If you want to listen to it instead of reading, play the link above
When I did my MBA, I remember we had a class called personal development. It was based on Eric Berne transactional analysis and involved setting your medium term goal. Mine was to become the best intelligent automation developer.
A good intelligent automation developer is one that develops automations that are very impactful, that get palpable results in shortest time possible and keeps them running. Therefore, it involves having a great idea, and then the discipline to keep the bot running. And then it involves tracking the results, and assessing it periodically to gauge if the impact is still worth the support effort.
My first project was one which involved sending mass emails. As a blogger, I sent people I did not know my articles. I remember I have taken a list of first names and a list of last names, and I have built gmail address by adding a “.” between any combination of first and last name possible and I have sent my articles via email to all these addresses.
My second project, based on the now defunct Internet Explorer browser, was a web scrape of the biggest Romanian online newspaper, where I kept a blog for a few years, of all the blog articles titles and authors of all time. I am not sure if I did scrape all the articles in the end, probably not, but I have still a database of over 17000 articles, with their URL, author, title. As a fun fact, I did now a sentiment analysis on the titles of these Romanian language articles, and the sentiment is negative with acceptable 68% confidence. Are Romanians negative people or was the sentiment analyzer inaccurate? Also, I did a histogram of the authors, and the two most prolific authors were a theather critic turned into society commentator and a journalist whom I suspect is part of the Romanian intelligence community, both of them very old guys, and who were always ranked by the newspaper upper than me or any other blogger, therefore leading to my disatisfaction with the blog and turning away to a more lucrative way of spending my time. Meanwhile the blog section of the online newspaper was quasi overhauled, to my delight.
The third project was the one which you can still find here on the blog, I tried to map all the intelligent automation jobs, therefore jobs in my field, by collecting a list of 350 big companies with a careers page, and going to all these pages and extracting how many intelligent automation jobs are there. In this manner, I could capture the trends in hiring in my field. This is still valuable today, as I can basically run them again, and see how many job are there by keyword/job title. It is basically a labor market analyser if you think about it.
As other side projects, I remember some Amazon.com bot that collected the number of political science books, other passion of mine, by title, author. And another one, which still has value today, is to extact the apartments from a renting platform to analyze price trends in your city.
You can use your bot machine to create a bot which extracts value from events. One idea that came to my mind after I went on a Milan trip when the Milan Fashion Week September 2023 started is to find out which of the tens of fashions events that have also men fashion is the most popular and also affordable, because these brands can cost a lot.
Therefore I created a bot would go to Invidious, aka YouTube without ads, and extract from the FF channel, all the data regarding all the fashions shows. Number of views and number of comments.
“Fendi | Spring Summer 2024 by Kim Jones | Full Fashion Show in High Definition. (Widescreen – Exclusive Video/1080p – MFW/MilanFashion Week) #Fendi #SS24 #MFW #FFLoved Blair Booth – Become Like Satin/Poison”
Tom Ford | Spring Summer 2024 by Peter Hawkings | Full Fashion Show in High Definition. (Widescreen – Exclusive Video/1080p – MFW/MilanFashion Week) #Tomford #SS24 #MFW #FFLikedalot
“Genny | Spring Summer 2024 by Sara Cavazza Facchini | Full Fashion Show in High Definition. (Widescreen – Exclusive Video/1080p – MFW/MilanFashion Week) #Genny #SS24 #MFW William Davies – Twilight Breeze/Dreaming Out Of Time/Rest is Noise”
“Nº21/Numero Ventuno | Spring Summer 2024 by Alessandro dell’Acqua | Full Fashion Show in High Definition. (Widescreen – Exclusive Video/1080p – MFW/MilanFashion Week) #N21 #SS24 #MFW #FFLoved Lotty Lush – You Want My Cyril Giroux – Il Tuoi Baci Alex Arcoleo – I Go”
“Aniye Records | Spring Summer 2024 by Alessandra Marchi | Full Fashion Show in High Definition. (Widescreen – Exclusive Video/1080p – MFW/MilanFashion Week) #Aniyerecords #SS24 #MFW Bethany Drewien – Blistering/Wild Thing”
95
A few rows of the data scraped by the bot from Invidious FF channel
Then extract how many of these had men fashion, because most did not. Sorry men who follow fashion. This step I will do now manually becasue I do not have a face recognition AI skill to do. If there is anyone who knows how to do this with UiPath for free, please feel free to contact me on.
After filtering the fashions houses that had men fashion as well, we are left with 10 fashion houses out of the 51 which I could extract from the FF chanell. Only 22% of the houses have men fashion. Which is pretty low for men, who are more preoccupied with losing their money in super corrupt sports watching, alcohol drinking and sports betting then on a such a nice activity as fashion following.
Let’s see how the table looks now. You will see I’ve added a column with points, just to keep track of the score.
Name
Number of Views in k on 11.10.2023
Number of Comments on 11.10.2023
Ratio comments per k views
Points
Bottega Veneta | Spring Summer 2024 | Full Show
69
99
1.4
8
Ferragamo | Spring Summer 2024 | Full Show
57
48
0.8
7
Jil Sander | Spring Summer 2024 | Full Show
34
38
1.1
6
Diesel | Spring Summer 2024 | Full Show
29
54
1.9
5
Benetton | Spring Summer 2024 | Full Show
13
17
1.3
4
Philipp Plein | Spring Summer 2024 | Full Show
11
26
2.4
3
Bally | Spring Summer 2024 | Full Show
7.4
13
1.8
2
Iceberg | Spring Summer 2024 | Full Show
4.6
5
1.1
1
Data only for shows that also had men fashion sorted by number of views at 11.10.2023 of FF channel
Name
Number of views in k on 11.10.2023
Number of Comments on 11.10.2023
Ration comments per k views
Points
Philipp Plein
11
26
2.4
8
Diesel
29
54
1.9
7
Bally
7.4
13
1.8
6
Bottega Veneta
69
99
1.4
5
Benetton
13
17
1.3
4
Jil Sander
34
38
1.1
3
Iceberg
4.6
5
1.1
2
Ferragamo
57
48
0.84
1
Same table but sorted by the ratio comments per k views
I will try to create a score for the best fashion show in Milan September 2023 Fashion Week in which there was also men fashion. A quarter of this score of given by the number of views, another quarter by the comments to views ratio, the third quarter is given by the average price of a dress kit (shoes, pants, jacket) on their online shop, and the last quarter by the sentiment of the comments on the FF chancel of their MIlan Fashion Week September 2023 show.
