Just another Temple Fox MIS site

Weekly Question: Week 11

Leave your response as a comment on this post by the beginning of class next week as well as your comments on your peer’s responsesRemember, you need to average four posts a week for a B. For these weekly questions, I’m mainly interested in your experiences and opinions, not so much particular “facts” from the class!

Answer one of the questions below:

  1. Give an example (at home or at work) where you’ve needed to determine the “critical path” to get something done, even if you didn’t use a PERT chart to do it!
  2. When you’re asked to give estimates of how to get something done at work, how do you go about doing it? What methods could you employ to “narrow your confidence interval” and provide a more exact estimate?

97 Responses to Weekly Question: Week 11

  • Give an example (at home or at work) where you’ve needed to determine the “critical path” to get something done, even if you didn’t use a PERT chart to do it!

    Give an example (at home or at work) where you’ve needed to determine the “critical path” to get something done, even if you didn’t use a PERT chart to do it!

    One example that comes to my mind is a project that I worked on during my internship last summer. I was helping a biotechnology company with its digital strategy. As the company had acquired many smaller tech firms in the bay area, there were processes that needed to be streamlined. One instance would be of a product launch. Because the products that the company developed were complex, and needed some kind of instructions, the product team would hold live seminars where they would not only launch the product, but also talk in detail about its specifications and use. The smaller tech firms that were acquired would also have their own product launches and online seminars. Because the smaller firms were planned to be brought under the company’s umbrella brand, their strategies had to be streamlined. Now, my role was to create a year-long digital strategy for the umbrella brand. While working on the project plan, I obviously had to keep in mind the different smaller companies that were acquired and come up with a plan that would encompass all the launches and promotional activities. In my plan, I included all activities that would be conducted, different teams that were or would be in charge of those activities, the amount of time they would take for the activities, the overlap, etc. Because there was some kind of projection involved, I had to keep in mind dependencies, and the “critical path” in every case. Sometimes it would be the marketing team taking longer to create a plan, and others it would be the product team giving in-depth knowledge to the marketing team about the product. Although I did not literally use the PERT chart, I extensively used the Project Plan format to ensure where the overlapping was, and where the slack time would be available.

    • Sid, this sounds like a huge undertaking, especially when you were new to the company and their processes! I’m guessing there were a lot of moving parts and one seemingly minor change on one component had a lot of affects down the line and on other aspects. An example like this really shows how important slack time is. This example shows how slack time is not only important when a process is not running according to plan, but when the plan was not estimated correctly.

      • Yes Lauren. It was challenging also because these other teams were not on the same site as I was. What really worked in my favor was that at that point in time, there was not much being done keeping everybody in sync. So in a way, I got to start afresh, and set processes which would help the teams work in a more consistent manner.

    • My goodness, Siddesh, sounds like a very involved project. Too much happening at once. This would definitely have to be planned out on a chart. The time frame could be infinite if there was no format or critical path. What was the time frame of this project? I cannot imagine the amount of discrepancies and misunderstandings on this project!

  • In my current project for Meaningful Use stage 3-CCDA I was asked to estimate rework that needed to be done due to a last minute change in one of the requirements. The estimate was broken down into several pieces. First the coding had to be broken down into 2 parts so I had to find available coding resources with the right skill set and who would provide the highest quality. After the coding resources were selected we estimated the coding based on the maximum amount of time it would take to be finished and unit tested. The minimum was 3 days and the max was 1.5 wks. Then I had to determine the amount of time to rebuild the content using the new code work. Since I was doing this work I knew the exact amount of time it would take with was 3 days max. The integrated testing workflow was still the same so nothing had to be added for that. After the estimates were complete, the team looked at the deliverable timeline to determine if we could make the release or management needed to request more time from the program. In the end we estimated 2 wks to get the task done and ready for integrated testing but overall it took 1.5 wks and we made the deadline with a cushion. In the end the estimating was done with the critical path method using the maximum amount of days to get to each point in finishing the task.

    • Hello Stanley, I am not sure how sourcing contract works for code projects. Are there different scales depending for the level of software engineer that is going to complete the job? Like one price for entry 1-3 yr experience, Masters level 1-5 yr and Senior software engineer and this forms the basis of figuring out the cost and duration of the project?

    • Hi Stanely,

      This is a good example of estimating a timeline by understanding your resources at hand. The breakdown looks like a lot of over time and mildly stressed workers, but only for a minimal period of time, but the planning seemed right on point.

  • When you’re asked to give estimates of how to get something done at work, how do you go about doing it? What methods could you employ to “narrow your confidence interval” and provide a more exact estimate?

    During my internship last summer, one of my role was to create schedules for projects. I would be given a date by which the project had to be released. With that date in mind I try to calculate backwards how much time could be allocated to each teams, and how much time could be a buffer. With these rough estimates in mind I would then approach the engineering, quality assessment (QA) and user experience team leads for an estimate. Sometimes, the engineering team may request a further breakdown of tasks to help them estimate their work. In such cases, I would schedule a meeting with the developers, team lead and the project manager to breakdown tasks and decide on an estimate. I also have to take into account availability of resources such as shared resources and vacation time. In case of shared resources, I need to understand how many hours per week they can spend on the project. The QA team is usually very busy and the testers are usually working on multiple projects. The sooner I block those resources, better chances they are available. Based on the time take by the same teams to complete previous projects, I have an idea of how long they should be taking.

    • Estimating is something which I have always found very challenging, Swetha. It is interesting that you had to break down the project tasks to very micro level. I believe it was also a cause of resources working on multiple projects at the same time. It must have been tough because sometimes it gets difficult for a person to change focus from one project to another. Also, if something is on fire on a different project, that resource might give your project little priority during that specific period. Did you face any challenging situations? In my experience, I have seen that developers usually give a larger range of estimate timeline than they would usually take. Did you face a scenario like that?

