We think, learn and love to share. Read the latest insights from our team and let us know what YOU think
I want to read everything
If you are unfamiliar with how digital designers work, wireframes might just look like a series of grey boxes with some random text dotted around. They can sometimes be as simple as a hand-drawn sketch on a piece of paper, or a drawing on a whiteboard, but you’re most likely to see them as low fidelity (simple) or high fidelity (more designed) digital designs.
They are in fact a super important ingredient when designing a successful user experience for a digital product and they act as a kind of skeleton structure for your site. We use wireframes to plan out what type of content will sit where on a page, based on how important each piece of information is. The content and functionality is mapped out, taking into account all the work we’ve done to define our users’ needs and the journeys we need to create for them. Think of letting a builder and an interior designer loose in your new house to knock down a few walls and start painting without first thinking about how the space needs to work for you with an architect and having plans drawn up. You wouldn’t expect the end result to be good, or to work for you and you’d probably just want to start again.
Because wireframes provide a map of how each page will be structured they allow us to look at the site structure with clients, and test journeys with users, changing things based on feedback, long before that becomes really costly. It’s much quicker and cheaper to change wireframes than it is to start messing with high fidelity designs or fully built products!
We often find that for people who don’t regularly work with wireframes it can be difficult to remember that they aren’t showing you how the final page will actually look. They’re showing you the layout, mapping out content and functionality, page by page. It can be difficult to get sign-off on the page structure without getting distracted and talking about the visual design. Testing with users can also be tricky for the same reasons so it only gets you so far. There will be things the visual design and the words will do that will help users find what they’re looking for that won’t be added until the next phase. It’s sometimes hard to find your way around a prototype that’s made up of only grey boxes.
We use wireframes early in projects to make sure we’ve planned out how we’ll meet all the users’ needs and the business’ needs that we identified in the discovery phase. If we can get things right at the wireframe stage it saves time and money later on in the project because we don’t have to change things when it’s harder to make changes. It’s also really helpful for the development team to have this to refer to as the project goes along – a map of the core structure and functionality of the site. Everyone needs a map to get where they’re going after all.
If you’d like to talk to us about how we design websites, apps, prototypes or anything else just get in touch.
If you’re sitting there thinking you didn’t realise a website could be environmentally unfriendly don’t worry, you’re not alone. Despite the internet being responsible for producing 3.7% of global greenhouse emissions each year (that currently stands at more than aviation!), there is very little awareness of the problem. As a designer, I design for the user and historically I rarely, if ever, think about the impact of my designs on the environment. However, I’ve recently became aware that there are things that designers, alongside developers, can do to minimise a website’s impact on the environment. To learn more, I spoke with some of our development team who already practice environmentally-friendly software development.
This is a good place to start because it’s only once people know how websites use energy they can begin to look at ways to reduce the amount used. Without getting too technical, the way a website works is this: a website sits on a server and every time someone wants to view a web page the browser needs to send a request to the server to ask it for information to display. In order to gather all this information, the server will compile and send back the information requested, routing back through switches, DNS servers, routers and caches. All of these machines require energy to transfer data, in the form of electricity, to do what they need to do. The more complex a website the more requests it may need to make and more requests & data = more energy. If you consider how sophisticated many websites are nowadays: videos, high-resolution images, motion graphics, forms and more, you can begin to get an idea of how much energy websites consume.
There are many ways but the below are the top ones…
Images and videos are often very large files with more data than any other parts of a website page and, in order to send large quantities of data, servers use more energy. There are different ways of reducing the data weight of an image or video; the first is to use fewer images. This isn’t always possible, especially for websites whose popularity relies on their imagery. A client of ours, the world-renowned architecture and design magazine, Dezeen, is a good example of this. While Dezeen can’t reduce the number of images on their site without affecting the popularity of their site, they are working towards reducing their impact. How? This leads us on to the second way of minimising the data weight of images: optimising images and videos in such a way that the file size is reduced without impacting their quality or the user’s experience.
