Why AI is more than cheap Prediction…its equally about self-organising dynamic workflows


I recently read the article The Simple Economics of Machine Intelligenceand the book Prediction Machines. It got me to think – the technology geek in me was of-course shocked at the over-simplification. And while I accept that the current state of AI is nowhere close to the myth of super intelligent machines, I still think that the reality is somewhere between the two extremes.

Five years ago, it was all about data and insights. Data mining algorithms were designed to determine patterns, correlations and convert data into insights and advanced machine learning algorithm predicted value (through ranking, relevance etc.) and recommended actions (using classification, context and value). Quality of Prediction was highly dependent on the richness and relevance of data and the buzz was about transforming data into informationand ultimately improving the decision making process. Better insights improved the context, and as the accuracy of predictions started to improve, it lead to more holistic perspectives– all this started slowly to make the difference, and the conversations gradually shifted to predictions.

If you sift through the AI hype, most of the products today sit in this space.  No wonder the economists are reframing AI as cheap prediction. They argue – the economic shift will centre around a drop in the cost of prediction (and an associated increase in the value of judgement) – which will in turn lower the cost of activities that were historically prediction oriented (like inventory managements, demand forecasting), as well as open up opportunities to use prediction as an input for things for which we never previously did (like navigation which now has moved out of the rule based programming framework to a machine learning based prediction framework that predicts human response to changing environmental data).

But limiting the view to predictions (and keeping actions triggered on judgement) out of the purview seems like an incomplete and a shortsighted view. It is like saying that we will replace bits and piecesof our existing processes by newer technology. It’s an oversimplification of AI and risks losing the bigger picture – the opportunity to reimagine business processes – not just with digital intelligence – but also through self-learning and self-organising workflows.

Of-course it is true that now prediction is becoming a commodity. It means that morecompanies can make betterdecisions, and fastertoo. But decisions do not necessarily translate into actions.

Some argue that automation of workflows is helping to achieve this. I hold a slightly different view.  Workflow automation has been practiced for decades now – and even I agree with reasonably good results.  But these workflows are not easy to setup. The design of a workflow is a slow, definitive process and requires careful planning and understanding of all possible scenarios and outcomes. This imposes a structure and boundary to the workflow and limits the flexibility. Changes to the workflow are slow to introduce and require both re-design and re-implementation followed by re-training. This often makes the workflows out-of-date in todays fast changing dynamics.

In the age of cheap prediction, new scenarios and outcomes are constantly discovered. Even if these predictions are fed into automated workflows the actions and outcomes are confined to a pre-determined set – the set that the designers anticipated as the range of scenarios and implementers programmed as rule based decision algorithms or multitude of if-then-else type statements. Conventional approach to workflow implementation and automation is simply too costly, low on productivity and offers limited flexibility to adapt to changing needs.

This is where I believe AI will have the biggest impact – the potential to take the predictions (or even simply data or insights) and change the workflow dynamically, the possibility to use the prediction to not just determine existing action paths but to learn new action paths and set them up automatically (by-passing the slow painful process of large IT projects with their 6-12 month cycles of consultancy, design, implementation, integration, staging and deployment).

AI has the potential to introduce this fantasy – the ability to dynamically create, manage and execute workflows – a paradigm shift into a constantly evolving and dynamic workplace. That is the vision that I am working towards and our early success makes me an optimist – our use of machine learning algorithms today allows us to correlate events with dynamic storyboards (thus setting up the full context), and with a simple application of prediction we are mapping intents to a multitude of actions. This is the first step away from the realm of data and insights into the world of real time actions and our machine learning driven framework is slowly evolving into a learning framework of self-organising and dynamic workflows and action paths.

AI will have a profound impact on the workplace – be it revenue or cost or customer experience – it will be more far reaching than the cheap prediction scenario envisions. The opportunity stands before us to turn business models upside down, to create new categories and overturn existing market structures. I can’t say if it will change everything, but it sure can change a lot of things. People and organisations that accept this and participate (beyond the hype paradigm), i.e. for whom AI is not simply a badge of innovation, will reign as future leaders.

And if you still do not believe me, take a moment to look at Amazon– AI is central to their workflow even today – some of this is highly visible (and an edge for the future), such as the autonomous Prime Air delivery drones; the Amazon Go convenience store that uses machine vision to eliminate checkout lines, but then there’s so much more behind the scenes – some to drive more traditional  predictions for process optimisations like demand forecasting, product search ranking, recommendations (product, deals), detecting anomalies and exception handling (e.g. frauds, asset management) and others that are innovating business models and personalising customer service… and many more that we are still to hear of!

This article was first published on @LinkedIn October 23, 2018.


