, NLP and text mining to kind of go through all of those results and really just deliver us the data that's relevant, the data that's clean. And then we can analyse that further and turn it into insights that then our clients can act upon. So it's kind of like the high level data cleaning, analysis and then insights if we kind of follow it through the lifecycle.

Paramita: Talking about the analytics and everything, we just launched a website on digital intelligence sometime back and I remember talking about real time brand valuation in there. I think for which you guys won a prize recently didn't you, Benedikt? Could you tell us a little more about that?

Benedikt: Sure. So real time brand valuation again was the starting point was a need from our clients. The product was born based on a client request. So the clients came to us and said OK we want to really measure our brands in a real time fashion and also understand what's the financial value right now of our brand.

Paramita: When you say "measure our brand" what does that mean?

Benedikt: So measure our brand is looking at some brand metrics such as like how known is my brand in the market, so brand awareness or like image strengths. What are the key attributes my brand is associated with in the market or how relevant the brand is in my industry. And for the time being companies are using brand surveys. So again you ask 200 people, what do you think of the brand.

So it's quite cumbersome, it takes some time and then you do that maybe once a year or once every two years because it's expensive. And the client really wanted to move from this infrequent brand survey which also kind of like the results were for them... OK like so what. What can I do with this to moving to something which is real time with some really brand metrics. So those are the metrics that we calculate. And then we have... It's actually a nice country cooperation with our German firm. So while we calculate the brand metrics, our German valuation colleagues calculate the financial value of this. So they take our input like brand awareness, image strengths and brand relevance into their financial model and the financial model goes kind of like the business plan of the clients, sales. And then based on this we are able to calculate the financial value as well as the brand value for the brands by country. We can do that also by product which gives the client like a clear view on how is my brand performing in this market versus the other markets. And if I'm running a campaign like a marketing campaign and a budget I'm spending for this campaign, how is that impacting the bottom line and the financial value. So to really have like a transparent view on the drivers and triggers of both brand strengths and financial value of the brand.

Paramita: Real time?

Benedikt: Real time.

Christine: Yeah that's especially important today with the virality and how things can go wrong very quickly. So your brand when you survey it if you did a normal survey it's kind of very static and you know the next day you could have a scandal or something could reappear in the news that brings your brand value down. But the survey's not going to catch that. So with this real time you can kind of go with the ups and downs of the market with awareness with your image strength etc.

Benedikt: And then we can link it to other data that the company has like sales figures to say OK what's the impact of the campaign in this market on our sales figures.

And again you can add different dimensions. You can add like even your client satisfaction scores in there. So that's with the objective to really provide a client with a holistic view of how the brand is performing in the market.

Paramita: And talking about because you spoke of client satisfaction what other kind of metrics do you take into account in this?

Christine: So for us with the metrics that we have has been set on the high level: awareness, image strength and relevance. So our task was then to figure out how we can measure those via public and social data. So how much people are talking about you. So we take into account things such as volume. How many articles about this brand is being published. How many times are they being tweeted, mentioned in forums and blogs. But then more than that it's also going into how much engagement do they get. So how many likes comments re-tweets do they get on their posts that they put out. How many likes do they get out on posts that other people make.

And then also how many authors. So it could be you know this brand A has you know 10000 tweets in the past two months but it's all kind of pushed by 10 or 20 authors. So it's not a big, a large audience that is pushing the content. It's just one or two influencers or authors. And then the other big thing which comes into play with image strength is if it's positive, negative or neutral in terms of the sentiment. So when people are talking about brand A, is it positive in sentiment, is it negative. So are they expressing feelings or opinions about the brand that are quite hostile or are they more favourable toward the brand. And so we calculate that using an AI sentiment algorithm that goes through and just based off of the language that is used.

Paramita: Is it just social media metrics that you take?

Christine: So for our brands score it's social but then also anything that's publicly available online. So blogs, forums, news sites...

Benedikt: It also yeah... It depends on what angle do you want to take. So we're differentiating between what we call owned media and earned. So owned, it's kind of like the brand's official communication which is not neutral. And then you have... you mirror that with OK what are actually the customers saying and are they engaging in this. So you have like owned and earned as differentiating elements that we look into and what Christine just said so like how are we doing this. So does this really kind of like where we have developed like a unique IP with the team of really creating this algorithm. So there was a lot of time and effort spent on really building the right algorithm selecting the right elements to build the KPI because as Christine said it's super important that yeah we're looking at unique number of authors. So you don't want to have one person tweeting a thousand times and influencing the result.

So there's a lot of effort and technology put into play to really make sure we have a clean data set so that we don't have any bots in there and that the data we take is clean. So there is a lot of effort that's put into there. In today's world there's a lot of buzzwords using AI and natural language processing. But we are actually applying that on a daily basis. That's the only way of aggregating and cleaning the large amount of data at scale in a clean way.