There are two types of fonts; fonts that come installed on computers; ‘system fonts’, and others that aren’t; ‘custom fonts’. While custom fonts allow for much more flexibility and artistic style, they can often be large files and require many requests to the server for the font file (as opposed to system fonts where all the data is already on people’s computers so no data request is needed). It might not always be an option to use system fonts as the choice is limited, but using custom fonts sparingly would be a good place to start.
How people navigate around your website is important. If the content structure is unclear people will click on numerous different pages before landing on what they wanted to find. By laying out your site’s content clearly, navigation will become easy and users will find the content they’re after more quickly, reducing the amount of energy used, both by your user’s device and the server.
Minimising the use of third party scripts can vastly reduce your website’s carbon emissions. Consider auditing your website to ensure all third party scripts are absolutely necessary. Having your website written in clean and streamlined code also decreases your website’s carbon emissions. Optimised code, written in a concise and standardised way (to remove duplication) requires fewer requests and less energy.
Much of the energy consumption a website creates is in data centres (physical sites where servers are held) through the transference of data. These days data centres have an energy-efficiency rating so you can choose to host your server in a highly energy-efficient one. Not only this, but you can also choose a data centre located as close to your target users as possible. As energy is used in transferring data, by reducing the distance data needs to travel between your servers and your users, you reduce energy consumption. So, if your users are located in the UK we would recommend hosting your server at a UK-based data centre rather than one in say, the U.S.
All of the above changes decrease the time it takes for a website page to load. We’re all used to such frictionless experiences on the internet that our expectations of websites have skyrocketed; we want to buy a product with one click, we want websites to remember our information so we don’t have to fill it out twice, we don’t want to be overwhelmed by choices and, heaven forbid, we have to wait for a page to load! There are lots of different facts and statistics on how many users will drop off after just a 3-second wait time but what they all agree on is that the longer a page takes to load the more people leave. It’s also worth mentioning that Google has used page speed as a ranking factor since 2010 so a faster page speed will improve your SEO which means you’ll appear higher up on Google search results.
This might go without saying, but if your users enjoy their time on your website – if they’re able to find what they’re looking for easily or don’t have to wait for pages to load, they are much more likely to stay around and do what it is you want them to do. This could be anything from buying your product, signing up for your newsletter or sharing some of your content. In short, the changes we’ve spoken about in this article will all provide your user with a smooth and enjoyable experience and this will increase your conversion rates.
We hope you have found this article eye-opening about the impact websites can have on the environment. I’m going to finish by saying that while we should all continue striving to make our websites use as little energy as possible, it is inevitable they will always use some. A way of tackling the remaining amount of carbon emissions is by offsetting it through investing in an equivalent carbon saving elsewhere. There are many ways of doing this but at Den, we have a subscription with Ecologi, allowing us to offset our carbon emissions. So far we have offset 31.22 tonnes of carbon and have planted 709 real trees in Mozambique, Madagascar and the UK.
You might now be wondering how environmentally friendly your website is and exactly how much energy it uses. If you’d like to talk to our development team about your website and reducing carbon emissions, get in touch here.
You might have seen the terms UX design or UI design popping up all over the internet in the last few years and wondered what they both mean. ‘UX’ stands for user experience and ‘UI’ stands for user interface and both are part of an approach to design called ‘user-centred design’ (UCD).
UCD is growing beyond design studios and start-ups. Now, many of the world’s top universities and business schools are including UCD modules in their programmes.
What is User-Centred Design?
UCD is an empathetic approach to design that puts the user and their wants and needs at the heart of every decision when building a new product.
A user-centred approach incorporates validation from the user at every stage of the design process, as opposed to showing a user a finished product and hoping, with fingers tightly crossed, they like it.
UX, UI and UCD defined
Put simply, UCD improves the user’s experience of a product. If a product has been designed in a user-centric way the chances of a user liking it go up. This increases the likelihood of adoption, conversion, or whatever your business goal is. Where does UI design come into this? UI design is a part of the user experience and is concerned with how the product looks (vs. how the product works) i.e. what the user actually sees when they look at the product – think colours, fonts and shapes. In short, the better the UI design, the better the user’s experience is and the more successful your product is likely to be.
What is the user-centred process?
