What do 1950’s IBM and present-day McDonald's have in common?
They're both great examples of getting the right information, to the right person, at the right time in order to drive good outcomes.
In the podcast episode below, Blue Margin CEO Brick Thompson and Business Intelligence Consultant Greg Brown discuss a critical question for companies seeking to establish a data-driven culture:
HOW do you get the right data, to the right person, at the right time (to power the best results)?
"If you can get data on time to the right people, you can actually affect the outcome of the current period and inform tactical changes and changes to your strategy that will get you much quicker and positive results than waiting for a retrospective report.” - Brick Thompson, CEO, Blue Margin
Do you have near real-time visibility into your organization’s data? Are you able to use that visibility to drive change, eliminate inefficiencies, and adapt to market changes? If not, the first step is to increase visibility in key functional areas and processes.
As a simple illustration, those looking to increase visibility around Sales might consider:
- Which factors most affect sales outcomes (e.g., lead flow, time-to-close, and close percentage)?
- Which of those metrics can be tracked and measured to monitor performance?
- Is there enough visibility in these areas to inform and improve strategy and execution?
When there is alignment on key metrics and a BI tool (like Power BI) to visualize and display them, critical information becomes available to the right people in near real time to inform decision-making from the board room, to management, to frontline employees.
This example is only the tip of the iceberg, so for further insight, listen to the episode below, read the full transcript, and explore our resource library.
For further listening/reading:
- Who Should Own BI In Your Company?
- Change Management and the Power of Data for Distributed Decision Making
- The Dashboard Effect Podcast
- How to Succeed with BI: Aligning on Metrics
Brick Thompson: 0:03
Welcome to the Dashboard Effect Podcast. I'm Brick Thompson, and today I have with me visiting again, Greg Brown, one of Blue Margin's consultants. How's it going, Greg?
Greg Brown: 0:13
It's going well, Brick, how you doing?
Brick Thompson: 0:14
I'm doing great. Thanks. So today you and I are going to be talking about what you've called the “age old bi challenge”. Why don't you start us off? What do you mean by that?
Greg Brown: 0:25
Yeah, thanks, Brick. So this is going back in time a bit. But I read something that really fascinated me. And I think what fascinated me is that the question that was posed all the way back in 1958, is just as relevant a question as it is today in 2022. And that's, "How do you deliver the right insight to the right person at the right time?"
Brick Thompson: 0:47
Yeah, I mean, that that kind of is the the central foundational question in BI that we're always trying to do as well as we can and do better over time. I mean, having having a report that doesn't meet those criteria, is not that useful, frankly, So So in the old days, oh, yeah, like 50s. Yeah. Yeah. So he wrote a white paper in 1958. And the paper was
Greg Brown: 1:06
Right. And if it's going to the wrong people, or it's going to them too late, before they can actually take action and change an outcome, then it's not really that useful of information for them. So it's always a relevant question. And what I read, it concerned, a former IBM researcher, his name was Hans Luhn. And he was one of the first are among the first to work out some of the basic techniques that are now commonplace in information science. Things like full text processing, hash codes, keyword and context indexing. Things like that. entitled A Business Intelligence System. And what that paper was about was an automated process he had devised to have computers scan an article, and then automatically write an abstract of that article to save people time, to not have to read the article, but to get the synopsis of it.
Brick Thompson: 2:03
That's so cool. I mean, we hear about text processing now, AI stuff and machine learning systems that do things like that. It's amazing to think someone came up with a system like that in the 50s.
Greg Brown: 2:13
It is yeah. And the white paper is great. I'm sure people can find it online. But you would take an article, in his white paper, he had an article, for example, on the messengers of the nervous system, this is from 1957. And it was scanned into an old data processing machine. And the machine would analyze which words came up most in the article, and then would analyze and rank which sentences contain those words that were most common in the article. And then those top sentences would be selected for a short, like two to three sentence summary of the article.
Brick Thompson: 2:45
Okay, that's cool. So this guy, Hans Luhn, I'm assuming then he did stuff specifically around BI.
Greg Brown: 2:53
Well, he did, and really, the system was its own thing as far as taking an article, making an abstract. So of course, saving people time, giving them the instant insight they may need. But then he had also went on to describe a system in which people could look up and access that information in the future, so that they could access that insight when they actually had a business problem to solve or something they were researching, for example.
Brick Thompson: 3:16
Okay, so how do you translate that, sort of into the modern world of getting the information you need when you need it?