To get the average price, I go to the e-commerce platform of these brands and compute an average price for a jacket, a trouser and a shoe.
And then split the brands into three categories, budget, mid range and luxury.
Name
Avg. cost jacket
Avg. cost trouser
Avg. cost shoe
Avg. total cost
Category
Points
Benetton
140 EUR
70 EUR
80 EUR
290 EUR
Budget
8
Diesel
545 EUR
200 EUR
575 EUR
1320 EUR
Budget
7
Iceberg
950 EUR
300 EUR
270 EUR
1520 EUR
Mid
6
Bally
1100 EUR
700 EUR
500 EUR
2300 EUR
Mid
5
Ferragamo
1550 EUR
620 EUR
845 EUR
3015 EUR
Mid
4
Jil Sander
3150 EUR
670 EUR
850 EUR
4670 EUR
Luxury
3
Bottega Veneta
3000 EUR
1100 EUR
790 EUR
4890 EUR
Luxury
2
Philipp Plein
3100 EUR
1200 EUR
1500 EUR
5800 EUR
Luxury
1
Data on prices and men fashion brands caterorizatio into budget, mid, and luxury
So, far the top three men fashion brands, before the final quarter, the comments sentiment is: 1. Diesel with 21 points, and Bottega Veneta and Benetton, both with 15 points.
What else is left? Do sentiment analysis in the comments of each fashion show. Most people post positive comments, but the score will give us the whole picture.
For sentiment analysis I have struggled a bit to find a free service. Google asks for money, also UiPath. But I have found one free on a Google search. Monkeylearn.com, thanks for the opportunity.
After analyzing all the coments in the shows, the most positive was Jil Sander, second Ferragamo and third Diesel.
Name
Sentiment
Points
Jil Sander
99.2
8
Ferragamo
99.1
7
Diesel
98
6
Benetton
96
5
Iceberg
92
4
Bottega Veneta
82
3
Bally
56
2
Philipp Plein
47
1
Sentiment analysis of the comments of the Milan Fashion Week shows with MonkeyLearn.com
Now we have the most affordable and popular men luxury fashion brands that will dress you like a champion for Spring and Summer of 2024.
Name
Final score
Stars
Category
Diesel
27
***
Budget
Benetton
20
**
Budget
Jil Sander
19
*
Luxury
Bottega Veneta
18
Luxury
Ferragamo
18
Mid
Bally
15
Mid
Philipp Plein
14
Luxury
Iceberg
14
Mid
Final score table
Therefore, the best fashion collection that contains also a men collection is the budget category Diesel. I recommend you to take a look at their online shop. Next is also the budget category Benetton. Please take a look at the new arrivals. And the top three is ending with luxury brand Jil Sander. We need to give a special applause to Jil Sander for making in in top three because my analysis is heavily skewed towards budget brands, because my top brands should be most of all affordable. But Jil Sander, which is not a budget brand, made it to the top despite very not affordable.
That’s it, here you have it! This is the only data based fashion journalism blog there is on the Internet with real data, guaranteed!
You can now repeat this process for every fashion quarter and location, year in, year out. Thanks to intelligent automation and the data on the FF channel (thanks a lot!), now an effort to compute all these will go towards zero.
It took me a few days to put this first blog post in order. But the next one will take much shorter as this process is massively automated already. The things I did manual were: checking if there is men fashion in their shows, formatting of the data and inputting these data in the tables here in the blog post.
Think about what you’re using now the PC for. Maybe you just input data manually from your human memory into a social media platform. Or you just extract data from a platform, aka consume, and store it in your human memory to form, share and shape political identity and cultural opinions.
This exchange above is the simplest exchange there is, nothing automated here.
Let’s try to automate some of the tasks of this exchange, one of the most important exchange there is in the life of a person from our Internet generation, from any place on Earth and who owns a laptop and speaks English.
What tasks can you do, on Facebook, for example? Few tasks come immediately in mind:
Automatic liking posts and / or commenting them
Mass messaging friends
Extraction of information from your friends posts
Next, we get to more than one application and add a human in between. Let’s say you want to find the coolest article on the LinkedIn based on a topic that you extracted from your friend’s post, and send it to your friend via message on Facebook.
But before sending it, you might want to ask the bot to send it to you for validation. Does the post qualify for a good impactful post that can be sent to your friends or it’s a shit? If it is a shit, stop there because annoying posts are actually costing you rather than creating wealth for you, which a good post can do for you.
Here, again, you get some limitations from Facebook on how many of your friends you can message a day (last time I checked was 150), but this doesn’t matter. The simple automation, will now look like this: automation bot searches a platform for content, but do not search on the main algorithm feed; instead, use a list of contacts that you personally know or trust, for example 300 contacts, and search on their personal feed for content that you can borrow, then send the content to you for validation and then search on another platform for the added stuff; so, for example, if you will send a video hosted on LinkedIn, then an impactful text related to the topic should be added; for example, the topic can be found easily if you scan the text of the post from where you borrowed it, and then you search for quotes about that topic on another quotes platform, for example GoodReads platforn. And there you go. You are now a posts automation bot or machine who can create conversations with your friends, without the emotions you can get while doing all this manually and without the energy you put when you do this by yourself.
But why would you do this? Why would James Watt invent an engine that uses steam to produce some energy? For the same reason, to create energy! And all the better if it is automated!
Now you get the power of intelligent automation?
From here, you can go expand the automation, for example you add a call to action if the post you send is about a societal problem, so then you add a third step in the automation, for example, search for petitions about that topic in another platform, for example change.org.