The first step in the process is establishing who your user is and answering the following key questions: Who will be using the product? Why will they be using it? What are they like as people? What are their daily habits? Do they have any fears or concerns that might be relevant to the product we’re creating? What problem/s are they trying to solve with the product? When and in which contexts will they be using it?
The second stage requires defining all the user requirements that need to be met in order for the product to be successful. This is also the time to define the business requirements too, for example, opportunity for upsell, data capture or others. While the users’ requirements should be at the centre of this process it is important not to disregard the business’ goals.
Now we begin designing! This stage requires problem-solving and designing different solutions to meet the users’ needs and requirements. This is best done in stages so as to make sure the product really is what the user wants without getting too far ahead and having to backtrack (backtracking can be costly!).
4. Evaluate & Iterate
Once we have a minimum viable product of a solution we think might work, we put it in front of a user; we watch them use it and we ask them what they like and don’t like. It’s also at this stage of the process that we often uncover unrecognised needs and desires, ones that even they didn’t realise they had. Through observing them and using tools like biometric testing we gain invaluable insights into subconscious habits and actions. Taking all this feedback with us we go away, make changes and then put the amended version back in front of them for a new round of feedback. This iteration process continues until we have a successful product, and beyond.
The user-centred process is a collaborative one between product creators and product users and in our experience it radically increases the chances of success of a new product.
How does it differ from other design approaches?
The opposite approach to user-centred design could be called ‘organisation-centred design’. This approach to design prioritises the business’ needs when creating a product and is the approach still taken by many companies. Businesses that take this approach first design a product and then find users that might like to use it. Unsurprisingly, approaching design this way often results in products that users don’t like or need, and of course, wasted time and money.
How a user-centred approach can benefit your company
A user-centred approach to design is a flexible way of working and that has a few advantages.
Faster to market
Minimising the specification means you get your product, albeit a simple version, in front of your user in a relatively short time frame and thus beating your competitors to market.
Keeps development costs down
With this approach, there is no wasted money on building features that the user doesn’t want or need. Getting the users’ validation at each stage means you only build what people will use. It also removes the margin for error so saves money that would otherwise have been spent on going backwards.
More desirable product
As a user-centred approach is guided by the user, the product is almost always exactly what they want. As we mentioned earlier, the user doesn’t always know what they want but through an iterative approach, we can uncover these needs and design a product that works for them.
Increased customer satisfaction
If a product satisfies all the users’ needs, you’ll have happy customers, which in turn means increased conversion rates, higher product adoption and realised business goals!
If you’re interested in our approach to design and would like to hear more, we’d love to speak with you, so get in touch.
Meet the team – Sophie Li Henry
Sophie joined our development team here at Den nearly two years ago now. Although it was a team effort, Sophie built this website for us and she’s become a super important part of the team, not to mention our first female developer, something we hope to build on!
It’s easy to showcase your design work as an agency, but much harder to show off your technical capabilities. Our development team are at the heart of the experiences we craft. Their thirst for knowledge, along with wanting every line of code to be better than the last, is key to our success. We asked Sophie some questions to find out more about her move in to web development.
So, why development, and what do you enjoy about it?
‘I was always interested in computers as a child but I never seriously considered coding as a viable career path. Growing up, it seemed inaccessible and out of reach. I subscribed to the misconception that you had to be the next Mark Zuckerberg or some kind of academic maths prodigy to be able to learn to code. I graduated with a history degree in 2018 and started to apply for roles in Social Policy, but I was struggling to maintain my enthusiasm. It felt like there was something missing.
It was so gratifying to write some code, tell the computer to do something, and then watch it happen on the screen in front of you. I soon realised coding isn’t about how good you are at maths – it’s about how you think.
I joined a coding bootcamp and then started creating one page websites for my friends to showcase their graduate portfolios as practise, before applying for junior frontend developer roles.
Technology is constantly reshaping our way of life and how we do things. Working in an industry where you can watch ideas unfold from a whiteboard, grow into prototypes and eventually transform into a real life product/app is so exciting, and I wanted to be a part of it.’
What are the lows of development – what do you wish was easier?