      • Hey Sid, yes it may be difficult for resources to focus on different projects. Sometimes its the same task to be executed in different projects, so its not too bad. Priority issues take place very often. On such occasions the project managers need to step in and agree to share resources a certain way. It is very important for PM’s to maintain a good relation with their peers and those above them. Higher ups tend to be more accommodating to change in release dates if the PM’s maintain a good rapport with them.
        Developers almost always give a larger estimate timeline, it is the PM’s job to stay alert to it. As a PM you are knowledgeable about how long such a task usually takes, and the level of people required to accomplish that task. It is definitely a tough battle to fight. On the other hand, developers cannot take too long, because then they will get a reputation of being slow and not abiding by the timelines.

        • Swetha, I agree that the PM’s job is to ensure capability but I know first hand, as you likely do as well, that not all team members are equally qualified or adept. In the perfect world, those not meeting or exceeding standards are released but we know that is not true universally. When I derive estimates ( as noted in my post), I stick with averages as a way to narrow my confidence level. I have to assume that the manager with oversight responsibility sees to it that the work is done. In some cases when it is evidently clear and more narrow estimate is required, I know the “go to” folks and estimate based on their known productivity level.

          • Maria-
            I agree, not all members of a team take the same amount of time to complete a project or a task making estimates difficult. Personally, there are some tasks that I speed through where some people would take days I can complete in hours. There are other tasks, however, that require more time because they are more detailed task and issues typically come up that have to be addressed. The issue is that some managers assume all tasks require the same about of time because they only think about the person’s individual capabilities completing it, not the task or potential problems themselves that may go beyond the scope of the estimates.

    • What you were asked to do is difficult. On one hand you’re given a release date and then told to do an initial rough estimates before pulling in the teams. Until recently the company I work for also did things in a similar way. We are supposed to be an Agile company but that is not Agile methodology. We’ve now changed direction and only “propose” a release date and no estimating is done without SMEs. If they are not available then estimates are only done if the project is similar to a past projects (OPA reports) and the managers know what the development time was on them.

    • Swetha – I really like your approach of “backtracking” in order to estimate the time taken. I believe that breaking down the tasks into micro level helps in understanding the dependencies of the tasks and also identify resources needed. As Rikki had mentioned, amount of time taken varies per individual depending on their skills and comfort level in working with the task. This makes estimating even more difficult. Having a buffer time is a good idea so that the actual time taken is not very high from what is estimated. I think experience plays a crucial role for this activity because one can learn to estimate accurately based on what worked and what did not work from past.

  • Sounds like your project was across companies and you controlled the WBS from the “control account” level with each account being a company then a more granular breakdown was done based on the activities each company had to do. Being a PMP I look at the CPM and PERT as two different things and used at different times. CPMs are generally used at the activity task level with a WBS to develop a schedule whereas PERT is used to estimate costs using 3 cost levels than a standard deviation should be calculated. It would be interesting to see how many dependencies there were between accounts, and how the overall PERT was determined based on the CPM.

  • In the past I have been tasked with doing quarterly data quality reviews of providers in our Registry. This can be a very time consuming process. When asked about how long it would take me to do this, I would often find myself mentally breaking down the process into different steps. Looking back I can see that what I was doing was trying to make a mental WBS. If a question like that is put to me again I think I’ll probably use a WBS to get a better sense of what needs to be done, and how long it will take.

    • I agree with you. When I first started studying for the PMP and learned about the WBS I realized I did this process even as a clinician at the patient bedside. I listed out the tasks that I had to do with the patient and prioritized them based on highest importance for patient survival.

    • John, I also worked on quality improvement looking at data, specifically for the purpose of reporting outcomes to funding sources. This indeed takes a great deal of time and you’re often on a tight deadline. At my old job, I was looking at a report and the data and the performance indicators did not make sense. I spent a great deal of time trying to break it down to understand where the problem was so I could create an accurate report, but the deadline was more important to my supervisor than the quality of the work. I agree that planning is necessary and often projects take more time than anticipated if issues arise that could not have been predicted, but sometimes at the end of the day, delivering on time whether the scope of the project is compromised, is what takes priority.

    • I do the same thing in terms of a “mental WBS”. I try to get down to the smallest unit of work I can and then multiply that out for the whole task. While it isn’t always that cut and dry, it can at least give a general ballpark estimate to start out with.

    • John, I think that to some degree we all do a “mental WBS”, but don’t actually go through the process of physically writing out all of the steps in the process. This is where people often come out with unrealistic estimates for the time it takes to do a work process or even daily activity. The WBS is something that I would like to start using on a more formal basis with both work and my personal schedule to become more efficient in the use of my time.

      • I agree, and I think the devil is in the details. When I give an estimate based off of a mental WBS it seems reasonable, but inevitably I run into issues I had considered, because I had not actually mapped it out.

  • I’ve done some estimates if certain resources are on the critical path for the sprint based on # stories they have and time estimate, they may depend on somebody else stories or may be others be dependent on them. There is a way you can determine what time take individual tasks based on prior similar task complete, but its tough to determine if the resource will be able to complete all his tasks because there are things which are coming from the outside like production support and bugs from previous release sprints. You need the input from the resource as well because he may see things you are not seeing to do better planing of the critical path.

    • I agree with you that a “cushion” is always needed when doing estimates. Also, in project management there is something called Enterprise Environment Factors (EEFs). These are things that may impact a project from an “external” standpoint (i.e. government policy) and needs to be factored into the estimates. Sometimes they can even drive the estimates.

    • Hi Bisser, you bring up a good point with saying that stories or projects can be dependent on each other to be completed. This can make estimating time more difficult especially if there is a large team each working on their own stories. Because of this complexity, sometimes it can help to go with an iterative sprint approach where if something is not ready for the upcoming sprint it simply goes in on the next sprint. This may not work with high urgency projects, but may help with normal fixes or improvements that are going in.