But the other direction in which you can go is to add another app. So then you find another thing, and only then you send it via Facebook Messenger to a long list of friends.
Imagine a post like this. A great poem on the top of a mysterious forest picture, for example the UNESCO World Heritage Bialowieza forest:
Know then thyself, presume not God to scan
The proper study of Mankind is Man.
Placed on this isthmus of a middle state,
A Being darkly wise, and rudely great:
With too much knowledge for the Sceptic side,
With too much weakness for the Stoic’s pride,
He hangs between; in doubt to act, or rest;
In doubt to deem himself a God, or Beast;
In doubt his mind and body to prefer;
Born but to die, and reas’ning but to err;
Whether he thinks to little, or too much;
Chaos of Thought and Passion, all confus’d;
Still by himself, abus’d or disabus’d;
Created half to rise and half to fall;
Great Lord of all things, yet a prey to all,
Sole judge of truth, in endless error hurl’d;
The glory, jest and riddle of the world.
The Riddle of the World by Alexander Pope
Natural coniferous stand of Landscape Reserve in morning with sunlight entering, Bialowieza Forest,Poland,Europe
If someone sent me a post like this, I would be inspired by it. I would dream a bit about life, meaningful or meaningless as it is. I would research the Bialowieza forest and I would find out that it is the last forest in Europe which was part of a giant forest which covered the continent. I would research Alexander Pope and find new pems by him. And so on. So the friend who sends me something like this all of a sudden grown in my perceptions He’s someone that can give you energy, not suck it. Someone that has a certain sensibility.
Everyone right now sends and receives posts like this on Facebook Messenger, on Whatsapp, LinkedIn or other messaging or social media platforms. This is a new form of status and wealth exchange, and can be automated easily with the help of intelligent automation. It is just a matter of how you compute the automation to give you maximum impact. Your friends will love you if you send them posts that they love, but hate you if you send them stuff that give them absolutely no value. You cannot make all of them happy, but all of them like something nice. So the intelligence of your automation is how you code the algorithm to automatically find the most impactful posts.
While computers are now used for personal emotions with the human body being the active mediator and interactive medium of inputting and outputting information which creates emotions, sometimes negative, sometimes positive, in the future I envision that automation will mostly be used to enhance wealth and popularity or status, with people using it less for the emotions, and more for wealth and status creation.
This was just an example of how you can create wealth and status automatically with the help of a well used computer. Next post I will give you another idea of how to use your brain and computing power to create value.
In a recent post of one of my RPA community colleagues, a traditional IT developer who pivoted his RPA company to an API company, told us to use API over RPA, whenever API exists. This is like saying always take the TGV if you go from Paris to London. This obviously caused me to search for examples where this is simply not true. As a person who is in the middle of this political battle, I have plenty of examples.
But before my example, I think it is very useful to paraphrase the ex-UiPath evangelist Guy Kirkwood, now retired, who told this simple analogy: RPA is like a regional train (he might have said car, but I adapted the example), while API is like a high-speed train. API is to be used whenever you want to go from point A to point B, fast. Let’s say you want to go from Paris to London. You do not need to stop in between, you are in a hurry and do not want to admire any landscape. Then, yes, that is a great choice, if you can afford the high price of the TGV. But if you want to make a stop in Belgium, to pick up a friend before going to London for another friend’s bachelor’s party, or you want to pick your father from Lille before going to see your sister in London, then the high speed train is not serving your purpose, and you will take a couple of regional trains that offer the flexibility you need, even though it will take you a bit more time to your destination. This is a good example of when to use API and when to use RPA, and traditional IT simply refuses this choice, saying it’s always more advantageous to take the high-speed train, which is simply not true.
Actually, I went to check this example with real data. I went to check the revenues of TGVs in France versus the revenues of the regional trains, and guess what I found. The regional trains have by far a higher revenue than the TGV, which is logical and obvious, except for the traditional IT who still thinks the TGV would be the only option possible moving from point A to point B. Even when it comes to job creation or profitability, API (TGV) cannot beat RPA (Regional trains), and this is something we see in automation or in transporation.
First column is TGV, while second column is Regional trains
With prices up to 10 times the cost of RPA, API based solutions are expensive. Please explain to me, as I am yet to understand, why would you pay 10 times more for something that does exactly the same thing as the other? Even though UiPath accepts that RPA has a higher maintenance cost than API, I respectfully disagree with that, as I found that you also need at least 2-4 hours monthly to maintain an API, which is in line with the cost of maintaining a mature RPA bot.
Trust me, I have done RPA since 2017 and I know what I am saying. A mature RPA bot needs at most 2-4 hours of maintenance every month, some RPA bots needing almost zero touch for as long as years. Another article I found spends some time comparing the two solutions and RPA clearly overtakes API in 4 categories (implementation complexity, time to market, cost, learning curve), while API is only superior in 1 category (maintenance). I excluded the scalability category from this comparison as I do not agree with the author of that article that you cannot scale RPA. Maybe this is a misunderstanding, that is why I excluded it.
In another article that promotes large hybrid approaches of both API and RPA when doing digital transformation, the author mentions that API is “strategic”, which cannot be more wrong. In my opinion, API cannot be “strategic” for the simple reason that RPA includes API, but API cannot include RPA. Therefore, the real strategic one is RPA.
Let’s move on. Another article gives three main advantages for RPA compared to API: being focused on a single platform, or how others call it, being platform-agnostic, the fact that is less reliant on IT experts, and mainly because of its cost advantage. Similarly, another article gives 5 reasons why RPA is better than API, among them the fact that API is risky, that it consumes more resources and time. Another opinion is more balanced and thinks a hybrid approach with both RPA and API is ideal. However, they prove the industry is now confused whether the costs of maintaining an API is less than the cost of maintaining a bot, with this article being opposed to the common belief in the industry that the cost of maintaining an API is less than the cost of maintaining a bot. They also end their view with this sentence: in case of frequent changes in the process, it is preferable to use RPA. With this sentence, it is clear that in an ever-changing environment, the use cases for API become less and less.