‘Something that I wish was easier about development is how to explain something technical to a non-technical person. Often when you are working with PMs, clients or consultants, translating what you have been working on in a concise and clear way to someone not familiar with coding concepts can be challenging. The best way I’ve found is to use analogies and try to be as jargon-free as possible. To be able to do this proficiently is a skill in itself but its very valuable in bridging the communication gap within teams. Explaining to someone else also helps deepen my own understanding as breaking the process down into bite size, simple chunks reinforces every step of what is truly essential to the process.’
What are your favourite kinds of projects to be involved in and what is it you like about them? What interests you?
‘I love working in collaborative/agile teams. As a junior developer, it’s the best way to learn. There is so much to gain from working with other devs. Daily stand ups where you discuss what everyone is working on are crucial to our understanding of how the system works holistically and how each developer’s role helps to join the pieces that fit the puzzle together.
Currently, I am working on the front end of a digital banking project. We work in fast paced sprints where teams are broken down into smaller units, which often leads to pair programming. This has been invaluable for learning best coding practices, methodologies and reliable systems for making sure your code and design is maintainable and scalable, as well as sharing recommendations for the best third party plugins, which are all dev approved. For example, in this project, I learnt defensive programming techniques which are designed to help your application behave in a consistent way even in unpredictable environments. I also gained experience in Tailwind, a utility-first CSS framework, Blade, a Laravel templating engine, as well as Cypress, an E2E testing framework which involves testing an application’s workflow from beginning to end.
Pair programming has also been a good excuse, especially now that we are all working from home, to catch up with the other devs – lots of screen shares and daily zoom check-ins help keep the office spirit alive.
Other kinds of projects I like working on are anything with a fun, quirky, interactive design. I would like to look into creative coding one day which has become a very interesting intersection between coding, design, art and technology. I came across creative coders on instagram like @tim_rodenbroeker, @zach.lieberman and @kiyoshinakauchi who use generative design software like p5 and Open Frameworks to explore the creative/artistic possibilities of programming, giving new meaning and interpretations to loops, algorithms and patterns.’
If you’d like to find out more about our development capabilities, or you’d like to join our team, Get in Touch!
If you own a successful small or medium sized retail business but you’ve never sold your products online, you might be wondering whether you need to. Or perhaps you’ve just bought a business that has a healthy balance sheet and proven track record but no online presence and you want to supercharge growth. Either way, it’s time to go digital. Here are our top five reasons to embrace e-commerce.
Some industries have been decimated by the rise of ‘digital first’ giants. It’s hard to predict which sectors could be hit next. From Amazon closing down department stores, to music streaming sounding the death knell for your local music store, if you aren’t yet selling online, you are more vulnerable than most.
Statista claims that in 2019 there were 1.92 billion digital buyers worldwide with e-commerce sales accounting for 14.1% of all retail purchases. If 14% doesn’t sound like enough of a reason to go digital then think again. This share is rising year on year and e-commerce is the biggest driver for growth in the retail sector overall, both in North America and Europe.
Every business wants to make more money. Numbers vary from business to business but BigCommerce says that physical stores that open an online store are making 28% more in revenue via their online channel within just six months. You are tapping into an audience who may only shop online (or even only on their phone) for convenience. A whole new audience you couldn’t get into your store. A larger customer base means more revenue.
The modern consumer has honed in on their affinity for certain brands. They don’t just choose the company that is geographically closest to them anymore, they choose the brand which is most aligned with their values. An online store gives you the opportunity to communicate to a wide audience who you are as a company, what your values are and why you do what you do.
Modern e-commerce platforms are sophisticated. They can sync with inventory management and fulfillment systems. You can build your audience with search engine optimisation, social media campaigns and paid online advertising. You can even send out emails to customers who filled up their cart and then got distracted, to try and tempt them back to complete their purchase. Opening your online store can form part of a wider project to digitise analogue, outdated processes across your business. Saving money and making more money at the same time. What’s not to like.
This is where we come in. Den can help you get your brand ready for digital, design and build your online store, or help you with digital transformation across your business.