    • Hello Bisser, nice point in taking into account external factors which could potentially influence the finish date for an estimate. I have found out from my work experience that these type of issue normally throw a project off course no matter how much planning you do. It takes team work to come up with a solid estimate on a completion date for a project.

  • The accuracy of the estimates is very dependent on external conditions as well. To be the most accurate your estimate you have to assume the you won’t have any interruption from external factors. So the assumption is you have zero interruption than you need a safety margin 20 more time at least than what you expect, because things may happen miss this or that.

  • The accuracy of the estimates is very dependent on external conditions as well. To be the most accurate your estimate you have to assume the you won’t have any interruption from external factors. So the assumption is you have zero interruption than you need a safety margin 20% more time at least than what you expect, because things may happen miss this or that.

    • Bisser-This is something I need to improve upon. I’m learning that I need that safety margin but am not sure how to know how much of the safety margin is enough to allow for any outside factors that would extend the time I’ve committed to completing the project. It is also not practical for me to set aside 6 hours a day in case I need to get something done I was not able to finish in my previously estimated time.

      • Hi Rikki, I agree that it is a challenge to find a balance of allowing enough time to complete a project while being realistic in terms of how, much time you actually have to set aside for it. For me it usually doesn’t go as I planned it (sometimes due to procrastination), but I somehow figure out a way to get it done.

    • Bisser – I completely agree that it is important to plan for interruption from external factors. However, I am curious how you came up with the 20% figure. I always seem to vary in my estimation of how much extra time I will need, depending on how busy the timing is otherwise and how difficult the project is.

    • Bisser, I also agree to your point that the external factors come into play while outlining the estimates. While dealing with the application issues in my prior job, there were times when I had to bring all the relevant parties on board which included vendors and other internal teams. Often I kept a backup session/meeting in addition to the scheduled one and also calculated the additional time that may be required to resolve the issue based on the urgency and criticality of the problem at hand. However, as Lauren mentioned, even I would like to understand the thought behind 20% safety margin that you have stated.

  • In providing estimates, I usually convert the time needed to Full Time Equivalency (FTE). For example, if I need .75 hours of Analyst work for every 3 hours of Coordinator work then I need a .25 Analyst FTE given every Coordinator’s 40 hour work week (i.e 5 Coordinators requires 1.25 Analyst). Average salaries for Coordinators and Analysts can easily be applied as well. In the hospital world, “productivity” is measured closely at the bedside and that scrutiny is applied to overhead endeavors as well. I once measured what it costs for paper forms to be submitted when an electronic process flow was available. Based on an estimate that each of the four required signees took 1 minute to review and execute a form, it was determined that it cost the organization

    1 minute * 150 forms per pay period * 26 pay periods + 3900 minutes/60 = 65 hours/2080 hours = .03 FTE per signee or .125 FTE given 4 signees. These may seem relatively low but given the salary levels of the signees, the dollar amounts are unacceptable as well given the existence and proven capability of the electronic process.

    • Maria – this seems like a really good way to evaluate the capability. The numbers on their own may not seem like a big deal but once the FTE is calculated it really sheds light on the whole process and sheds like on efficiency, or the lack thereof, for a full pay period. I hope the hospital recognized the room for improvement and made some changes!

    • Hi Maria, I think that your example shows a good way to quantify productivity in terms of numbers. At my job our 40 hour work week is split up into percentages based on what we are working on. They do this since departments are charged for the amount of hours spent working on their related projects. However, in reality my actual work week never matches up to my allocated percentages since we often get pulled into various things that aren’t planned.

    • Nice job Maria. If we could take your point and apply it to the PHS scenario where an organization is dealing with large volumes of forms and a wide spectrum of employees ranging from highly paid specialty physicians to a general practitioner at the end of the day, paper work adds up a sizable amount of money.

  • When you’re asked to give estimates of how to get something done at work, how do you go about doing it? What methods could you employ to “narrow your confidence interval” and provide a more exact estimate?

    I actually am terrible at estimating the amount of time it will take to complete an assignment or project. I typically look at the assignment and the due date and establish the amount of time I think it will take. I sometimes only allow for a small window of time before I start looking at the requirements for the assignment and figure out I need extra time for team collaboration, readings or searching for resources. My issue is that I make my confidence interval too narrow. I’ve been learning to look at the scope of the assignment or project earlier to have time to plan when I can work on and factor in extra time needed as a buffer to complete. What is a barrier for me is that I typically work on something in one long sitting. It is difficult for me to break a project or assignment up into multiple sessions which explains my narrow confidence interval. If I devoted small amounts of time over a longer time span that would work out, but if I continue to have my one sitting approach, then I have to establish a time frame that allows for all the work that needs to be completed, the prep work and also any breaks I may take.

    • Rikki
      Your way is not incorrect. It is a different way of doing things. The important thing is that you realize there are ways. We all learn things from observation or experience. In your example, i feel that our surroundings and our limitations cause us, to plan differently. Projects flourish due to different views of individuals. It is a daily, ongoing learning experience for all of us. No one knows everything!

    • Rikki, I find myself taking the one sitting approach as well, but it is largely because I want to put my full attention on a project. For larger projects like a research assignment, I will break it up into sections and complete each of those in one sitting. When assignments at school stack up, I prioritize them by when they are due and how long similar assignments have taken me in the past. I also consider time of day. For example, it may take be 1.5 hours to complete Accounting homework in the morning right after coffee, but 2.5 hours after night class and dinner. Sometimes, my schedule only allows me to complete Accounting late at night so I need to know both time frames.