RPA is by far the more flexible solution. As I explained in the beginning of my article, API is useful when you really want to go fast from A to B, which is not the majority of the cases. More often, you use the RPA platform and within it call an API where it exists between two applications and can be easily used, which is not so often. A process includes some data exchanges between two systems, but the data exchanges between two systems are never the whole process.
Usually, the process starts with some insights and decision making and that should be left to the human, which can be automated with RPA. Finally, another article recognizes that in most cases RPA is preferable to an API, which is in line with what all the practitioners in the field of process automation observed. Except some people, like the one which I mentioned at the beginning of this article and who pivoted his RPA company to API. But it seems he did not read any literature, he was just misinformed by the likes of Microsoft, Apple, Facebook, SAP, Salesforce, which want RPA to disappear, but which obviously will not happen any time soon.
Unless IT starts understanding that RPA is here to revolutionize the whole office work, they will be required only to set a couple of APIs once in a while. That will be the new role of IT, together with security, which will be a far cry from its role it pretends to have today in driving business growth.
Maybe the Orchestrator platform, or in terms of software for developing these bots. UiPath solution is something that is kind of written in stone, or are there some other software that maybe are having additional features or a bit more flexibility. I don’t know this is something that is in the scope, or maybe should constrain ourselves to UiPath solution. Overall, I think that the current processes are working quite fine. Obviously, not all of them, but in many cases we have seen that it’s rather related to the currently used business processes and systems.
I think we need to gather the feedback, what went well, what not so well, then we analyze the lessons that we learned from the previous projects and it really depends on what went well or not so well. You should establish a plan to improve implementation. Which really depends, because there is no fixed answer for that. But I think the process is to learn from the past and improve.
What I have in my mind is the documentation. It’s not even who should do, but how should we do it. In my opinion, the way we do it, huge documentation in Word is not appropriate.
I think that RPA could be used to replace all the HR related things. I know that many companies, I know ours as well, are still relying on human workforce, for example to pay salaries, which isn’t bad necessarily, because a human can control all the numbers and so on, but at the other side, human can make mistakes with numbers. So, I think companies should focus on automating salary payments, or payments in general, and anything related to money could be automated, because if a robot is fed well, with details, then from that moment, it should work without any problem. And there are still humans to help. And I heard that many companies still rely on software or hardware which are from the 1970s, 1980s or even older. Because they are working fine, but you never know what happens. And such things can be done by automated robots. And I think companies should focus on such things, not improving the software, but immediately implement automation in such cases. So financial things actually.
Before going to the next step, we have to all learn from the failures of the past implementations, from the previous phases. So make sure you try to avoid these kinds of issues in the future, make sure you got everything written in a professional way, and the level of communication could be accurate and understand the business needs. There is way to implement future projects, but you have to avoid the repetitive issues appearing, and understand the rationale behind it.
Currently we are in the era of stupid bots, right? The next one is supposed to be smarter ones. We’re supposed to have smarter and smarter bots and more abilities for the bot for example. Currently the bot cannot understand human language. If our customers write an email to us, the bot cannot read it, they don’t understand. The next generation of bots is supposed to have more on AI, more Machine Learning and more Natural Language Processing. So, they can understand the email, and then we come to non-human work only.
I remember there were some chats about the optical readers. So I think that the optical reader field can open a lot of opportunities. Because we still have many processes which are based on the paper. And on the paper means on the scanned documents, for example. Having an optical reader for that it can improve a lot. And of course, I am thinking about the cash application. So having the bot able to read what this payment is for would help a lot. Because we still have manual cash application performed.
I can think only about one. It was strangely easy to build a bot, but it’s very hard to maintain a bot. So, I would go in the way where the necessary changes that are coming out from the business could be more easily applied within the process. Meaning, there are some working parameters of the bots, and they could be hardcoded within the scripts, could be exported to external config files, and a way easier how to maintain the process is when we could work on external files and the logic was clearly described. So, I would go in this place as a next step of RPA process. To make it easier to maintain, to apply any changes in the future.
What does the industry say?
According to SearchITChannel, next step is what Forrester calls “digital worker analytics”, or what has recently been called process mining or task mining. This kind of analytics is not only great for discovery, but also for the much needed standardization. Additionally, intelligence can be pursued to accommodate for the exceptional scenarios, according to the same article. Finally, an alternative direction for a part of the RPA vendors is ease-of-use.
According to RockingRobots which quotes various characters in the industry, it is various kind of intelligence that is the next step for RPA, be it document understanding or conversations AI.
According to Dynatos, combining RPA with workflow orchestration, unstructured data capture, OCR, machine learning, advanced analytics and e-signature converges into the next level of RPA, intelligent automation.
For planning, we always need to take into consideration how long it takes a bot to run. My idea will be to create multiple processes, but very simple ones. I mean, multiple automations, but simple ones. Not to create one very complex and time demanding automation, but to create multiple, simple ones. And this will help you to better place the schedule during the week, for example. And the other thing is that the processes that don’t need the results to be visible as soon as possible could be run during outside whttps://open.spotify.com/episode/6FcpwXIOswpCRSdxb26OLM?si=kAwInuWkSTawqkX7B8NJrworking hours. In the middle of the night for example. But of course, it’s different from the business expectations. In Accounts Receivable, we have some processes that we need to close them or react much, much faster than with other processes in different businesses.
So, the answer is with planning. We should plan exactly what we want. I think that we should first do a draft. And then go back and re-test. Because what I saw and what I could experiment, and I was joining the team when the RPA already started so I did not take part in the first phase, but the main prerequisites of some bots were just not valid after four, five, six, seven months. So, yes, scheduling, planning, but then go back before launching the bots and re-validate all the steps.
With a smart scheduling you can utilize more of the robot used. Less IDs, less licenses, and that’s how the scheduling can help.
First of all, to understand the logic behind each process. I guess most of our issues are coming from misunderstandings, unclear SDDs. That’s why you have wrong results. If the SDD is written quite clear and understandable, it will be easier and better for business and for IT as well. Difficult to understand code, unclear SDD, these kinds of things, should be taken care of from the beginning. To make sure that quality is assured for these kinds of documents. For SDD and for code. If it’s done well from the beginning, I believe you can save a number of issues after production.