Creative thinking can have a huge impact on business and in a climate where businesses must innovate or die, this is more true than ever. Aspiring creatives leave university or college with a basic set of tools, but what they do with them and the impact they can have are really wide open. Den went to the University of the Arts London (UAL) Creative Futures week to inspire the next generation to think about design thinking and the impact it can have on business.
Our growing team of designers and developers have between them an amazing breadth of skills and experience. One thing they have in common is that they never imagined the impact they could have on the business world, by applying the kind of thinking they have learnt in their education and work lives to date, to business problems.
Den regularly undertake transformational brand and web projects for our clients but since teaming up with Elixirr, we also take part in the design thinking process that Elixirr, the challenger consultancy, have made their own.
We headed over to UAL at Elephant & Castle to tell the story of this kind of work.
I acted largely as a kind of master of ceremonies, with the real content coming from Cat, an Elixirr Consultant, Vanessa, one of Den’s Creative Designers (and incidentally my inspiration for putting us forward for the talk) and Rob, Den’s Creative Director, there to answer questions about career success in the creative industries.
We took the assembled students through our process:
At the start we help our clients to define their challenge (a problem they want to solve). Then we aim to expand their thinking by showing them cross-industry content about innovation and how consumer’s expectations are being raised by other industries to the point where they are no longer happy for signing up to a bank account (for example) to be a lengthy, paper based, in-person process.
With minds expanded, our team take the clients through a process of refining a solution. The idea is not just to improve their current offering but to move into a space they might not have imagined before they defined the problem, but to really aim high for their consumers. The outcome of that workshop process is often some very low fidelity wireframes of a potential application, dashboard or similar that is going to do more than just solve their problem.
At this point the design input really ramps up. We go through a process of rapid prototyping and user testing until the design is ready for the team to build the MVP (minimum viable product). Sometimes we can take advantage of time zone differences and clients will go for dinner leaving hand drawn app screens on the wall and wake up in the morning to a working clickable prototype. We move from low fidelity prototypes to high fidelity designs making sure that the UX is slick and the look and feel is inline with the brand and enhances the user experience.
We then talked through translating the designs into the final product through the build phase of a project. As our developers are part of the Den team, they work closely with our designers and the consultants to make sure development is not only fast but accurate. We can sometimes go from defining a problem to launching a new application live in the space of four months.
Vanessa inspired me to suggest Den take part in Creative Futures because she said to me that she loved the work she was doing. At the time she was working to solve extremely complex user experience issues on a trading platform. She added that never in a million years would she have imagined that this is what she would be doing a few years on from leaving her design training at UAL. It feels really good to be making such an impact with the tools she left university with and adds to every day.
We had a great audience who asked lots of questions about how their skillset might fit in to the journey we had discussed, as well as seeking more general career advice. It felt important for Den to be part of this Creative Futures event. If we don’t reach out and tell our story then we only meet the creatives who have already made the leap to see their value in the business world, and we aren’t necessarily meeting the very best thinkers from the creative talent pool. We always look forward to welcoming new creative and technical talent to our team!
I am not a savvy shopper, definitely not a committed bargain hunter, but I am obsessed with brand identity, authenticity and the emotional connection between brands and their consumers.
Everything about walking into a shop in sales season puts me on edge – the jumbled rails of mismatched items dangling by a shoulder from the hanger, the elbows, the frustration and palpable anxiety bubbling just below the surface in my fellow shoppers. There’s also the stories of people getting stretchered away after midnight stampedes and security guards getting battered out of the way, around the world. Online sales are a little easier for me – less elbows, less bubbling rage, more filters. But, you can still trawl pages and pages of ‘load more’, find exactly what you want and then discover it’s only available in a size 6 (not my size by the way). I’ve often used up the last of my short attention span to get to this point and the frustration is too great, I bounce.
So, it’s no surprise that on Black Friday, rather than queuing from 3am to get my hands on an Xbox, I’m analysing how brands have approached the dangling carrot of a revenue feast and whether they’ve managed to hold on to the meaning of their brand, the message they want their consumers to believe that they believe in, or if they’ve left it trampled on the floor below the shelves piled high with goods, bought in to flog to the frenzied crowds. Have they harmed or enhanced the loyalty consumers feel towards their brands?