    • Rikki, I also at times fall into the trap of not estimating for break times, meals,travel times, and other activities that can detract from my work. I think it is most helpful to really think through the entire process when planning and be honest with yourself about how long the total amount of work including breaks etc. will actually take. Hopefully one day I will start to use that line of thinking myself.

      • Oh geez, that’s a good point. I was trying to plan out my school work for this three day weekend and I never even considered taking meals into account. I ended up doing the bulk of my work on a single day, but I probably would have had a better estimate had I considered non-work related activities factoring into my available time.

      • That’s a really good point! We often tend to forget to include the time taken for “unproductive” works which results in inaccurate estimates. Even I do the same thing while estimating the amount of time I would take to finish the assignments. Most of the time, I tend to go beyond the time limit that I have set for myself.

    • Your response was interesting to read and rather real. I tend to try to get things done in one sitting many times too and that can be tiring and not always so effective, but sometimes one does not have a choice. At times it is better to break up the work in portions so one isn’t overwhelmed and can be more productive too. It is good that you found out what you can do differently to get better results.

  • Give an example (at home or at work) where you’ve needed to determine the “critical path” to get something done, even if you didn’t use a PERT chart to do it!
    In one of my last jobs I was tasked with a number of responsibilities that needed to get done each day but there was not enough time to complete all of them. I was a social worker on an inpatient unit. I had to conduct psycho-social assessments on patients within 48 hours of their admission, I have to obtain clinical reports from all 4 physicians during their rounds and I had to contact managed care for each client requiring approval for extended stay as well as document everything I had worked on and the outcomes in the patient chart. Since health insurance providers are only available between normal business hours and I was the only person who did my job I had to be smart with my time and focus on the tasks that absolutely needed to be done. My critical path included only sitting to listen for clinical updates on the patients that were due that day and begin calling managed care right away to leave a message leaving time for them to call me back. While I waited I met with doctors and nurses individually to get clinical information I may have missed from doctor’s rounds. Then I checked the EMR to get the rest of the medical information I needed. By this time, managed care would be returning my calls and I was able to provide medical updates on the patients. If I had obtained all of my clinical info and I was waiting to hear back from managed care, I would then conduct my psycho-social assessments for the day. I focused on the tasks that had to be done that day and so I was able to get all of my work done.

    • Rikki – Since you were the only one with your role I’m guessing you needed to figure out the critical path on your own, through trial and error. It sounds like you found a rhythm that worked for not only your needs but everyone else’s as well. However, I can only imagine being new in the role and trying to get a handle on all the tasks and ending up calling managed care too late or having missing information. It sounds like this critical path would have taken some time to figure out!

    • Hey Rikki – since you were the only person who was responsible for those tasks, how did you manage your time when there was a delay in the critical path? Looks like you had a pretty short turn around time – 48 hours, before you had to get to the next set of tasks. I believe there were a lot of factors that were not under your control and that had to fall perfectly in place for you to keep up.

    • Wow Rikki that sounds tedious. Looks like there were many factors that were not under your control. How did you manage your time in that scenario? How did you estimate which part of the process will take how long? Were there times, where you could not finish a certain task, and that led you to miss the deadline for another task, due to rollover?

  • While working at one of my part-time jobs here at Temple, I was assigned the role of Project Manager where the project revolved around identifying technological solution to eradicate obesity in teens. The various tasks that were involved but were not limited to conducting interviews and surveys, engaging in design thinking with team to outline wearables, products evaluation, self-daily monitoring, and journey maps. The team I worked with comprised of high school students, technical architect, and a designer. While outlining the project plan, I took into consideration the estimates for each of these tasks keeping in mind that students had limited time apart from school. Additionally, I had to also think about the availability of the technical architect and the designer as they worked on multiple projects. So taking into account all of these variables and also the slack time around these aspects I outlined the plan for the project.

    • Hello Rishabh:
      Nice mix of stakeholders for this project. How did you go about estimating how punctual high school students were for the various tasks compared to the adults like the Architect and Designer?
      I can see it must have been challenging to figure out the right amount of slack time for various tasks. Was there an app for this or was it a cloud based application?

    • Sounds like a lot had to be done with not as much help or time. Did you face any difficulties while being the project manager? Especially with the high school students since they go to school too was it hard to manage their timings for the work to be done?

  • Hi Rishabh, looks like you had limited resources and time – that is two factors in the triple constraint. Were you given a deadline to deliver the project, and how did you balance that with the available resources?

    • Swetha, there was a initial timeline set considering all the mentioned factors and also the two critical resources – designer and technical architect were made aware that they would be required for the project the following week. So they managed their assignments accordingly. However, later the technical architect left the job and the timelines were then massaged to incorporate the leftover tasks.

  • In my previous position I utilized the idea of “critical path” when placing my equipment rental orders for events. After figuring out all the event details and what rental items I needed I typically placed orders with two different external vendors (each carried different types of items). One vendor typically got back to me with an invoice within 48 hours and the other typically had a 72 hour turnaround. Once I received the order invoice I always checked the document line by line to ensure I was getting what I needed. It was too easy for me to forget something on the list that I emailed to a vendor or for a vendor to make an error on their end. About 25% of the time there was an error so the double check was necessary. I made sure to place my orders early to ensure there was sufficient time for changes to be communicated and carried out, before receiving an updated invoice. During busy times it was common that a vendor did not have enough of an item to complete an order, which required a phone call about alternatives and added additional time to the turnaround. It was necessary to have the final invoice amount for the client’s final invoice, which was sent a week before an event. However, while I was waiting for final invoice from vendors I was able to carry about other processes related to finalizing event documents. I learned that it was most efficient to build my rental order lists while I was going through preliminary event details and then send my list to vendors before finishing the other activities. If needed, I could always add an addendum to either order. This way, I received the orders from external vendors with time to spare. All other processes were internal and had a much quicker turnaround.