It really depends how it’s scheduled. I mean, this is something which is mostly done by a human, so if the scheduling is done in a way that it’s not efficient, then the production won’t be efficient. It really depends on the skills of the scheduler, how it’s scheduled. And I think this is something which cannot be done via RPA or such things, these are almost soft skills like the ones we use at customer service. I don’t know how you do it in IT, but I think this is comparable to the soft skills we have at customer service. And I think this is more on the human part, again, soft skills.
I always suggest for the business to ask themselves when are they supposed to have this or that job and they can always suggest when they are doing that so that’s how you improve with scheduling. Oh yes, and you don’t have to start it manually.
There are many ways to improve efficiency by better planning. There are many examples. So, if you talk about efficiency, less cost and more outcome, if I am right about that, you can plan to utilize the resource more effectively, for example you need to plan to minimize idle time of the resource. So, when you plan to use someone’s time for the implementation, you will not want them to sit there idle or doing nothing, right? So, you need to plan carefully between the multiple resources you have in the project or the team. So that you have a good work assignment, process, so that when someone is free or have free time, they can use their idle time to something else or some more process in the queue without keeping them waiting. That can be a better task planning or assignment process, it can be Agile process, you can utilize Scrum or Kanban board to help on that so that you have visibility on people’s workload. Yeah, that’s one example.
Again, this is something related to process. If everything is mapped at the initial phase correctly and if there is a clear picture on what amount of workload it will take from a developer to start the development of this script, and what could go wrong, and at which stage of the development for testing, then it should be quite ok, and I believe these kinds of initial mappings or planning, if done in an efficient way, with close involvement of the actual people processing the manual workaround, the operational teams having to perform this and that activity, because there can be, let’s say, one key business contact that needs to communicate to the developer about twenty processes, but if it’s not the contact that runs the manual transaction or the manual workaround, there can be something in the picture that comes right after implementing the robot, and when it gets to the actual people using on a day-to-day basis that specific robot, so that’s planning in terms of IT side, but planning in case also of the business. Maybe the responsibilities here have to be very clear in terms of what is mapped by the developer side, what is mapped by the business side, so this is to be aligned to what it should be, what is actually is. I think it should be covered by both developers and business.
Yeah, it will for sure, because most of the simple, repetitive things can be replaced by the bot, and that will give employees more time to spend on harder cases, and it will for sure decrease the closure time of the simple requests that need to be done. And we can already see the effect of those.
It depends on how the automation is managed. I am not against the automation at all, not at all, but if automation means as it is, the last fashion, bots will reply to a call, of course, it won’t improve the quality of the customer service, but if automation means that we are not sending one by one the dunning letters, because a bot will do it for us, yes, it will result in an improved customer service because the customer will receive a punctual email, because at 9 o’clock the bot will send out the dunning letters. But if then the dunning letters shouldn’t have been sent, because the customer already paid, then the bot did not check the general ledger, this is a problem. So, it depends.
You’re supposed to have the transaction processed faster than human at 24/7. So the answer would be yes. I don’t know how it will be but it is yes for now.
Yes, definitely. Customer service is a service department, as you know. So if you take of some routine work, which can be done by RPA, and team will concentrate more, have more time, of course, it will have great improvement.
If automation is done correctly, definitely yes. Because automation in our case is not to replace us, but to assist us. So possibility of human error is definitely less. Because if you want to feed the robot with all the information necessary and all the process details which are necessary to work, from that moment the chance of having issues is much less than with a human. Because with a human, there is, of course, a higher failure possibility. But also human aspects, I mean, if someone is not really concentrated or not feeling well, even such things, then it might be that error happen much more, which could have a negative experience on customers. Only if you think that you are having your fiftieth issue on a day, and you can’t really think about it like it was your first one in the day, even if you have to, should do so, but normally you can’t. And this is something which is not the case with robots, because they just do the work, so, in the end, if the automation process has been implemented well, then it’s definitely improving customer service.
This is not an easy, white or black answer, because on the one hand yes this will result in improved customer service because of the faster reply and this can be really tainted by the facts that the bots are misbehaving, but I still think it’s worth to use the bots. And make some occasional mistakes, even in production. So the answer is a clear yes.
It can, in many ways, depending on how smart you use automation. For example, if you want your customers to get a faster response for simple queries or questions, you can have automation like bots or other automation tools to handle the request from the customers, maybe by chat or by calls, that’s one example. So to me, yes it can, but you need to be smart in choosing the solution.
If someone wouldn’t have to take care of activities that are taking certain amount of time regardless of it, half hour and six hours a day, if someone doesn’t have to deal with these, then there should be something else, something more advanced or more analytical that the customer service representative can spend his or her time on. So, definitely that should result in a more sophisticated customer service or improved service. On the other hand, at this point, I am not quite sure we are expecting that it will be robotics purely bringing this improvement in the service level. It’s not because of robotics not being a good solution but definitely not everything can be addressed with the same RPA solution. It again links back to knowledge retention. So, it partly covers the picture, robotics and automation, it should result in an improved quality of service. On the other hand, to be honest, at this point, I do not really see it as happening, it’s not the fault of automation, and not the fault of RPA, but something that is the human factor, I would say. So, you can have like, ten or twenty automated processes working perfectly fine without any issues, but if some other skills, not necessarily technical skills related to using the robots, but other skills, are not in place, then it’s not fine in the customer representative mindset. So, automation should result in an improvement, but it’s not the only important factor here.
What does the industry say?
According to a Capgemini study, 63% of organizations have improved their customer service with the help of intelligent automation. It has resulted in improved response times, personalized service, 24/7 availability.
Other report by TCS found that those companies which invest in intelligent automation can not only cut their support costs, can provide proactive and even predictive customer service, boost loyalty and revenues, but also gather intelligence to improve future product designs.
In my daily operations job as intelligent automation product owner, I have experienced several failures that made me question sometimes my whole career choice.
