Doug Den Hoed | AtTask Blog

About Doug Den Hoed

I used to subscribe to a magazine called Fine Woodworking. My favorite part was a section called Methods of Work. Craftsmen would share their techniques and tricks on how to make woodworking more enjoyable. The elegance of their solutions fascinated me! Now, I'm no master carpenter. But I am getting pretty good at using @task. If you're an enthusiast who enjoys learning how it's done, I hope you'll enjoy the Methods of Work I'm sharing.

As Soon As Impossible

26 February 2011 - POSTED by Doug Den Hoed

Faster than ASAP?

Enterprise Project Management Software is helping teams accomplish their Project Based Work more efficiently than ever. Most Projects are designed to be done As Soon As Possible (ASAP). The PPM Software that teams use may even have email alerts to expedite Tasks that can start early.

But a colleague recenty posed an interesting challege: "Is there a way to easily reset the start date of successor tasks when a task completes early, essentially setting planned dates to the projected dates? I thought this would be down by looking at recalculate timeline, but this does not seem to do it."

The Handoff Date

Imagine a Project with Task A, Task B1, and Task B2 (successor of B1)

ASAI1

The @task "Handoff Date" is a Task level attribute that notes the earliest date that Task could be started. When a Task (A or B1) has no predecessors, its Handoff Date is equal to its Projected Start Date. When a Task (B2) has predecessors (B1), its Handoff Date remains null while its Predecessors are < 100% complete.

ASAI2

Once all of a Task’s predecessors (B1) are complete, @task populates Handoff Date of the successor Task (B2) with the Actual Completion Date of the predecessor (B1).

ASAI3

At this point, the @task Event Handler would kick in and send emails, if configured to do so; however, most of our clients do not use the email Event Handlers (gets too spammy).

Do The Impossible

I fall in the "Revise Planned Dates as I get Smarter" camp myself, so — heresy or not — if you are also interested in synchronizing the successor Plan Dates, I have a solution. Change the successor Task Constraints to FIXT, and the Planned Start Date to the predecessor Handoff Date. The effect is that AtTask stretches the successor earlier in time, visually highlighting the opportunity to compress the timeline.

ASAI4

If you like this approach, you could also create a Task report with a filter that restricts to Tasks where Actual Start Date is null (not started), Handoff Date is not null (could start), and Handoff Date is earlier than Planned Start Date (could start earlier than Planned). Run the report daily, and do the impossible, as above. Our clients liked it so much that we even used the AtTask API’s to automate it for them.

0 Comments »

How’s My Portfolio?

6 December 2010 - POSTED by Doug Den Hoed

TSN Turning Point

Our local sportscaster uses the phrase "TSN Turning Point" to highlight the pivotal play that enabled the win. In my Project Based Work, I use it to refer to that pivotal event when someone I’ve been coaching thinks of their own solution in a way I wouldn’t have. I always like to make a big deal of it, both to encourage them for taking ownership, and to celebrate their progress.

Chris Connoly, one of my @task clients, sent me just such a TSN Turning Point, and has given me permission to share it here.

My My My

@task has many "My" features built in to help users with Work Managment, such as My Tasks, My Reports, and My Issues. However, where Chris works, the emphasis is on the PORTFOLIO portion of the Project Portfolio Management Software, which doesn’t really mesh with the "My" concept. Here’s his story, and his solution:

We have all of our projects grouped into Programs and then Portfolios, and we have Program Managers and Portfolio Managers. It’s very easy to create project reports based on the Portfolio Manager or Program Manager since they are only one level up. However, it seems pretty much impossible to pull out task information based on those managers.

Since most of the information that we want to report on is generated in task reports, not being able to have a "My Program" or "My Portfolio" on task reports is actually a huge problem. And it annoys me that I can see who the Managers are in the view, and group by them, but not include them in a filter.

[One option would be to] create a filter that uses project IDs with an "IN", and then just include all the project IDs that you want. Not exactly useful for a large number of projects, and still doesn’t take care of the "My" part.

So what I ended up doing was creating calculated custom data fields at the project level (in my ever expanding "For System Use Only" section) for "Program Manager ID" and "Portfolio Manager ID", and using "Program.Owner.ID" and "Portfolio.Owner.ID" as the formulae.

I can now just use DE:project:Progam Owner ID=$$USER.ID for my filter, and I have the My Program filter, [without] even having to use text mode.

An Elegant Solution

In short, Chris was able to use @task’s Custom Parameters feature to point "up" to the data that he wanted, effectively extending the data model to suit his needs. I’ve always admired AtTask’s stance that (I’ll paraphrase) "We can’t think of everything, so we’ve included Custom Data so you can tailor @task to what you need". That’s a collabrative invitation that works.

0 Comments »

Get Some Professional Help!

18 November 2010 - POSTED by Doug Den Hoed

Logo Loco

If you’ve been using @task as your Project Portfolio Management Software for a while, you might remember the old layout, where you could put your own company’s logo in the top left corner. As the administrator, I used to keep a couple of versions of my logo handy. I’d set the regular one for normal operations, but set a black and white one when I was doing maintenance. It was a way of broadcasting to my team.