When the workflow becomes a competitive advantage…

Is it a missed opportunity?

I find it fascinating that very few business leaders talks of workflows or processes in their business updates. It is rarely a subject of discussion in strategic planning and board meetings. The leaders are engaged (and often obsessed) with visions and milestones and the mantle of setting up the day-to-day working system (to realise those goals) is left for others to work out.

Even in today’s world where digitisation and transformation are buzz words, most leaders are following the hype curve and adopting the processes, tools, technologies that they see around them. Induction of new technology – digital, mobile, automation or AI – is seen as a checklist, something that has to be done to present as a badge of innovation. In most cases it seems to be viewed with the lens of cost reduction and the first instinct is to explore automation of existing workflows and processes; very few organisations invest in building it up as a competitive advantage or an opportunity to innovate customer experience.

And yet, every now and then we come across companies that surprise us with the power of process innovation. I was very excited when I read of an anecdote related by Hans Roslingin his book Factfulness– Hans tells us that he was invited by UNICEF to investigate a bid for a contract of malaria tablets to Angola – an unbelievably low bid by a small family business (Rivopharm) based on Lugano in the Swiss Alps. The bid caught everyone’s attention, as the price was lower than the cost of the raw materials.Hans visited the factory to inspect and came back highly impressed. They had achieved this impossible task of delivering a cheap solution through two unique strategies – first they had used technology to automate their process(which allowed them to manufacture and prepare the shipment in just 3 days); and second they had converted their payment terms into a business advantage– where UNICEF payment was received within four days, the raw material supplier gave them a credit of 30 days – and they simply routed the interest earned on 26 days to pass the benefit back and cut down the prices! This is a perfect case study to demonstrate the power of a process – any process – be it manufacturing, development, logistics or business.

If you are still not convinced and somewhat skeptical that this may work on a small scale but is not a practical approach for a large or complex business, then there is no better example today than Amazon– even without having the benefit of internal insights – I can say with conviction that they owe their success to process and technology. They have oriented themselves to let technology drive their processes(and not the other way around) – and that has made all the difference. Whether it is the use of technology, robots or AI – its central to their process. Some of this is highly visible (and edge for the future), such as the autonomous Prime Air delivery drones; the Amazon Go convenience store that uses machine vision to eliminate checkout lines, but others are like the hidden layers (of a neural network) that run their magic of optimisation and deliver predictions with the least cost solution – be it the machine learning algorithms that drive demand forecasting, product search ranking, recommendations (product, deals), fraud detection etc. or other automations that enable Amazon to improve core operations.

It has been seen – again and again – that the workflow (or process) is a leading indicator of success. And yet, it is always relegated to the side – why? Is it because it falls way behind in hype and glamour? It surely is a missed opportunity, as it seems to hold the edge for the future– especially with the growing applications of AI, predictions and automation.

The workflows in the future will need to be dynamic and adaptive to constantly changing inputs and predictions. But we have to be careful not to force fit new technologies into existing processes – that will only return incremental improvement. Instead, we should view technology and process innovation as a tool to create a step change.

Technology driven Innovative processes will turn out to be the biggest competitive advantage in the new era of AI driven automation. But a word of caution – even as we design and deploy these new processes, we should never forget that process is not a proxy for results… we should always stay vigilant – constantly reviewing the outcomes and adapting the processes…

This article was first published @LinkedIn on August 13, 2018.

Shifting gears… driving all efforts to maximise every client interaction…


Be it optimising processes or driving revenue growth, the key is focus on the client conversation!

It all started – many years back – with sales force automation which laid the foundation for companies to drive their sales processes. The challenge was daunting – managing opportunities, accounts, contacts and pipelines for sales forecasting, and the goal of sales force automation was to improve operational efficiencies. Over time, many peripheral sales processes such as lead management, quote management, territory management, strategic account management and partner relationship management were also integrated to build a single platform for managing the entire client journey. Recent innovation is looking to add predictive analytics for opportunities, business process modeling, sales-methodology overlay and vertical specific products to further optimize workflows.

But removing operating inefficiencies can rarely – on its own – deliver revenue growth; and this realisation shifted focus to sales efficiency (mostly under the head of performance or productivity) and brought in initiatives to develop sales capability. It led to the emergence of new tools and processes – all within the SFA environment – starting with sales content management (creation, curation, and analytics for usage and influence), feeding it into onboarding, coaching, role-play, training and slowly extending to applying this knowledge to specific sales situations with curated account intelligence, just-in-time training, recommended sales content and many similar functions. Increased experimentation with such tools can be seen under broad initiatives of sales enablement or sales excellence (or even the not so fancy old-but-still-relevant sales training).