    • Lauren, looks like you also had to use estimation in times when the vendor had to relook into his inventory that he had missed the first time. How did you manage to estimate the time for that? Also, the critical path was the vendor who took 72 hours for a turnaround. But how did you ensure that the other vendor delivered things on time? Also, any more delay by the vendor with 72 hours turnaround time would mean a mess. Were you able to manage that well? How challenging was that?

    • Lauren, I used a similar process when I was working on magazine publication. I would stagger writing and editing stories so that I could send the art director a few stories at a time and spend time writing and editing the others while the art director worked on the completed stories. Even when there was a backup in the art department, we had a constant stream of work and little lag time.

    • Lauren, this reminds me a lot of my work experience. When planning ahead for events, I also found that ordering items and accounting for possible issues with those orders took the longest time and had to be done up front. Other internal tasks as you mentioned took significantly less time and therefore had a good deal more slack time in which to complete them.

  • When you’re asked to give estimates of how to get something done at work, how do you go about doing it? What methods could you employ to “narrow your confidence interval” and provide a more exact estimate?

    I first try to break down the project into smaller tasks similar to a work breakdown structure, and then estimate the time for each task since it is easier to estimate smaller pieces than the whole thing at once. I also try to take into account any potential setbacks, challenges, or unexpected events that may come up and work that into the estimate. To narrow the confidence interval I can look to see if this is a project I’ve done before or have done something like it. Then see how long the previous project took to get a baseline of how long it might take. If it were a project I have no prior experience with, I would look to see how long it took others to complete something similar. An example would be when I worked on IT support tickets. After doing it for a while, I was able to come up with my average number of tickets completed per day and average time it took to complete one ticket which gave a pretty narrow high confidence interval. On the other hand, a project I’m working on now involves developing a script to track our code deployments. The script needs to be written in Jython, with which I don’t have any experience. In this case I would have a wider low confidence interval since I’m unfamiliar with it. I would estimate that it would take longer than normal because I anticipate there will be some setbacks.

    • Jonathan, one question keeps coming back to my mind. Looking at WBS, one breaks down a project into smaller tasks. In times of stringent project deadlines, does it not make sense to work in reverse- we get the deadline from the client, and then work with various stakeholders in the team to figure how much time each one of them would want to complete their respective tasks? I think in some cases, this would lead to a more efficient team, and would help complete the project on time as well. I’d also like to hear Prof. Flanagan’s views on this.

    • Jonathan, I feel the same way. I can estimate with a narrower confidence interval, if I am well versed on the subject. If I have no clue about what is happening, then I would have a very wide confidence interval. Experience gives us the ability to quantify, quality of projects.
      Education and experience are key factors in narrowing our confidence interval.

    • Jonathan, I too agree with your point of knowing about something enables to narrow the confidence interval. When I worked with consumer management company, I had to utilize Tableau tool in my day to day duties. I had limited knowledge of it initially as part of which I had a wider confidence internal in order to complete the tasks that I was assigned with. However, as I got more familiar with the tool the confidence interval narrowed down for me.

    • Although I was completing a very different set of tasks, I briefly spoke to the effects of set backs in my example as well. I agree that it is much easier to estimate times when you are completely familiar with a task. The difficult part is estimating how long a setback will take because they are at their very nature an unknown event so their is no great way to estimate an unforeseen issue.

  • When you’re asked to give estimates of how to get something done at work, how do you go about doing it? What methods could you employ to “narrow your confidence interval” and provide a more exact estimate?
    In my role as a Clinical Document Improvement Specialist; there are reports generated by revenue cycle which tell us how each specialty service for example, surgical services, is doing as far as documenting appropriate diagnoses for treatment given in the hospital. I review random charts of selected patients daily, based on a daily in-house report. The report indicates whether or not diagnoses is documented. I can estimate whether or not, the report is correct by reviewing the patient charts. My confidence interval is usually wide. I have been training doctors by rounding with them on the patient units, giving feedback, at the time of documentation and giving them monthly training sessions. I can see that there is a more narrow confidence interval when I continue to monitor the charts daily, after the training sessions and one on one discussions in person or via email.

    • Hi Susan,

      Would you agree that these reports are help in improving the time and documentation done by providers? Do they see your rounding and training as valuable? I’ve supported go-lives where I’ve observed workers as yourself come in to address charting corrections to providers and most providers are not happy with those discussions.

  • Give an example (at home or at work) where you’ve needed to determine the “critical path” to get something done, even if you didn’t use a PERT chart to do it!
    Running the production cycle of a magazine involved finding a critical path to the print deadline. When I first started, I adopted the system of the person previously in my role and found several steps with idle time. For example, after a story was edited and ready for art, the production editor (me) would place the story in a red folder with a label indicating the InDesign file name and page numbers and then place it on the desk of the art director as an indication that the pages were ready to be laid out. The idea was that the art director would lay out the folders in the order he/she received the folders. The problem was that many other publications at the company used email to send changes and instructions, so the art directors would check those first and make them. Another problem with the folder system was that art directors would often have questions about the document or handwritten instructions, which would further hold up the process. After a few cycles of barely making the print deadline, I started emailing detailed instructions to the art directors. They were able to reply quickly with questions and work on the documents earlier in the process. Removing the red folder step saved a lot of time and removed an unnecessary step.

    • Hi Colleen – I agree that it can be very challenging to move a task when there is a lot of back and forth. By eliminating the task that took the maximum time on the critical path, you quickened the entire process. You replaced it with emails – which facilitated quicker response time.