Some brands have nailed Black Friday. They’ve cut through the noise and most importantly, they haven’t lost themselves along the way:
Dewerstone are an outdoor lifestyle store and brand. The first line of their ‘About Us’ page on their website says ‘We’re people that love the outdoors.’ And what did they post on Black Friday? ‘We’re shutting the shop, office & warehouse doors on Black Friday, we’re just not into it. We’re going adventuring instead.’ They don’t just talk the talk, they walk the walk. They are prioritising their team and the outdoors for a day. As a brand who want to be part of the outdoor community, who typically are interested in quality, and the environment, the risk to their revenue in not taking part in Black Friday, was outweighed by the risk to their brand if they did. In case it isn’t clear enough whether they care more about their customers’ loyalty or their bottom line on Black Friday, they even take a moment to thank their followers for being customers every other day of the year.
Patagonia, another outdoor brand, state very clearly on their website that their mission is that they are in business to save ‘our home planet’. In 2016 they launched their Black Friday campaign ‘100% Today, 1% Every Day’. They pledged to donate 100% of their Black Friday sales to grassroots nonprofits working on the frontlines to protect our air, water and soil for future generations. And guess what the final sales tally was for Patagonia that day? $10 million. Good call.
Pieminister turned Black Friday in to Black Pie Day in 2018. Since their inception, Pieminister have donated pies to good causes, from homelessness hostels to community fairs to fundraising banquets. They call these donations Little Acts of Pie-ness and their website states that they’ve put over 200,000 pies to good use in this way already. Every Black Friday they give away pies in their shops and pop up shops around the country in exchange for donations to Shelter, raising thousands of pounds. The purported values of the brand, what they care about, what they think matters is clear to any consumer. It isn’t a gimmick, it’s who they are. So, they lose a days sales, but they gain a loyal band of pie eaters in the process.
They have always maintained a policy of no deals, no discounts, for their game. Before Black Friday 2013 the game had always been $25,and the makers felt that doing any kind of deal or discount would undermine the simplicity and honesty of the game. This is a really important brand value and really difficult to uphold in the maelstrom of deals that is Black Friday. In 2013 they settled on their angle, they raised the price of the game by $5 and saw a huge sales spike as a result. Their campaign parodied the slightly stomach turning marketing concept of establishing scarcity, leading people to purchase when they wouldn’t (or shouldn’t) normally, due to a manufactured sense of urgency. They crossed out the $25 dollar price tag and listed the new $30 dollar tag, with phrases like ‘Today only!’ And ‘Once-in-a-lifetime-sale!’.
This year they’ve gone fully dystopian pitting their human staff against a bot to see who can come up with the most popular pack of new cards. Again, no discounting and a perfect illustration of their playful brand identity, hopefully creating new encounters with their brand for new customers.
A sustainable fashion brand, founded by a group of surfers committed to protecting the oceans have turned Black Friday in to Green Week: ‘Black week turns green – one order = one tree’. They offer to plant a tree for every order placed in Green Week. They haven’t ignored the discounting element of Black Friday, offering ‘special discounts for a better future’, but they’ve added meaning. To date (since Sunday 24th November – in 6 days) they have planted 4673 trees. That gives me, as a consumer, 4673 reasons to believe them when they say they care about the oceans, and that they believe that without green, there is no blue. They are acting in support of their words and getting my buy in as result.
Defining what your brand means, and what you believe in is the first step in any branding journey. Long before you have a logo and a colour palette, you define your identity. Then, you need to build and protect consumer belief in that message. Once lost, never regained. The excesses of Black Friday leave me feeling a little queasy and, particularly in a time of environmental disaster, a burgeoning debt crisis in the UK, and increases in homelessness and poverty, the idea of manufacturing a sense of urgency to encourage people to spend when they might not if they had longer to think about it seems wrong. It leaves me cold.
Offer to plant a tree with every order? I’m pulling out my wallet. Shut up shop to go adventuring on Dartmoor? I’ll be queuing up the next day for you to open the doors. No brands can afford to ignore huge retail events like Black Friday but what you do about it can help you define who you are as a brand, or it can destroy the identity you’ve worked so hard to communicate.