It’s quite fascinating to watch these innovations. It sounds so cool, so intelligent – a system that can dynamically deliver contextual content to me, that too just when I need it. However there is one big catch. The efficacy of these targeted push solutions is highly dependent on the quality of data that exists in the system – the data needs to be accurate and timely to build the contexts and identify the co-relations, else it will end up delivering static set of pre-defined resources or running predictive analytics on an incomplete or even worse irrelevant data set. And this is where most of the current systems start to fail to deliver on the promise.

In this scenario, I am reminded of a lesson I learnt (as always the hard way) in the early days of my career – a lesson I have never been able to forget – that you can be highly efficient but continue to do the wrong thing (!) and that the measure can never be actions but has to be only-and-only outcomes.

The focus really needs to shift from efficiency to effectiveness… You can argue that many of the sales enablement solutions are designed to improve effectiveness as well… and it may even be true for simple out-of-the-box solutions where you need to deliver consistent messaging about what you do and what you offer and deal with a set of known objections. However, my experience has been that in today’s complex business scenarios, selling is not about talking of what you can offer but instead understanding the client’s pain points and challenges and often extrapolating the implicit needs from those conversations. Solutions are not static and a solution that fits one set of conditions may not be effective in another.

The key to effectiveness is to understand the situation and ask the right set of questions at the right time. It’s a challenge not just to know what to ask but to figure out what is more important and relevant in the current context.

With growing sales cycles and increased cost-of-sales, the need is to maximise every client interaction. But there is little support from the environment, which continues to be weak on managing and driving client conversations as also extracting actionable intelligence from sales interactions.

It is not hard to understand the real problem – all these systems are designed for management and there is very little for the person in the field to gain – a person whose life revolves around clients – he meets new clients and stays in touch with old clients, building relationships, understanding their needs, finding about their pain points, talking about solutions, understanding other options and in the process collecting so much of data and bits-and-pieces of market action, competition strategy and emerging innovative alternatives. All this real intelligence just sits in people’s heads and never gets shared into the systems. If it did – it would open up a plethora of options on which to respond and stimulate business decisions that can transform the win cycles.

The crux then is to capture the client conversations – accurately and in-time. This will bring the larger organization into the decision making circle and also drive these next-gen intelligent systems to drive the cycle of contextual discoveries.

Accurate and timely capture drives contextual discovery, which in turn drives dynamic responses and maximises client conversations.

But the big question – how? Hardly anyone has been able to drive client conversations into the CRM.

Arti is the co-founder of humanLearning – a fast growing UK-based technology startup – setup with an earnest desire to make the life of busy professionals simpler and more effective. hL is disrupting business workflows thru short structured and searchable videos and event-driven SaaS. Arti can be reached at arti@humanlearning.com. Arti has spent a major part of her working life in the field trying to create new opportunities in the leading edge high-technology complex solution space and used this experience to design their new platform vyn.

vyn, a new tool, tries to address the challenge of feeding client conversations into CRM by embedding the way of selling into daily action. Centered on client conversations, it asks the right people, the right questions at the right time. Structured storyboards ensure that no important aspect is missed out and by making sure that one speaks for less than a minute, it keeps it just to the point. With short, storyboarded videos, it is now easy to capture updates, insights or any other information directly into the CRM – wherever one may be – and allow all stakeholders to participate and contribute to moving the conversation forward.

But where the tool really scores is that it slowly starts to assist the sales person in daily chores… its like a super efficient virtual assistant – that mostly leaves you alone but in between sends a few prompts – triggered on key events in the client journey – these provide timely reminders for urgent actions and make sure that you don’t miss anything. Even better, when you need it most, it guides you with simple storyboards at every step and ensures that you are fully prepared with the right set of questions and answers at the right time, and can maximise every client conversation. And of-course it makes life simpler…

[This article was first published @LinkedIn on 03 March 2017.]

What if sales transformation could become just another every-day project and not be a big managed initiative…?

Is this just a dream…? What if a CRM system automatically prompts me at the right time and guides me with a simple (not more than 2-3 lines) template, telling me what to do… I will always on top of the client journey! I will always know what to do, how to do and when to do…

on top of the client journey

Sales Transformation is more than deploying of new sales methodologies… its really about changing sales behaviour… and driving sustainable change… far bigger a challenge than most realise!