    • Colleen, glad you found a way around. I had a very different experience while I was working for an advertising agency. The agency was very quick to execute the tasks, but the client look a lot of time to get back with approvals. For instance, if we pitched an idea to the General Manager, he would seek inputs and changes from all stakeholders possible. This would lead to the approval time to go up from a day to a week or even more. Quite a few of our projects used to be delayed due to this, as execution would start late, or approvals in between the execution would take time. It was really hard to estimate how long the project would take. Most of the time, we would end up waiting for the client’s approval, or switch between accounts to avoid downtime.

      • Hi Sid,
        I met the same problem while I worked in TV station for government event planning before. Even though the team was efficient to execute or revise the tasks, our counterparts in government took time to integrate their ideas for feedback. The team eventually figured out that sometimes our clients could not even address their need precisely. In order to improve the situation, besides the formal proposal for clients, the team would also provide the designs in the past for other government department as the samples. We also conducted schedule check in every two days and sent out the progress document for every party involved. After the approach, the team reduced the rate of delay.

  • When you’re asked to give estimates of how to get something done at work, how do you go about doing it? What methods could you employ to “narrow your confidence interval” and provide a more exact estimate?
    I had to give time estimates to clients when we began new projects. I would use a mix of hard deadlines and past experience to estimate the time it would take to complete a project. For a catalog project, I would use the date when the catalog needed to be sent to customers as the hard distribution date and work backwards to identify a start date. Once I had the start and end date, I would use my past experience to set time frames for each step. I would then create two timelines: one for the client that gave me extra days between each step in case anything went wrong and a second one for the internal team with earlier dates in the hopes that we would pleasantly surprise the client by completing the project early. I was able to narrow my confidence internal with more experience as I recognized different types of clients. For example, some clients needed constant communication and review of our materials, which added time between each step. Others only looked at the final product, which allowed us to complete internal steps more quickly but could add time on the back end. Identifying the type of client allowed me to plan when I would need extra time.

    • Colleen-
      I think having two timelines is a great idea. Did you experience any conflicts in trying to reconcile two different timelines? Did team members slack off on the internal deadline because they knew they had more time according to the client timeline?

      I think using client types as a way to manage time is also a great idea. It seems a bit hard to quantify, and as you said definitely seems like an experiential endeavor. Do you think establishing more concrete time allocations for client types would have been beneficial, or is it truly an individualistic meter?

      • Peggy, I actually had a mental third timeline for myself that was the real timeline (which sounds crazy when I write it), but I knew that many team members did slack off knowing the real timeline so I had to adjust for their behavior. Some team members were reliable regardless of timeline so I would share real dates with them. Much of running production was figuring out personalities and who could have certain information to ensure that the project could be completed in a timely manner without staff issues or other setbacks.

        We had more concrete time allocations based on project types. I would be worried that organizing clients into “types” in a formal matter could be an HR or PR issue for the company, but it could be an interesting exercise.

    • Having two different timelines is a good approach as it would allow some additional time for you incase everything doesn’t work out as planned. In my past experiences, when I had to come up with timelines for a task allocated to me, I first divided the task into subtasks and verified if there are any dependencies on other team to complete the task. In case, I had a dependency, I would make sure to add a few extra hours than what it would actually take to finish the task. For the tasks without dependencies, I used to estimate based on how much time it took in the past.

  • Hi Colleen – I like your idea of having two timelines, one for the client and one for the internal team. I used a similar tactic in my previous job. I would have a schedule for the upper management, which included slack time and another schedule that had fewer slack time.

  • Colleen, the situation that you have narrated reminds me of a time when I got stuck at the New York airport. Not sure what the issue was that day but immigration people were randomly picking students (mostly) and doing random checks on the validity of the official documents. As part of this process the people were taken into a separate room where they were validated one person at a time. Each of the passports for these people went into a file and was kept in the order in which they were received. Although there were three people processing the passports, there were delays as well in the process. The delays primarily occurred as the officials had to answer the calls they received for a particular case that was put on hold for further inquiry and sometimes an airport staff might come to advocate for a particular individual which was way behind in the pile. Clearly, the process could have been made more efficient which would have allowed the passengers to board their next flight on time and also for others to leave the airport early.

  • When you’re asked to give estimates of how to get something done at work, how do you go about doing it? What methods could you employ to “narrow your confidence interval” and provide a more exact estimate?
    The best example that I can think of from my work experience for giving estimates of how long it will take to get a task done was when I scheduled for tournaments. A typical tournament can be broken down into 3 stages: prep work, the actual event, and clean up. The method that I would use to narrow my confidence interval was to look back on past experiences and use those to plan accordingly. If I had kept more exact data from past tournaments I would have been able to give more exact estimates, but I would typically round to the half hour. Of course their is always the chance that something will go wrong so you need to account for buffer time. Estimates would change depending on the size of the field, the amount of staff, how involved the event was, and several other factors.

    • Hello Thomas, golfing events are interesting because the club gives you a window and there is always a group that just can’t finish on time or play too slow. Does the estimate take into account like how many players attending the event and charge accordingly? I can see that depending on the size of the event one would need more FTEs per crew work on the event and cleanup.
      I agree on adding a buffer as a way to account for unforeseen issues that may surface.

  • Project to Scaleup Preclinical Drug Candidate
    Contract with CRO in Hydrabad, India
    15 Step Synthesis, 5 FTE’s includung a project manager in Hydrabad.
    Procedures Validated in Boston
    Tech transfer to Hydrabad from Boston

    Sourced in India
    Backup Supplier in US
    Sourced in India
    Backup Supplier in EU
    Sourced in India
    Backup Supplier in US
    Sourced in India 100%

    Custom Building Block(1 kilo of crystaline powder) sent to India
    Customs Building Block clearance into India

    Customs Clearance into US
    Quality Assurance Checks in Boston
    Shipping Final Drug Candidate 1 kilo

    Duration 60 days

    Project Plan draw on a white board in Boston Office
    Ciritcal path dtm 25 days for Synthesis of Custom Building Block in Indiana Site.
    Shipped to Boston for QA and then to Hydrabad.
    Project meetings Daily or as needed via Webex, Skype or Landline calls.
    On call 24 hrs.