We as humans have started using natural language processing commonly in our everyday lives. It’s a digital trend which has emerged alongside the prevalence of AI. It is predicted to be a large part of businesses innovation going forward and we, as consumers will be interacting with and using it on a daily basis in addition to implementing it our products.
Natural language processing is essentially voice or text recognition software also known as NLP. It’s the process of taking a human user output processing it with artificial intelligence, transforming it into something that a computer can process and determine an appropriate action to take. For example, when you talk to you mobile phone and ask siri or the google assistant to write an email for you, this is considered NLP.
For NLP to understand human language the program needs to be trained for it. This usually means processing thousands of items of text or speech in order to get a base understanding of the human language. This includes words, grammar and names in addition to the natural flow of the language.
NLP services are usually trained with text books for example since these have correct spelling and grammar throughout. It’s important to remember that the AI will be trained on a specific language such as England or Spanish so if it’s to be used in another language / culture it we need to be trained specifically for it. It can be trained on anything, you can make up your own language and then train the NLP model to respond in the way you want but it might not be a very popular service.
For example “I would like to buy size twelve shoes”.
The intent might be “purchase item” and the entities would be “size 12” and “shoes”.
If this sentence was used with a bot on ASOS for example it may then direct the user to a page of shoes size 12 that they can purchase.
Often NLP models will be designed to work for a specific purpose so they will be expecting a type of intent, this makes it much easier to train and use. For example an ASOS bot will be expecting questions about its products, company or orders and not questions about if the earth is flat.
The NLP software is likely to be running on a server and accepting requests from lots of sources, this means any users with a device connected to the internet should be able to send a request to the software.
Natural language programs that can process human speech usually work by being trained on transforming the voice speech into text. Once they can transform the speech into text they work the same was as other NLP services by processing the text as intent / entities.
A simple use is to direct a user around a website, when they first navigate to a site they might not know where to find what their after so if there is a chatbot on the website they should be able to ask “Where can I find the benefits application form print out” and the bot can reply with a link to the correct page or download.
A very practical use is being able to talk to a GPS in your car, you can ask for directions to the location of the distance left on your journey all via voice speech. This means users don’t have to take their hands off the wheel or their eyes off the road making it much safer.
Another use could be analysing all the posts in a forum, this could generate lots of useful information such as the general sentiment of what users are discussing (games / sports / clothing) their attitude (happy / sad / angry), this feedback could be used to make appropriate improvements. This could also be used to discover anyone using the forums for nefarious uses such as scamming or planning terrorism attacks (yes really!).
There are a couple big players in terms of NLP, most of them are paid services, lets cover them first. This is not a comprehensive list of all the NLP systems available just a few examples.
With the emergence of NLP there has been an explosion in the amount of companies trying to offer their software solution. Several of the big players like Microsoft and Amazon have very capable systems. I would recommend if your looking to experiment and get started in NLP to use Rasa since its open source and has a lot of tutorials and guides on getting started in the world of NLP.
They are usually trained for specific uses so will not be able to handle anything they are not already expecting. For example, if you try asking the ASOS chatbot bot for directions to Leicester, you’re probably not going to get a satisfactory answer.
They also need to be trained for specific languages so if they need to be accessed in an alternate language they will have to be entirely retrained for it.
One issue I encountered when developing a chatbot for a website designed for British Muslims looking to regain their faith was the issue of dual language. I found that often British Muslims will use a mix of English and Islamic words and this caused confusion in the processing. The solution to this was to use a more generic AI to process the input which didn’t rely on knowledge of a specific language but instead used traditional AI techniques of comparing raw characters and words with a very large training data set in order to best match up the user input to a defined intent.
NLP can take a large amount of processing power, training the model to process the inputs can take some time depending on the complexity and the amount of training data. Requesting the model to do the processing on the input can also take a lot of processing power but nowhere near as much as the initial model generation. While this in preventive it is something to consider when developing an NLP system.