In recent client interactions, I have found many clients investing in expensive sales methodology trainings, setting up new workflows, building checks and balances… driving their teams to comply to newly deployed systems or processes…

There are many different sales methodologies in use – from the more widely known ones like Miller- Heiman’s Strategic Selling, SPIN Selling, Target Account Selling (TAS) etc., to a few that have their own niche following such as MEDDIC, The Challenger Sale (TCS, a successor to SPIN), Value Selling Framework, Solution Selling, SNAP Selling, Sandler, CustomerCentric Selling etc. All of these require the sales teams to adopt a structured approach to the sales cycle… expect them to get acquainted with complex decision matrices, question checklists and develop the capability to select from a vast suite of alternative scenarios…

Lots of investment in time, resources and processes is done by organisations (big and small), and yet the returns fail to match the expectations or the potential – and it always takes longer than projected! This is not from lack of commitment or motivation from the management, nor from selection of incomplete or ineffective methodologies – for each methodology has years of study, research and results that have led to its formulation and evolution. While, some methodologies focus on communication and messaging, others give more attention to important aspects of the sales process like discovery, qualification, decision criterion, decision process, and a few also cover account mapping, identifying champions and developing effective relationship maps.

The biggest stumbling block in effectively embedding these methodologies (or for that matter any new process or workflow) is the standard deployment approach – a few days of targeted intense workshop training (with tons of reference documentation to read and refer), followed by managed introduction of the new process, and setting up internal programs to drive adoption and compliance of the workflow through monitoring, tracking and regulating actions.

The reality is that however good the methodology and the supporting frameworks or worksheets, it is one thing to know what is required, a completely different thing to overcome inertia and an even a bigger ask to develop the capability to select from a vast range of options to put that into action. It is very natural to unintentionally fall back into the comfortable and tested mode of operation… and even the most effective of all monitoring and tracking is limited in effectiveness to simply managing and rewarding activity and very rarely get further down into quality or tangible results.

My personal experience has shown that the key to success is to try to embed the new methodologies into every day action. Do not simply rely on the quarterly, monthly or weekly reviews of plans, strategy or approach – big scary excel sheets, fancy blue sheets or other frameworks – as these are offline snapshot activities (and after all reviews!)… Instead enable every update, every action to automatically answer the questions and capture the information relevant to the stage of the client journey… Of-course this is easier said than done… but is achievable by supporting human intelligence with elements of collaborative intelligence and machine intelligence… support the user by giving him prompts – at the right time of the client journey – that guide him to the relevant questions… (Believe me – all sales users will love the fact that there is no need to remember the questions or the criterion or the complex matrices, or refer to offline reference tutorial to understand the definitions of confusing terminology or fill up documentation and worksheets!).

Remember that these prompts and questions cannot be big word documents or excel worksheets – that will turn off all of us… given our attention span 2-3 lines describing the question will work best… what if don’t even have to write down the answers and can instead capture the response as a short video? Well! We have our new template or worksheet – a short structured video! Where every question is just a segment in the video, and as the question is prompted we answer the question on video and move on to the next question… so a series of 3-4 questions gets recorded as a short structured video… just imagine the simplicity over all the documentation!

Is this doable? What does this require?

  • Take the client journey and map it to our own definitive stages – stages that we have identified for our custom workflow (say discovery, definition and scoping, qualification, positioning and validation, negotiation, closure etc.)
  • For each stage, building upon the methodology in use, list down the applicable questions and qualifiers (e.g. Is this an opportunity? Can we compete? Can we win? Is it worth winning?)
  • Map the questions into templates. [Each template is a set of 3-4 questions (2-3 lines max) which get encoded as segment descriptors for a structured video]
  • Identify triggers that mark the start or end of different stages and associate the templates with the triggers.
  • Integrate the triggers into the sales workflow (as simple service extensions on the CRM)
  • The occurrence of an event or a trigger will result in an automatic notification to the user, feeding him with the prompt and appropriate template

So, what does it deliver? As a user, I don’t need to refer to offline documentation or associations or supporting apps and toolkits, I simply engage with the defined process of updating client interactions in the CRM system. The process automatically activates on changed triggers or new events and notifies me (the user) on specific action, prompting me with the appropriate template to ensure full coverage of the questions to progress the opportunity or reduce the risk.

The notification acts as a natural check and reminder to engage with the process and the prompts drive the quality of client conversations… The methodology gets embedded as a way of working and requires no offline effort from the user. The underlying goal of building sales capability and moving the organisation towards consultative selling gets gradually into the social fabric of the organisation.

The improved effectiveness of sales and better understanding of client needs opens up new opportunities for accelerating business.

This is not just a concept today but tested in the field through SmartVideoNotesshort, structured and searchable video bites that are embedded into client CRM systems.

SmartVideoNotes (from humanLearning) are being used by global organisations today to drive sales transformation and embed new sales behaviour through use of customized client journey templates and associations. It is also being used for digital transformation to embed technology and innovation into everyday actions.

This article was published @LinkedIn on August 05, 2016.