What strategies do you use in your work as a statistician to ensure successful outcomes?
How do you set expectations with stakeholders when tackling a project?
Are there any common mistakes statisticians make when setting expectations in their job that you have learned to avoid?
How setting realistic expectations can positively impact your work as a statistician?
As I embark on this journey of discussing expectations, it reminds me of an incident that happened to me recently. During one of my business trips, the coffee I ordered at an airport was quite enjoyable; however upon opening the cup – much to my dismay – only half was filled with liquid gold! The size and shape established a certain expectation for what should be inside which created feelings of dissatisfaction due to unfulfilled hopes. It goes without saying that if they would have had appropriately sized cups, then there wouldn’t have been any misalignments in regard to meeting expectations. Fascinating how something small can make or break experiences!
As statistical professionals, we frequently encounter situations where our technical expertise is required to communicate complex data insights to a non-statistical audience. Creative data visualizations, tailored to the audience, are an effective tool to help us overcome this communication gap. By presenting information in a clear and understandable manner, we can over-deliver expectations and make a real impact on our work projects.
In this episode, I give more tips on how to set expectations and the following points:
- What are the key elements to building trust with coworkers and supervisors as a statistician
- What role does setting expectations and abiding by specific timelines play in the success of a project as a statistician?
- Some experiences where data visualizations are demonstrated to an advisory board and evoked a “wow” effect for their surprising quality as a statistician
- Why is data visualization especially effective in communicating complex ideas in a manner that non-statisticians can understand as a statistician?
- How can investing in data visuals help one stand out and over-deliver expectations in certain cases, potentially leading to promotions as a statistician?
Setting Expectations – An Art and A Science
[00:00:00] Alexander: So in today’s episode, we are talking about expectations, setting expectations, fulfilling expectations, and as we dive into this, I want to tell you a short story is that happened to me recently. One of my business trips, I was buying a coffee at an airport and I got this really nice muck and then I looked into it and it was only a half full the muck was basically setting for me the expectation of how much coffee I will get, and then I looked into it and was only half full. Although the coffee was nice, it really led to, yeah, me thinking, I’m dissatisfied, I’m frustrated. If the muck would have been just half the size, the expectation would have been different and I would have been satisfied. Interesting, isn’t it? Same amount of coffee, different marks, different expectations, different experience.
I once was at a training at my, one of my former companies, and because I was working in the affiliate, it was a lot about marketing and communication and these kind of things, and the trainer mentioned that if you want to create a wow effect, then a wow effect was a customer with a client, with a coworker. With an employee, you always need to deliver above the expectations. Otherwise, you’re just satisfied or you’re even dissatisfied. Think about yourself. When did you have such an experience? Maybe with the recent salary increase. Yeah, maybe you expected more, and then when you got your X percent salary increase, you said I’m not really happy about it, because it’s less than what I expected.
We have this all the time. Yeah, I once had a person in my team who was on a, yeah, very difficult career trajectory, and so were already discussions about him leaving the company and then he got this really challenging project. I thought he was the right person for it and he overdelivered on it. He really went into it. He solved all these problems and enjoyed it. For him, this was really great because he now had something that he was enjoying and he over-delivered and over-delivered on the expectations on him, and so he got a couple of years later, a promotion on it. Actually, three years later, after we had kind of discussions about him leaving the company,
I also recently had a coach and during the first session with the coach, she did some guided meditation with me. I’ve never done that before and I was skeptical about these kind of things because it sounded a little bit esoteric. But it was really good. It went far beyond my expectations, and I had says, wow, effect sets a really good coach.