  • While reading the posts on this blog it occurred to me that one time I think we have all attempted to calculate a critical path is for travel. International travel especially is dependent on a person figuring out a critical path. I need to be able to figure out the minimum amount of time it will take me to walk from the terminal in which I land to the terminal where my connecting flight is. If I fail to figure this out the result I miss my flight. Which may not be a big deal when traveling domestically, but can be a huge deal for an international trip.

    • Hi John,
      It is a great example. How to design the route in the trip is very crucial especially while you travel in countries with complex transportation system, such as Japan. Take the Tokyo city as an example, there are at least 5 different routes in subway systems that you can utilize for traveling from Shinjuku to Tokyo Disney land. The price and time of the routes sometimes have huge differences. As the result, Yahoo! Japan provides the route design platform in their search engine to help tourists to identify the options among cost-and time-saving. It also offers the related information of the destination and recommendation for visiting. The tool is really helpful for determining the critical path for travel schedule while you are not familiar with the country.

    • Hi John, just coming up with this thought and putting it out there, don’t you think that Google maps kind of do the job of PERT. we feed in the requirements and it gives us different paths and estimates and based on the timeline, we chose the critical path. Generally we tend to chose the minimum time as the desired path. It also gives us options, (relating to the skill level or resources available, haha), that we can chose a car, public transport or walk!

      • Perhaps I am misinterpreting critical path, so please clarify:
        I thought a critical path was relevant when there are steps executed simultaneously, and you have to determine the length of the project based on the path that takes the longest- thereby also calculating slack time and efficiency. A lot of these examples seem to be linear processes that have one path- which I suppose could be considered critical.
        I was just wondering if there was a distinction, or if the critical path is relevant in a single linear process (ie a>b>c vs. a>c; b>c)

  • 1.Give an example (at home or at work) where you’ve needed to determine the “critical path” to get something done, even if you didn’t use a PERT chart to do it!
    While I conducted marketing campaign including physical conference or event, the team mostly would use backward planning for time management. After confirming the scope of work, event location and date, the team would draw out the schedule and identify the possible critical paths based on the experience in the past. Take new branches opening event for example. The most time-consuming and uncertain procedure would be producing the print materials and event souvenirs. Because the procedure includes internal image design and production outsourcing in China, we usually added extra 1 or 2 weeks to deal with delivery delay or other urgency. Then according to the time reserved for outsourcing, the company would also identify the internal critical path to allocate the staffs to make sure that we can complete the design and get sample products in the shortest time. The main strategy is to assign all the team members and shift staffs from art design department in brain-storming as well as design stage. Even though in marketing filed, determining the critical path may not be that crucial for cost in one event, the work method is still important to be adopted as the model for time management.

    • Pei – I like this example of forcing yourself to at least recognize that there are a series of sequential steps needed to carry out almost anything! My entre into quasi- project management started with the idea of moving backwards and i think this is good practice for business novices so that they can recognize the number and series of steps needed for a particular outcome. This is especially helpful in identifying questionable steps.

    • Pei Yen, confirming the scope and details is an important first step, otherwise you will have many setbacks during the process. Additionally, though planning for delays seems like it would be interfere with the quickest path, it helps in the real world when the unexpected happens. We often did this during print product of the magazine, because sometimes an advertiser was late submitting an ad or we would have a backup in the art department. Building in that extra time allowed us complete projects on time (or early if somehow there were not setbacks).

  • 1.Give an example (at home or at work) where you’ve needed to determine the “critical path” to get something done, even if you didn’t use a PERT chart to do it!

    Creating a new EAP (component record) to use as an option for ordering a new medical supply in the system.
    1) Pull the next available procedure record from the build tracker
    2) determine if there is a similar record that exists to mirror the new component after
    3) determine if this order will be orderable, performable, and/or chargeable
    4) will there be any default settings for the order status (normal, future, STAT)
    5) add synonyms for easy search options
    6) submit built to Change Control for approval
    7) get assigned a Change Control # to present new build
    8) present new build at Change Control
    9) Approval given
    10) add new build to import spread sheet
    11) new build gets imported into the Build environment
    12) test new build
    13) request migration of new build into Production

  • 2.When you’re asked to give estimates of how to get something done at work, how do you go about doing it? What methods could you employ to “narrow your confidence interval” and provide a more exact estimate?
    Personal experience in the past is the first thing that I would examine. Then the internal record is also a good resource to look at. If the company already has particular goal in time or cost, these two resources will set the baseline for me. Additionally, I would also reach out to related stakeholders and generate their opinions. For estimations of project with large scale, I will also consult with third party experts. Even though I can always find information or method online or in some professional social media, I will not take it as main source unless I confirm the materials with reliable resources.

    • Hi Pei,

      I think these are all good starting points for estimations. Do you think you would find some barriers with the stakeholders, company goals, or cost given to estimate a timeline?

      • Hi Brinn,
        Good question!
        I did find some barriers especially dealing with stakeholders and cost given. In marketing filed, all the company sponsors hope to gather the largest impact with the lowest cost. When I received a scope of work with short budget or nearly impossible timeline, the best strategy for dealing estimation is to provide another the design with smaller scope along with the original estimation to address the backup plan. As for dealing with stakeholders, it is more about politics in the organization. Sometimes, they did have the skills or knowledge to shorten the timeline. However, they may not want to share because they are afraid to lose their “power” or simply lose their face. I also tried to be careful about not only taking one person’s opinion while I conduct the estimation. Sometimes the stakeholder cannot see the whole pictures and just answer the question from their position.