Greg Brown: 3:24
Yeah, it's an interesting example. And analogy, or corollary to our modern times, because he was considering a system that would automatically ingest information, would summarize, would basically give people the big picture, high level detail of what that was all about, and then to make it consumable and useful for the end user. And part of that, of course, was presenting it to the right person at the right time. And this is very comparable to what we consider in modern day and age with BI, in terms of how do you take large data sets, different sources of information.... How do you condense that to make it consumable and useful for the end user? And again, how do you deliver that to them at the right time?
Brick Thompson: 4:10
Yeah, exactly. So what what you describe there, as you were describing it, I'm just thinking about automated ETL systems that are extracting data, say, from ERP systems, or whatever transactional systems you might have, and then automatically populating, say, data cubes so that users can then go and look at reports based on those cubes, whenever they want to. Seeing that seeing that information, very timely, when they want to, presumably at the right time, hopefully not too late, but when they needed it.
Greg Brown: 4:43
Well, that's right. And when you think about taking that data architecture, and then pushing it up into a reporting layer, when you look at a dashboard or an executive level summary, it's not going to be all the detail of the underlying data sources. It's going to be that high level, abstract, so to speak, like we think of an article abstract is saving me time, I understand what's going on, maybe I don't need to dive into that full article, take the time to consume all of that. And so just as his system was doing, modern BI think tries to condense and summarize insights, but also allow the user to drill into more detail, just like someone could read one of his abstracts and say, you know, I need to read that whole article because that's really pertinent to what I'm doing. Today, it's, well, I can see that insight on a dashboard. And then if I need to drill into it, if I need to look into detail for analysis to take action, then I can do that. But if I don't need to do that, I have the summary, it's quick, and then I can move on to the next thing.
Brick Thompson: 5:38
Yeah, I love it. Do you know whether Hans conceived of this around business data? Or was he more theoretical, and so thinking about his language processing?
Greg Brown: 5:49
I think he was more theoretical and thinking about the language processing, and also just thinking about what computer systems in that day and age could really do in an automated way. Of course, the technology has evolved, you know, beyond even what we could describe, but he was considering how do you take larger sets of information, have machines process it so that a human can get a summary, can get an overview of it.
Brick Thompson: 6:16
Yeah, I love it. And I think of myself, you know, I'm quite a bit older than you, so I entered the business world, late 80s, early 90s. And, you know, we had systems for being able to get at data. I remember, in the early 90s, being exposed to SQL, and it being kind of sold to me as like, "This is the business person's tool for being able to get at data and do their own reporting. It's really easy." It wasn't as easy as they said. It became easier, but I think the average business person doesn't use that. But we didn't have the types of tools we have now. And even twenty years ago, we didn't. I mean, they were available, especially at larger companies, very expensive installations. But the type of stuff we have now that makes this within reach of any business is quite amazing. And it's kind of cool thinking about the fact that this IBM researcher back in the 50s was already thinking about, alright, how do we solve that problem?
Greg Brown: 7:18
Well, yeah, and he was thinking about archiving data, keeping useful summaries and abstracts available, so that you could access those. But you didn't have to spend a lot of time digging through all the data, so to speak, all the words and the articles, to get what you needed. There was a system that you could reference and query to get those insights that you needed. So yeah, that's why I found it fascinating. And again, the connection there is that I think he really put it out in his white paper that it's about getting that information to the right person at the right time. And that's still the challenge we face just with very different technology. And of course, as you note, the good news is that the technology has evolved to the point that it's not super expensive. It's not something that's too complex, for your average, you know, middle market company to access and to use to help their employees and to help their users have better insights.
Brick Thompson: 8:15
Yeah, exactly. You know, data is so key to informing strategy, obviously. And it can inform strategy after the fact for future periods. But often, if you can get data in a timely manner, to the right people on time to the right people, you can actually affect the outcome of the current period and inform little tactical changes and changes to your strategy that will get you much quicker results and positive results than if you have to wait for a retrospective report. And I think that's kind of what you're referring to here, what you're talking about, right person, right time, right data.
Greg Brown: 8:56
It is because the right time is always, I mean, it's either in real time or as close as possible. Because if you're really looking at monitoring your strategy, monitoring your growth plan, or the initiative that you're pushing, you don't want to monitor that every two weeks when you actually get a snapshot into the data that's showing performance. You want to see that as in real time as you can, as possible for the business. And so you have to think about how do we leverage those modern tools that we have nowadays to make sure that data is flowing through the business to allow users and to allow stakeholders to have that real time view, not a retrospective view. That, yes, you can make adjustments, but what opportunities and what time have you lost? And what's the value of that, before you get to make that adjustment or that change.
Brick Thompson: 9:43
Right. And I think any of us in business have experienced that, where you get to the end of a period, you get a report and realize, "Oh, if I had known about this three weeks ago, I could have affected the outcome."