Managing expectations is really difficult, especially if you set big goals. If you set these audacious goals, these moonshots. I was once working at a company and in an organization whereas the new leader, the new product leader, global product leader was it, okay, we are here shooting for the stars. She set really high goals, but she also made it clear that these are really high, and she communicated in such a way that she said, okay, we really wanna reach these. But they are really difficult to reach. That I think is a really interesting thing. Yeah, because it communicates that you want to reach far and it’s very much the opposite of what is very often done. Yeah. Goals are very often sand backed. They are set in such a way that you can really achieve them, that it’s easy to achieve them. I can think of a scene in one of the Star Trek movies where there’s this old generation, it’s a huge generation, and of course like always there’s some kind of emergency and the engineer get asked by the captain, when will you be able to fix it?
And the new engineer says, we only need two hours. Okay. And then the old engineer looks at the new one and said, you just told the captains that it will take two hours. It takes two hours. Yes. I know it takes two hours. Then you only have these two hours. I usually said I need three hours or four hours. And when I was then was delivering in two hours, I was a hero. So this kind of old engineer, always sand backed stuff, always wanted to make sure that he would over delivered. And he, of course, knew that the captain would always push back. So that was a typical game. And that’s a little bit the problem on always sandbagging expectations. Now, because if you are always sent back or sent back too much, Then you get into this problem that you always get pushed back on things.
Another problem with, especially setting timelines and things like this, we are notoriously bad for in terms of planning. There’s some kind of bias that we always think everything will go smoothly and we are overly optimistic, usually on when we will be able to deliver. And then in reality, of course not everything goes to plan. Of course not. So in these kind of scenarios, sometimes it’s good to give some kind of range. Yeah. And to say, okay, if everything goes more or less to plan, we’ll hit here, but realistically will deliver at that date.
Now it’s really also important to think about what happens thereafter. I once had this discussion within a company where the regulatory people sent backed quite a lot, and then were super happy. We have received approval two months earlier and the whole organization was not ready for it. Everybody had thought, okay, this is an approval date. Okay, we’ll have, recruitment lined up by then. We’ll have training lined up by then. We’ll have promotional material lined up by then. We’ll have our reimbursements lined up by then and nothing was ready. Yeah, great. You had a approved drug. But you didn’t have anything to launch it, so there was no really additional benefit with the earlier approval because it couldn’t materialize. We couldn’t, give any drugs to patients. So the regulatory people were partying and saying, oh, we were faster than expected, much faster. We over-delivered and so on. But for the company as a whole, it didn’t make a difference. It actually led to a lot of frustrations because now of course, the people in the marketing and the PRA departments were asked, why are you not ready to launch? And they of course said we didn’t expect it to be ready now. It’s a lot about trust within companies. That you can say really okay. These are different discussions to have. These are different negotiations to have. So whenever you talk about timelines, talk about the deliverables. It also depends very much on trust. And trust is one of the key things that we speak in our leadership program about. Trust has on one hand something to do with your competence, of course, but it also has something to do with your character. It also has something to do with does the person believe you care for them. And so these are the three things that we actually dive very deeply into in our leadership program.
In our leadership program, which is really about influencing. It’s really about not, supervising people, but about these crush functional relationships, these typical things when you as a statistician worker, and a cross-functional team, and you need to convince. Your clinical operations people, you need to convince your regulatory people. You need to convince your physicians, the pharmacologist or whoever you work with, or you need to convince upper management. Yeah, so this is what the leadership program is really about. If you wanna learn more about this, just check The Effective Statistician homepage. And we mostly actually offer this nowadays to companies and companies purchase them in bigger chunks for their teams. If you’re interested in that, just reach out to me and we can have a discussion.
In terms of character, competence, care, it’s really important to keep people updated. Don’t have any negative surprises. Maybe you have heard that already from your supervisor. Supervisors can deal with more or less everything, but not with negative surprises. Yeah, if you mess up, tell someone about it. Tell your supervisor about it. Tell the stakeholders about it. If you will, need a couple of days more before you can deliver. Don’t tell the people at the time of the delivery, tell it earlier. This is really important.