    • Katty, i like your enthusiastic approach to consult different sources to produce your exact estimate. But don’t you think that all these will be required while you are gathering and eliciting requirements and early information about the project/job. Based on all the information collected and the rescued and infrastructure available, you will be able to determine the estimate and the timeline. Let me know your thoughts in this !

      • Hi Snigdha,
        You are absolutely right. I think the difference about the approaches comes from what kind of project do you estimate. In software or product development project, the project manager may have more time to gather and clarify the requirement about the job. However, in most of the marketing projects or government bidding program that I participated, the team may need to work under the condition without clear requirement and budgets at first. Take the government bidding event for example, the first requirement posted on website may only have the event goal,date, location, budget and limited conditions. The team needs to come up creative detail in the short period to defeat our competitors in bidding round. And even we can get the bidding with the template/model that we used multiple time. Our clients always come up something new after we won the bidding.Nearly in all the event, parts of the “creative” design such as combining interactive technology in the event were totally new to the team. As the result, it was common for me to do the estimation in this approach.

  • A personal and recent example of determining the critical path involved a painting project that was to be carried out by my two teenage sons. I embraced this project as a means of helping them learn a skill so i participated very minimally 🙂 Interestingly, the older son wanted to just plunge right in and “get it over with” whereas the younger son intuitively knew there would be a process. As you all likely already know, the most time consuming part of a paint job is in the prep work. Knowing that we were hosting Easter dinner, and that they would be off school Thursday and Friday, I started to help them plan two weeks ago. The PERT chart was of course in my head but I had them write down the tasks and plan accordingly: remove the existing paintings/nails, remove certain furniture pieces, prime walls, spackle/sand holes, and tape woodwork. given schedules, we certainly needed the 2-week prep time! Most of the painting was done by Friday afternoon with just touch-up work needed to be done on Saturday. The job was well done and hopefully they both learned the value of sufficient prep time and prep work.

    • Maria, so glad to hear that your sons excelled in the paint job and that it involved PERT methodology 🙂 Also, to see an example thats not work or IT related is so refreshing. You were acting as a manager and your sons as employee and you managed the work and started planning two weeks in advance. Unknowingly, we always follow this approach in our day to day work and i especially do it in planning my vacations (even more enthusiastically!). We plan in advance and start prepping by gathering and accumulating things that we might require. We have a list which can be compared to (requirements) documents. We set a budget aside for the vacation and things involved in that. And after the vacation , we set up an album (similar to user manual)

    • Maria-
      I think this is a great example of time management/project allocation. Can you clarify how you determined the critical path and utilized slack time, if at all? Did you have your sons working on projects simultaneously, each with different time commitments? Or did you work as a group linearly?

    • Thats a great example Maria. Outside of my professional work, I went ahead and volunteered for an non-profit organization where I gave a session on India and its culture to children. To ensure that children get enough exposure out the session I started the preparation a week earlier to list out the tasks that I want to do as part of the session. I outlined a presentation and then further divided into different sections which revolved around the customs, dance forms, foods etc while the other task was painting a butterfly. I ensured that I wrapped up the presentation followed by questions on time so that children can get enough time for their paintings. All together it was a nice exposure and my preparation helped to deliver what I had planned.

  • When you’re asked to give estimates of how to get something done at work, how do you go about doing it? What methods could you employ to “narrow your confidence interval” and provide a more exact estimate?

    In my first company, we used to follow the Microsoft tool but customized by my company to provide any estimates for the upcoming project/enhancement/change. All the work was IT software development related, hence over the time the company did evolve with certain standards that helps us be in check and determine the place and the category where the task/job will fall into. Based on the work that used to come in and the requirements associated with the work, we used to determine how critical the work is. Also, based on the requirements we used to have a basic idea that what would be the size of the project. If it was an enhancement, but exceeded the company set standard hours, we would negotiate it to get it as a project rather than an enhancement ticket. If it was a change ticket or project, the requirements and the modules that the change will effect and work involved in the effected module would determine the criticality and the size of the project. Generally the project would of small, medium, large size. The hours would be determined by selecting the type of modules under work and the type of changes or additions to be made in those modules. The hours would also (help/suggest) determine the skill level of the resources and the management people involved. The addition of buffer time was pre set with the amount and the level of work involved. So ideally, there wasn’t much to play around with the timelines and the software usually gave estimates that were “ideally” accurate and the buffer time keeps in within the intervals!

  • I had to create a timeline for aggregating data related to a certification proposal I am working on. The timeline was based on my assumptions of how long it should take to collect the relevant data, and on what I was told was readily available information. Needless to say, this method was extremely inaccurate. The model did not take into consideration other high priority projects that the team was working on in conjunction to this project.
    I could have asked for a weekly time commitment from the team that would provide a better context in which to budget time. I could also create soft deadlines for certain deliverables, which would have encouraged the team to keep up with the timeline outlined. However, I am a volunteer working with upper management, so my deadlines would be easily disregarded if other commitments arose.

  • Working on group projects requires a consideration of the critical path. Often, our GMBA group delegates duties. Each duty is not equal, and people work at different speeds. In coordinating our internal soft deadlines, we consider whose piece will take the longest and schedule accordingly. Where there is slack time, group members can work on other group projects in attempts to balance out the overall work load.

    • I agree working on projects with others divides the work and with each person their working speed is different as you state. I have seen this occur in my work place where working together is key to get things done and keep up with the work level for the day.

  • 1. When you’re asked to give estimates of how to get something done at work, how do you go about doing it? What methods could you employ to “narrow your confidence interval” and provide a more exact estimate?
    At my previous job there was some data entry that I had to do using Microsoft Access for a social group project. I was given pages and pages of information that had to be documented into the Access system. I broke down how many pages I could do on the days and the times I was working so that I had a target of how many to get done each day I was at work.

Leave a Reply