Paramita: Let's take a concrete example and let's say that firm XYZ they want to apply this digital intelligence. What would you recommend them? Do you have any best practice? Any tips?

Christine: I think for me I mean there is a few important things that come with digital intelligence. So one important thing is that it's not just the data. It's not just saying OK I have access to all of these public web sources. The intelligence part is important. So being able to analyse and filter through all of that data. So having the resources or the knowledge to go through and filter out what's kind of you know just rubbish and what is actually strategically important for the brand. And then I guess another thing that's important for digital intelligence is again we're dealing with lots of data. So if you come at all of this data with kind of too large of a question you know like if you go with it saying I want to know what's going on in AI what you're going to get out is not going to be strategic or concrete enough for companies to act on it. So it's really coming at the data with these strategic kind of targeted questions that you have that you want to answer and then figuring out how the data can answer that question or how you need to analyse the data or segment the data to answer those strategic questions.

Paramita: Yes probably as in like you need a specific objective before starting the entire exercise to know what you're looking for.

Benedikt: Exactly and that's where also like we are helping our clients to formulate the right question as Christine was saying like to be really specific and really trying to find out where to help. What are they... What is really helpful for them and to link the big data and insights topic to the strategic business part. Which means let's take an example like sustainability and environment, if that's at the core of what the brand is trying to have as an objective in terms of sustainability. So it's discussing with the client OK what angle should we take. Do you want to look at your supply chain and your suppliers. And that we filter and apply those sustainability environmental filters to see if your suppliers are actually performing according to your standards or if they're if we see issues that come up with some of your suppliers on environment. We can flag that.

Or if the angle is we want to understand how important the sustainability topic is for our clients we can then again filter another perspective which is how often sustainability and what type of topics within sustainability is discussed among the clients, in which context, on which channels, which audience... to help on different scenarios. So to help OK like on which communication that the information is discussed. Also understanding, and that's an important point, how are people interlinked. What we call like the like influential mapping or stakeholder mapping. Who are the key stakeholders in my industry who's talking, who's an opinion leader and how to link that back. So that's another way of helping our clients to understand the communication ecosystem and how it works. This is also very relevant if you're in crisis communication mode. If you have like an issue to see OK where's the communication and we have had some use cases also where we have analysed the type of contents. Is it authentic content or is it just like re-tweets?

Paramita [00:18:44] Fake news?

Benedikt: Fake but also which is re-tweeted. Which has a different level of quality than if I take the time to write something personal. So helping then our clients to estimate do they need to react and also to see where communication is flowing.

Christine: It's important especially like Benedikt said to understand the data that you're dealing with. So it's a lot of just people putting their own views out there. It might be bits, it might be groups... It's companies. So it's this huge data set that I think it's important to always keep in mind you know the kind of lens of this is what people are seeing and what people are believing whether or not it's fake or it's true. And then being able to analyse. OK is this authentic. How many people are going to see this and really cut through the noise which takes some time. The Internet is a mess in a way, a good mess, but so it's important to have that...

Benedikt: And then for us it's to see also like there are as many use cases as kind of like creative examples we have in our head. So you can do that on a brand level, you can do that at a product level and we could even do this on like product features. If you have like a product and you want to understand what features are most discussed in the market and which of them are positive, which of them are negative.

So then again kind of like this can loop into R & D and product development and say actually a lot of people are complaining about this product feature. So it's really another way of getting unprompted customer feedback and integrating that in the loop of like product development or product distribution. So really helping to get the customer experience right.

Paramita: OK. I'm just probably going to be the devil's advocate here. While I understand that you know it's all really about understanding the customer and understanding how they're reacting to certain kind of things. Is there any risk in doing... in listening to what people are saying?

Christine: I think for me it's always important that we stress that it's publicly available data and especially with GDPR in the EU now, everything that we get all complies with what people are publicly putting out there unfortunately if they know it or not it's ambiguous. But so it's important that we have that kind of lens that it's public data that we're getting at. And then again like I said it's understanding the data set that there are certain people who go on social media and who are most likely to talk. Especially you know social media tends to be a negative place. So you're going to find a lot of people complaining on social media. You know that the two ends of the spectrum people who were super happy and super dissatisfied with the product they got or the brand. So kind of having always that perspective and the understanding and expertise to know kind of what types of people we're dealing with and what types of content is more likely in this data source.

Paramita: And it's public like you said it's publicly available.

Benedikt : I think it's just like reality today if you really want to capture the customer it's less on paper survey. So you have to be there where the customer is and the customer is predominantly on social media almost 24/7. And so it's being there where your customer is and then capturing that information and as Christine said it's like public, it's publicly available information. So if I tweet about my bad check-in experience at the airport, I'm conscious. And I want to know like I want that the people know that I had a bad experience that I'm sharing my frustration hoping that someone will pick this up to improve it for the next customer.