Now also trust character comes from taking responsibilities. If something goes wrong with your project, take responsibility for that. Don’t blame others. Don’t blame the CRO, the programmers, the data managements, whatsoever. Yeah. Make sure that you have a good relationship with those people that work before you, that need to deliver to you so that you can deliver. And build trust with them. And make sure they also understand that you need to work on this. You need to then deliver once they have delivered and that you don’t want any negative surprises either. If they say we’ll deliver on Monday, then you expect things to be delivered on Monday and not on Tuesday. And if they can’t deliver on Monday, then at least you want to know on the Thursday before, not on the Monday morning. These kind of things build trust, and of course, if you always deliver, if you always maybe a little bit over-deliver, that builds a lot of trust because then people see. You say what you do and you do what you say, and this kind of alignment is really important. It’s especially important if you have people leadership responsibilities.
So if someone reports to you, what you do speaks much louder than what you say. Because what you do is fulfilling on the expectations of what you have set. If you say something, but then you do something else, you don’t deliver on the expectations yet, you have set yourself. Maybe you have seen that in politics, or maybe you have seen that in upper management. Yeah, where they have on a Sunday speech says this, and next Monday they do something else. Say, you know the company values about X, Y, that and then you see some decisions and you think like, how do they go together with these values? This is where, expectations are not met, and that hurts trust quite a lot.
Now, timelines is of course, one thing in terms of expectations. Quality is a completely different thing. One area where you can really wow people, where you can really over-deliver in terms of expectations are data visualizations. I worked once in an area where, everybody was just doing line charts, bar graphs, because mostly we looked into digitalized data over time. So there was some kind of continuous variable and that got pre skewed distribution after treatment, and therefore people would look into how many patients have at least 75, 90 or maybe even complete response. And that of course, led to lots of bar charts or line graphs of how many responders do you have over time. And everybody used the same. Yeah. If you go went to a conference, all the presentations more or less looked the same.
Of course, it branding was different, but also designs were very similar. So you had a dejavu every time you see it, saw a new presentation and I was thinking we should do some things that stands out. We should generate a data visualization that shows more senses, digitalized endpoint. Because of course it’s much more happening. Yeah. If you think about how patients develop over time, do they really stay where they are or do they, go up and down and just because on average the response rate is the same, you know that not, it doesn’t mean that the patients really always have a response. Yeah, because flip around unless you have 100%, 100% responders, and of course nobody can flip around anymore. But usually you don’t have that. And so we looked into an animated scatter plot and this animated scatter plot showed much more the detail of the data showed it in a completely new way.
When I first showed that at an advisory board, people were amazed they didn’t expect something like this, and this was really a wow effect for them. Data visualizations are area where expectations are very low. And where also non statisticians can understand how something is more valuable because you can directly see it, every non statistician will understand. This is a great data visualization, and this is a poor one. They see it, but they don’t necessarily see that if you have different multiplicity adjustments or if you have different designs. Yeah, of course. You then see the operating characteristics of a design are different and better and yeah, you have some savings in terms of money and things like that, but data visualization is really compiling. These kind of data visualizations actually got me promoted a couple of times, but that’s another story.
It just shows you that in this area, if you invest there, you can easily over-deliver, you can deliver more sense that people expect and set creates a wow effect that really helps you to standout. So if you wanna learn more about data visualization, I’m offering different data visualization workshops. So mostly companies actually purchase them and these 15 people online or 20, 25 people in person that attend this free for our interactive data visualization workshops.
If you’re interested in something like this, reach out to me. Almost, there’s also the opportunity if you are located in Europe, you can come to ones that we are organizing in Berlin on the 21st of June. So head over to the show notes of this episode and you can learn more about these interactive data visualization workshops. So to sum it up. Always make sure that you deliver on your expectations and check with your stakeholders that you have the same expectations in terms of quality, timelines, budget, if that plays a role, these kind of different things. And if there’s anything changing, keep people updated. No negative surprises. So with that, have a nice time and talk to you again.
Never miss an episode of The Effective Statistician
Join hundreds of your peers and subscribe to get our latest updates by email!