In the first example, the automation we built for our business team was very complex. I go to meet my business, as we scheduled periodical calls with this group, which was not in my city, but another European city.
Business does not show up, which I should have understood what that means, that moment, on the spot. But I did the mistake of keeping the flame of hope on. Big mistake.
I obviously went to their supervisor, asking why my business clients do not show up in the meetings we scheduled with them, for us to discuss the health of their automation. I was already thinking that my business does not show up because either 1) they are afraid of losing their jobs, 2) they simply despise us, the foreign aliens from the automation team or 3) they are just unprofessional.
Eventually, their supervisor “forced” these business contacts to behave nicely and show up in the meetings we set up for them.
While showing up in the stewardship and bot monitoring periodical meetings, my beloved business found some problems with their automation. The bot was not behaving according to their standards. The same standards that they set for the bot while in development, few months before. Second red flag I should have recorded, but I was not aware yet of what is going on.
Next thing we did for them was to put together a huge enhancement, to re-design the bots radically, giving them chances to fix the bots and make them more user-friendly and user-centered, just as they would like them to be. Complete failure.
My business was sabotaging all the efforts. We ended up in disaster. We spent huge amount of resources on this bot, and it all failed. We threw money on the window. Why?
Because this business was not motivated to use automation. They were so reckless, so careless, but at the same time only had the wit to sabotage the bots to get to the results they hoped: total failure of automation so that they can continue to do the process manually, so that they “prove” to themselves and their managers that they are “worth something”.
Overall, our company lost big time here. Why? As I explained in my previous post, lack of proper governance leads to these kinds of situations, where my automation team gets super frustrated. We all lost.
Without the people, and people can be motivated to automate only through a good governance, automation efforts fail.
The next example I will give is one in which we did not even reach the point in the previous example, it failed in testing phase. Business was so negative and emotional about the whole automation thing, but they forced themselves (or were forced by their managers, we’ll never know) to show up in meetings.
However, this business had other plan. To simply say the results of the bot are not ok, that is, pure sabotage in the testing phase. They either gave wrong data or kept lying that the test results are not ok. How do I know this? All of us in the developing team were wondering what is wrong because all results seemed ok, even by the standards the business gave initially. Plus, several months later when a new round of negotiations got in place, the bot was by miracle ready to be put in production in a few days.
Again, this business only wanted to see me and my team simply disappear. They kept lying that all test results are incorrect, and then went to their managers to complain that it is taking too much time from their precious time to give feedback to testing developer, so they would like to be left alone. Flabbergasted?
What could you tell them more? We left them in peace, thinking what did we do wrong. We did nothing wrong, they simply were against automation, were not motivated to use automation, they had the wrong mindset towards life and work, and the whole automation program did not have a proper governance to solve these deadlocks. And several months later, we went again to discuss.
In the corporation, conflicts of this nature are not solved in the courts, like in the nation-states, day-to-day lives of a citizen. There is not any court in the corporation, only the diktat of the manager. And the higher up the manager, most of the cases, the less they want to argue with their people and force them to give it another try to automate their process, and just prefer to listen to whatever their people tell them. Which most of the cases is they do not want to use the bot, and they invent some ridiculous reason which does not have any base in reality.
Managers prefer working with people than with bots, and this is a governance problem that needs to be solved in good intelligent automation programs, if a company is really serious about getting to the next level. Automation must be an all-in strategy, supported by a great governance and trainings of the people, however in reality the efforts are timid and governance does not really have coherence and proper structure. In the end, who wins is the one who shouts louder and has more influence on managers, and 99% of the cases in which business does not want to use bots, they end up doing the process manually. Simple as this.
The final example I will give is a personal one because I contributed massively to the development of that automation script. It was a very complex and key process in the corporation.
As the process was very complex, it had also a very unique design, and was failing a lot of times due to this frail design. But as much as we were thinking, we could not come up with a better design, so we needed to “swallow” the failures due to huge benefits we were reaping.
One day, the process changes and business decides not to use the bot anymore and go back to manual process. I was shocked. What were they thinking? My team is there to fix the bots, for free, and they chose not to. What was going on?
The business was so negative against bots, that they did not want the bot to be fixed. They wanted the bot to disappear. Why? Even until today, I do not find the answer. Most probably it was not doing what they were expecting. But then why didn’t they come to my team to find the right solution for them? How on earth is it possible to find economical reasons to throws tens of thousands of dollars on the window except malice and poor understanding of the situation.
Why does business prefer to do the process manually instead of automated will be a huge mystery that needs to be addressed by all of us in this industry. Why are they having the wrong attitude towards work and getting upskilled in general?
Automation should be a goal, but in reality, we see sometimes the opposite. This not only means huge losses for the corporation, because of causes that are not the scope of this article, but also self-sabotaging your own careers, because instead of upskilling and getting competitive so that you’re getting some nice raises in the future, you end up in the bottom, without marketable skills, at the mercy of management. Why do they self-sabotage is hard for me to understand.
I have been saying it for a while now, governance is the most important aspect of implementing intelligent automation. Governance is like rule of law in the nation-state. But who assures rule of law in the corporation? Because there is no police, like in the nation-state, and there is no written constitution and laws, like in the nation-state. I am not a fan of the nation-state either, I consider it a mistake, but I do not prefer totalitarian dictatorships as well. However, without governance, that is, proper process and clear rules accepted, acknowledged, and followed by most of the people working together in a corporation, the corporation is nothing but a big dictatorship at the mercy of influent older managers.
Without governance, people or leaders will rebel against bots and will demand the reversal to the manual process, because the repetitive manual process is a big part of the modern worker’s toolkit to fight his miserable anxiety. We also understand people being afraid of automation, but leaders? It must be older women leaders who are rejecting innovation the most, as some studies of innovation have shown, like UTAUT model of Venkatesh et al. Why would be older women more prone to bash and trash automation? Is it because they are more humanistic, is it because they believe more in humanism, a secular religion that rejects traditional monotheistic religions and authoritarian (prejudiced) personality, or, on the contrary, is it their traditional monotheistic religious beliefs that makes them reject innovation as the “work of devil”? We do not know.