Christine: Exactly. They usually tag the brand and all of the different companies that are involved anyway. So it's I think it's kind of responding to the customer's needs. You know today everything like we said is real time. Everyone wants things on demand. So it's just being up to date.

Benedikt: And it's about like easy to use. Probably will not go to the complaints office or get like a form, wait 10 minutes and then fill it in 20 minutes. I want to tweet that in 30 seconds knowing I can tag the brand and making an impact kind of like in 30 seconds.

Paramita: Yeah I've done that actually.

Benedikt: So would you... let me turn the question then to you... Your complaint that you tweeted about and if we would use that would you feel...

Paramita: No. If my objective really was to grab the attention of the airline company where I did actually tweet I would rather that it gets picked up by something like that and you know it gets reported back to the company so that they can improve the customer experience. Of course...

Christine: Yeah and I think it's important sometimes. So these companies they have social media teams that usually reply. Like I've had a few times where I tagged a company and they do reply sometimes in 30 minutes but it's having again the insight of stepping back. You know sometimes they deal with request by request. But digital intelligence is really about stepping back and seeing the whole picture. But then you can see oh wow we got all of these tweets at this time, in this store. So then you can kind of pinpoint larger issues or bit more strategic things that then you can address.

Paramita: So to wrap things up what are the trends? What's the future like in digital intelligence?

Benedikt: So I think we have like three priorities that we see. One is really bringing the voice of the customer as close as possible back to the brand. And in a structured way. So this can be across the customer journey and across the touch points.

The second one is and we spoke about this like real-time brand valuation to really measure the strength of a brand and replacing survey data.

And the third one is really I would say a trend that touches every one that was discussed on the previous podcast like the sustainability topic. So how sustainability is impacting customer behaviours. It's impacting kind of like the markets and the brands and how we can help from our insights to also shape the debates in identifying issues in terms of like sustainability perception of your brand or your products. Providing the right insights saying I will not buy this product any longer because it's using a single plastic usage in terms of like packaging. And really using technology for good to improve really on the on the product side.

Maybe a fourth one just to add is also like what we do is like building some custom indices meaning that if you're and I'm speaking out like if you're a financial company and you want to invest in a couple of companies, to understand which of those companies is best prepared in terms of climate change risk for instance. And there really get the ranking and get some insights again on kind of like methodology that we have developed to see OK those are the most sustainable companies according to market and customer perception. Not according to their own sustainability programme but really what the market and the customers are saying. So it's bringing the power back to the consumer in the markets and taking that data to help them make decisions.

Paramita: So would you guys really say that today digital intelligence is kind of becoming imperative for... quite important yeah for brands to be self-aware let's say and to improve their customer journeys?

Benedikt: I think like in today's world no company can really ignore the customer and the industry or markets that it's operating in. And so they need to find one way or the other to capture this. So one way is through our digital intelligence service. There might be other ways like I said like somewhat traditional ways of surveys.

But if you want to have this as an ongoing basis that's something that we see clearly as a trend in the market.

Christine: I think it's an important part of a comprehensive data programme if I were to say. So if you don't have that type of data in the data you're analysing then you're really missing out. And in today's world, the Internet and the public web has become so important to us that it's important for companies as well to use that and to see what we're putting out there, what we're consuming you know. And when maybe the Internet isn't important anymore then maybe it's not as important of a data source but I think today and I believe into the future that it's going to remain you know and increase in importance of using this data.

Paramita: And I think maybe the... Even you know in our last podcast it was on BXT and the phrase that came up all the time is "real time" you know, the real-time analysis, the real-time knowledge on what's going on because it is evolving so fast.

Benedikt: Yeah. And I think BXT is a good... It's kind of like how we are operating. So like you need to have the business knowledge, the business experience. You need to have the technology part, people to understand how the algorithm works. And you need to understand the X-part, the experience part. Like how is this linked to customer journey and customer touch points.

So I think that was a good reference of also like how do we work in different competencies and disciplines like within the firm but also like within the network. Because the service that we provide it's not just like for our Luxemburg clients. We are supporting the entire PwC network with the expertise that we have built here which is again kind of like one of the strong points in Luxembourg where we have like a multilingual workforce to also support the network on this.

Paramita: Well thank you guys. It's you know I really should say this that I really love doing this because I get to know about all these stuff and you like you're saying it gives me a holistic point of view on different aspects. So thank you very much.

Benedikt: It was a pleasure. Thank you.

Paramita: And until next time.

Benedikt: Absolutely. We'll make it... we'll not wait as long as this time.

Paramita: That's all for today. My thanks to Benedikt and Christine. I'll be back next week with a new episode of PwC TechTalk.