When it does go it wrong it can fail spectacularly, for example “how I can I deal with bullies” could be interpreted as the user wants to sell their bike to an bullie rather than how the user should act when confronted by a bully.
A key aspect of the NLP models and technology is that its constantly being improved. As people use and play with the NLP services available, they are being tracked and when a model predicts something correctly or incorrectly, this is being logged down and can be included as training data in the next iteration. As time goes on the NLP services as well as the models we are training are going to get better and better at predicting our language.
As the software and methods for NLP improve over time we will be able to use it in more sensitive areas for example someone’s entire medical history could be processed by NLP and the important and relevant points to the task at hand can all be surmised to a doctor.
We don’t know all the areas we can implement NLP yet but as the world’s data is consistently increasing in quantity (expected to double every two years), we will need to find methods of processing this data in an efficient and error prone manner, NLP could be the solution.
By 2020, it is predicted that 85% of consumer interactions will be handled without a human agent (Chatbots Life, 2019).
67% of US millennials said they are likely to purchase products and services from brands using a chatbot (Chatbots Magazine, 2018).
With four key speakers from various industries, including JustEat, there was plenty to take away from last week’s Tech Talent Forum hosted by Makers Academy.
Just off Commercial Street near trendy Brick Lane, Makers nurtures technical talent and partners with global clients to help with the hiring process and placement.
Focussing on creating a culture of learning, the evening was sure to be interesting for anyone from a technical background, but also as a manager, improving culture is an ongoing item raised at monthly meetings.
In all honesty, I wasn’t sure what to expect from the event. A learning culture can be tricky to get right when faced with ongoing client requirements and project deadlines, however I’m glad to say I’ve taken away some key thoughts that we will aim to implement across our business. It can’t hurt to try, right?
A-ctivity – Inherited attributes that are difficult to teach
S-kill – Needed but can be taught
K-nowledge – Required but will grow
In short, creating a good culture starts with a good hire. We know this already, but what is that in reality?
Looking for a team member that has an abundance of skills and knowledge, is not necessarily going to add to the culture. The third piece of the puzzle is Activity and inherited skills. Is this person going to be difficult to manage or do they already have the attributes to integrate smoothly with the team?
A bold statement which will ring a bell with many hiring managers. How we define a problem is down to the individual company but in a nutshell, Jeremy Burns (News UK) described this as a team of perfectly lined bowling pins being knocked over by the ball. Disruption to the team can have positive and negative effects, usually the latter.
Instinctively we don’t like change. Occasionally we need to throw a curveball at the team but in a positive way. A disruptive addition to the team can have serious consequences. Follow your gut instinct and watch out for a lot of “I, I, I” and also a lot of “we, we, we” answers when interviewing. A balance in the middle is usually a good start.
Kate Richardson from Just Eat spoke about how her team have reviewed job titles and aimed to demystify senior positions by introducing tiered mentoring within her team. Allocated time and tasks are distributed and solutions are thought through and tested.
In theory this sounds great, but in agency land, we tend to require the team to utilise 90% of their time on billable work. The other 10% should be spent on personal development but striking a balance between client v internal tasks makes this sometimes impossible.
Institutions are failing our teams. This was the view of Fred Scholldorf, (Global Relay) who stated that it’s a digital world and still our education institutions are not prioritising this in the way they teach. In the technical world, teaching students how they learn opposed to a specific subject, will nurture a more useful skill set for budding developers.
Strange huh? Well, Makers believe that every error provides a learning experience and now it makes sense. “Why did this happen and how can we avoid this in the future?” A normal reflection we have when something goes wrong. Pointless if we don’t learn from it.
Build, learn, improve. Don’t engage with an error at a concrete level, the black and white, right and wrong way. Pass on your knowledge of how to learn a process to fix the error and your team will enhance their skill set along the way. This applies to all roles within your company.
Give a developer a solution and he’ll apply it, teach a developer the process of finding a solution and he’ll always (eventually) complete the puzzle…
In summary, there were a lot of points raised that I already see in practice here at Den, but with an ever-growing pool of skilled resources and varying skill sets, consistent communication and knowledge sharing is first on the list to get right.