Now, there are people who want RPA and intelligent automation destroyed. We do not know who these people are, as “they” are more groups, but we have a clue. Might be socialists who are alarmed by media, and rival traditional software developers, who feel their influence (and salary, to be honest) is decreasing.
Some businesses, who at the moment do not consider intelligent automation as strategic, prefer either to do the process manually, or prefer spending huge amount of dollars in so-called “modern” apps with APIs, so that … people will use them later. In either case, you can’t really compare their cost with costs of intelligent automation or RPA, because, at least in my view, RPA is years light ahead of these “modern” apps. I have never seen cheaper projects and ROI than in RPA, and ROI should be still king in business, at least this was true last time I checked.
Instead of investing massively in RPA and intelligent automation, what we see in these corporations who do not invent in RPA and intelligent automation product growth anymore, is a plethora of managers, most of them close to retirement, badmouthing RPA and planning to destroy this program, that is, throw hundreds of thousands of dollars worth of code on the window (and who knows how many millions of dollars in future potential unrealized savings) in which they were forced in the last years to invest by C-level people who were visionaries and prefer less cost and more productivity, and which were sabotaged by business folks stoned by the fear of losing their jobs, and in which their managers had nothing to object to this public and transparent sabotage.
The bright future that IT managers promise is a future of expensive “modern” apps supported by some kind of expensive software developers that hopefully business people will not sabotage like RPA.
How much do these “modern” apps cost? According to this site, a medium complex “modern” app costs between $80000 and $150000. How much does support cost? According to the same website, a freelancer on the market costs between $75 and $400 per hour. If you hire them, this cost can double. Then compare to how much RPA costs and RPA support costs. A medium complex bot costs between $20000 and $40000. Support costs between $35 and $55 per hour for a hired employee. For a freelancer, those number will most likely be 50% of the permanent hire. So, what does the strategic planning of getting rid of RPA and intelligent automation, and building “modern” apps that will be manually operated by people hope to achieve? Nothing but a permanent increase in cost, driven by the IT industrial complex, led by bribe-induced monopolies like Microsoft. How is that even strategic? Where does the money come from for this socialist experiment? Maybe for shared service centers this can be, but for profit centers? Profit centers do not afford to do socialist experiments.
Instead of doing these comparisons and deciding on an economical way forward, currently there is just bad-mouthing of RPA in big companies’ shared service centers and the very slow building of these “modern” apps, who are awaited by the businesses like the new messiahs, and which will be nothing more than the new disasters that nobody will use. No wonder success rate of IT projects is 20%.
Let me tell you a story, because this type of fight is not happening only within IT or business process world. I go to the beach club this summer. As I stopped drinking almost 2 years ago, I drink solely non-alcoholic drinks. So, I went to the bartender and I ordered a non-alcoholic cocktail. The bartender, who was obviously having a great time, but he seemed also very drunk, almost did not want to serve me, and tried to humiliate me in front of other clients because I was ordering a non-alcoholic cocktail. He challenged me and my “manhood”, and I had almost to beg him to prepare me a non-alcoholic cocktail, after I waited for approximately half an hour in the queue. This is exactly what is going on with the traditional software developers and intelligent automation within technology world right now. Bullying for obvious political decisions. Some alcohol drinkers and traditional software developers feel that their world is shaking and they are not so cool anymore. They feel their world of privilege is under attack by some people they disrespect (we are not so technical as they are, my oh my), and who, by the way, chose better lifestyles with nicer outcomes and more fun. Because trust me, intelligent automation has a lot of fun right now, inventing the processes and the companies of the future.
The managers who are obsessed with working with people will try to destroy RPA, and this is happening, while the managers who care about productivity will embrace RPA and will want more and more. The managers who are young and ambitious, and who care about productivity, will want productivity and will obviously understand that people and technology combination are the key to their success and will not want RPA and intelligent automation gone in 5 or 10 years, but will choose the best tool for the best situation.
Why isn’t there a higher authority giving a mandate to automate to everyone? That would be governance, good governance, based on statistics and the well-being of everyone. And in business well-being usually means productivity. And excuse me, but RPA beats “modern” apps at productivity, for the simple reason because they have exactly the same benefits, but much lower cost.
In a state of anarchy, without proper governance, everyone does what they can to keep doing what they’ve been doing all their lives because they feel comfortable doing. But what do we do in the state of anarchy with people like traditional IT software developers, who have a direct interest in destroying RPA and intelligent automation, who started to become more authoritarian and prejudiced because they want the whole cake for themselves? Who is the authority to discipline and police their violent attacks towards RPA and intelligent automation folks? Should we have a civil war, or someone who has enough authority and power, like a president, make peace and institute a proper governance? Win-win, or lose-lose.
These situations expose a very nasty view on the corporation. There is no law in the corporation, because decisions are made exclusively by decision-makers in their 60s who have their own agendas, and who haven’t touched a new technology in years.
As I explained before, older women prefer people, younger men prefer productivity. Generally sparking, of course. Depending on the these things, there is just the political will of the manager or group of managers, usually all same age and with same prejudices, and you, as young person who understands some things, cannot fight that. According to the official ideology, you need to wait your turn until maybe one day you will make the shots. You cannot sue these decision, like in the nation-state, and go to trial. Corporation is a big totalitarian dictatorship, at least American ones. And American corporations are everywhere.
And there’s a lot of corruption in these corporations especially because of the lack of transparency, aka governance, aka rule of law, aka clear processes that cover 100% of the cases.