Greg Brown: 9:55
Sure, absolutely. Or, "I thought this initiative was going to bring this sort of improvement or these results. Now I see that we have to make that change. I wish I would have seen that three weeks ago. We could have, you know, executed so much better. Or we could have made that adjustment. And we're going to make it now. But we've lost that time we've lost that value because of that."
Brick Thompson: 10:15
Yeah, exactly. I think you had an example, around, I think it was McDonald's, around delivering insights at the right time.
Greg Brown: 10:22
Yeah, and this is kind of a funny example. But it really struck me. And this is from McDonald's. And this is from a process they have in place. And again, this is about the way that they've set up data to flow through their organization and to flow quickly enough to get to the right person at the right time to make a change and to lead to some sort of business impact for them. So what they have is a system that can tell from like mobile orders, DoorDash order, that an order is coming in from a hospital. And when they have that data, they send that data, they send that indication, to the store to say, this order is coming in from a hospital. And what that cues the workers at the store to do is to include a nice note into the order that saying, hey, it seems like this is coming from a hospital, we hope you're doing well, and this order is on us. It's free. And they package that note and then deliver the order. And what results from that, we would presume, is customer loyalty and satisfaction, and brand loyalty. And it's a small example, but what's interesting to consider about the example is that they've engineered a system so that the data that they collect and use is moving fast enough through the organization to get to the right person at the right time. If that person were to get that information, five minutes after the order is delivered, there's no opportunity to create that brand value, or that excuse me, that brand loyalty anymore. They've lost that opportunity. But by having a system in place where that insight is delivered to the right person at the right time, they can take action, which you know, enhances the value of their brand and the loyalty of their customers. And so it's a very small example. I, you know, it's one of those things you read that kind of puts a smile on your face, because you're like, "Wow, a huge company incorporation is really just thinking about what a person is going through, you know, at the hospital, maybe with your loved one, and you're obviously not having a great day." But it's also a very smart business thing to do. And it requires the movement of data at such a speed to get that insight to the right person. So that's why I thought that was such a cool example.
Brick Thompson: 12:21
I love that. Yeah, that's, I think of real-time often as sort of next day, but that's a perfect example of no, it's in the minute.
Greg Brown: 12:29
Yeah, that's seconds, maybe.
Brick Thompson: 12:30
Yeah. Yeah, that's fantastic. So Greg, let's, let's bring this home a bit. If you're a company, and you're just embarking on a BI initiative, how would you recommend that they go about thinking about how to get the right data to the right people at the right time?
Greg Brown: 12:51
Yeah, I think it's a matter of considering what visibility you have. And I think one useful way to look at that is to look at the different value creation or value delivery processes at a company, and to think of them like equations with different variables that impact the end result. And you can say, "Okay, well, we have visibility into this variable, but not this one." And when you can kind of chart it out that way, and say, "We don't have visibility here, or we don't have enough to execute and improve our strategy." That's a good idea of where you could start in terms of building that process to get the right insight to the right person.
Brick Thompson: 13:26
Okay, so you're referring to functional areas of the equation. So think about what key functional areas do we have that we're not getting information we need to be able to pull levers when we need to. Is that right?
Greg Brown: 13:39
Yeah, exactly. And you can try to break those down as much as it makes sense. You don't have to go overboard with that, but to consider the different components there, like variables in an equation, and where you don't have visibility. That can be a really good way to decide, where do we start with this?
Brick Thompson: 13:54
Okay, so as an example, maybe your functional area, sort of the main equation, is sales. But as you're looking at it, you're realizing sales is not trending the way you'd like it to. So you start looking at those variables within the equations like sales leads, and close percentage, and time to close, those types of things that are going to allow you as a manager to hopefully pull some levers to start to improve those things.
Greg Brown: 14:23
Yeah, that's exactly right.
Brick Thompson: 14:25
All right. Well, why don't we wrap up. It's been an interesting discussion. I love how you started us in the 50s with old Hans there. It's amazing that he was thinking about this almost exactly the same way we think about it now. It is sort of the age old question in BI. And it's one I know as a consultant that you're thinking about a lot. It's so tempting to sort of dive in and say, "Okay, what do you want," but instead of doing that, really thinking about okay, what are the goals here? What's the right data? What's the right time? Who are the right people?
Greg Brown: 14:59
That's right, Brick. And the good news is that unlike back in 1958, we have a lot better technology and a lot more capabilities to help businesses do that, to deliver the right insight to the right person at the right time.
Brick Thompson: 15:10
Yeah, that's right. All right, Greg. It's been a great discussion. I look forward to having you back again soon.
Greg Brown: 15:16
Thanks, Brick. Appreciate it.