The concept of good governance is the most important aspect of having a good intelligent automation product and program. In my forthcoming book, enabling the right governance is a key aspect of conquering the challenges in implementation of intelligence automation. Good governance has 7 pillars in my book:
Collaboration between IT and business is essential
People should learn to love robots
The people involved in the intelligent automation project should be great communicators
Have an intelligent automation strategy (preferably developed together with top management)
Have a working change and stakeholder management and communication approach
Use a balanced scorecard to evaluate the performance of your intelligent automation
Success of RPA should be first measured by the results of every process. Meaning how many of them were processed by employees, and how many of them were processed by the bot. Without a proper mapping of activities, without a proper mapping of roles and responsibilities, it will be hard to determine the full success of RPA. RPA was also one of the key factors telling business that something could be wrong with the process itself. So, during the development period, the whole process helped businesses identify things that were not so logical or could be done differently. And this is one of the successes of RPA as well. But without proper data from business employees, how many hours daily they spend on some activities and how many hours per day they spend after the bot implementation, I would count the success like this. But it will be a subjective opinion.
For success to be measured, we have to agree on what is the success. What is the success for IT, may not be a success for business. It depends on the parameters for measurement. So, for IT, success is zero percent of failure, which for business is totally meaningless because this is just the success rate of the process, of the bot, but it is not about what the bot delivered. And the way of mitigating investment risk is to define the success before implementing the bot, so what is really the scope, what is really the goal for business, for example.
The way to measure success is measure the same thing, with and without bot. To mitigate that is how good you select the process to be automated. If you select a bad one, that bot is not usable.
Success can be measured by reduced number of bot failures, and making sure whenever there is any kind of bot enhancement, that it’s completed timely, and everything is clear. Wherever there is an issue or constraint, it has to be resolved in a professional way, and completed as it should be. And of course, repetitive issues give harder time to business, especially if they face it a lot.
If we mean that a process has been implemented successfully, doesn’t mean that it will work well in the future, so what I think that success is that once it is implemented, it’s working on a longer term, and it’s also valued by customers, or employees, and I think, how it will be measured is not really measured in terms of money, but in terms of how people appreciate it. Because if people do appreciate it, that means that RPA gets even more attention, leading to more opportunities and so on. But if something is implemented incorrectly, that doesn’t help at all, that means that trust could go very quickly. So, it really depends, how we define success. And I think that success in our case is more like trust, and longer term, time efficiency. And risk, yes, of course, investments always holds risk, but I think we only see it once we go further. If at the beginning we see that an investment might succeed, but then we go on and see that it turns out that they are showstoppers, or at the end, it doesn’t really help at all, then this is something which has to be considered whether it’s worth, and I think this is also something which builds on trust. And it’s not really a question of money, it’s really about building trust among employees.
Measuring the success for me is based on the customers. For me, that’s the most important thing, because honestly I don’t care about the project manager, we are working for the business after all, and they are the ones that must be happy at the end of the day, not the project manager. Ways to mitigate investment risk, I think a good way is doing a really good documentation of the process and how we want to automate it. This way, if we are really going to throw RPA to the thrash, we will still be able to automate.
Success, for two different people, is different. If the goal is to increase the speed, RPA can increase that and that will be their success. You can measure it before and after the implementation. For some people, if you want to free up their people’s time to do more complicated work to help their company move on and win in the competition, then RPA can do that and do some time consuming tasks. And then how to measure that is to measure before and after and calculate how much time RPA can help. So these are two examples, I guess there are more examples. It depends on what success people would like to get. For the second question how to mitigate the risk, that depends on the goal of RPA, which can be different. I think the risk in general can be mitigated in many ways, I will give you one or two examples. One example is, if you want, if the tool is new and you don’t have any people who are familiar with the RPA tool, you can start from implementing RPA at a smaller scope or smaller area. So, that, when you complete that, you can learn how to manage RPA. And you can scale up to the bigger area later. That can help you mitigate the risk. Some other examples may be, if you want to make sure that you get the results you need, you probably need to get the expert to help you. For example, within the company, if you have a network or other teams that have implemented RPA, you should partner with them and learn from them before you get started. That can help you to jump start, and reduce some risk of failure. So those are two examples. I believe there are more types of mitigation to the risk, depending what risk you are focusing on.
I believe the success or measuring the success is an unlucky one, because if everything works perfectly, without any issues, no one is talking about this and that robotic process. Of course, the most noise is coming when something goes wrong. And everyone wants to measure the case when something does not bring the value, the initial assumption that it could bring. So, yeah, measuring the success of the automation is something more challenging, and to be completely honest I was not that much involved in calculating the FTE savings, so even though I had some initial information, in many cases it turned out that we had to readjust those calculations, and because of this, I am not quite comfortable with saying anything about the measurement of the success or measurement of the efficiency it’s bringing. Constantly changing business processes, and business not being aware of system replacements taking place can be mitigated by signaling that, ok, we as business are aware that there might be some changes in the future, then maybe if there is a bit more focus on what exactly is changing, I mean, is it the system, is it the entire process, is it just parts of the process, and any kind of, you know, a bit more explanation on what is the change or what is expected to be changing, and by when, so what is the timeline of that change. I am not sure how much I was involved in the different processes, but maybe that could have helped a lot. Track at the beginning. If there is a kind of routing in place to, you know, ask the right questions, if they see this process to be the same in one year, two years, three years, or something that will have an impact on just six months, then maybe if it’s tracked at the beginning, and constantly readdressed, then maybe it’s a bit more straightforward, even at the development phase.
What does the industry say?
iGrafx, a process excellence consultancy, writes via its CTO, that success can be measured mostly by increased process efficiency, so more output with same or less input, or same output with less input or cost. And as a way to track success, KPI dashboards seems an obvious choice.
The USBPort, a digital platform for independent journalism, writes that there are four ways to measure success of intelligent automation: speed and productivity, accuracy, compliance, and the value of freeing up humans.
UiPath, the market leader of end to end process automation, gives few clues where to look from operational perspective and business point of view. Operationally, looking at errors, utilization and success rate, but also process duration. From business perspective, ROI is king. However, it is again UiPath who comes up with note that it’s not so easy to measure RPA ROI, especially because you need to decide if you do it per process or per whole program.
Finally, Wavestone, a consultancy, in line with previous thoughts, proposes to measure the KPIs before the automation implementation to have a baseline. For example, measure performance by time. Also, measure number of errors a human does, and the FTEs savings.
Intelligent Automation Product Owner 