Although the top left logo is no longer an option — to be fair, that really is the primest location of them all — I’ve figured out a slicker way to broadcast announcements. And share Help Files. And get my logo back.

Click Help

There are lots of steps in this one, but if you scroll down to the screenshot, I hope you’ll agree that the end justifies the means.

Write your own help manual (e.g. in Powerpoint; you’re on your own for this one)
Create a new Project (e.g. Help Docs)
Attachments > New Document > upload your help manual
Attachments > View Documents > Click help manual > Hover on Preview for ID (e.g. 816046)
Paste Link into Helpdoc.HTML (e.g. see Internal Help Manual href)
Repeat previous 3 steps for Company logo (e.g. see img src)
Test Helpdoc.HTML (e.g. layout, links should work)
Attachments > New Documents > upload Helpdoc.html
Attachments > View Documents > Click Helpdoc.html > Hover on Preview for ID (e.g. 816117)
Create > External Page > Help Page, with defaults, plus URL (e.g. https://lumina.attask-ondemand.com/attask/documentDownload.cmd?boolean%28viewinline%29=true&ID=816117)
Create > Dashboard > Help Dashboard, with Help Page as the only report
Note ID of Dashboard from URL (e.g. 118321)
System Settings > Preferences > General Preferences > Help URL (e.g. https://lumina.attask-ondemand.com/attask/dashboardRun.cmd?ID=118321)
Click the Help link within @task to view your new Help Dashboard/External Page/Document

Download Details

You are welcome to download the simple Helpdoc.html file from this article and modify it for your own purposes. You could share status updates, broadcast upgrades and outages, link to a training video, or even make a virtual team bulletin board. Just remember that each time you upload a new version of Helpdoc.html, you’ll need to update the ID in your Help External Page. It took me six versions to get this one just right <grin>.

0 Comments »

Rekindling Burndown

27 October 2010 - POSTED by Doug Den Hoed

Is the Campfire Cold? Match Burning Down

Back at the 2009 AtTask User Conference, there was quite a bit of buzz around Agile Project Management tools. One concept I like was a burndown chart: pick some Tasks, choose a timeframe, then use a picture to confirm you’re on track to meet your goal.

I haven’t heard much about burndown charts since then, so I wonder…has the fire gone out?

A Spark of an Idea

Most modern Portolio Project Management Software does a good job of highlighting current state: what has happened, is happening, and will happen, from a particular current point in time, at the time it occurs. A burndown chart, however, has a historical state to it, allowing Project Managers to compare progress over time, not just at the time. That’s the tricky part. However using some concepts I presented in my Project Management Time Travel post, I’ve had some success. Here’s a picture. There’s a groovy animated browser version, too.

Burndown Example

Warming Up to the Concept

To be fair: this was just a prototype. But I’m considering taking it further. If you are interested in building Burndown Charts in @task, I’d love to hear from you. Specifically, how would you suggest I prioritize these alternatives?

When running a burndown chart, the majority of users will want:

A. To choose a single Project and generate a burndown chart run for all Tasks in the Project for the entire duration of the Project

B. To choose a single Project, provide a sprint name and associated date range within that Project, then generate a burndown chart run for all Tasks in the Project that are Planned [vs Projected] to finish within that date range

C. Same as B, but based on Projected to finish within that date range

D. To multi-select one or more Projects, provide a sprint name and associated date range, provide a Task Category, then generate a burndown chart run for all Tasks with that Category across those Projects, providing the Tasks are Planned [vs Projected] to finish within that date range

E. Same as D, but based on Projected to finish within that date range

You can either leave comments so others can read them, or email me directly at doug.denhoed@gmail.com.

Thanks in advance!

2 Comments »

Face The Interfaces

13 October 2010 - POSTED by Doug Den Hoed

Mirror Mirror Interfaces

One of the happy paradoxes of good Enterprise Project Management Software is that the more information users find and believe, the more information users believe they will find.

That extra information often lives in other tools; and rightly so. But to be effective, it often needs to be visible along side our Project Based Work. For those of us managing such Project Management Tools, this leads to an inescapable truth: eventually, we’ve got to face the Interfaces.

A Spectrum of Options

Modern PPM Software offers a plethora of choices when it comes to data integration. Recently, I had a client ask me to help them decide which route made the most sense for them. So I took the opportunity to build up this Interface Options matrix in Excel:

The Requirement section:

Has a Description of the overall objective (e.g. To Push @task Hours to an External System Twice per Day)
Lists each Option in increasing degree of technical difficulty (e.g. Manual Push/Pull by User, Email Pull to User, Email Pull to [Email] Agent, Webservice to WebService, or WebService to Database)

The Interface Steps section:

Lists the functional steps in the interface
Notes for each Option which resource has responsibility for each step

The Comparison section:

Lists the difficulties in the interface (e.g. Reliances, Manual Efforts, Programming Efforts, Costs of Developing, Costs of Operating)
Uses a somewhat subjective (but rather telling) High (Red) / Medium (Orange) / Low or None (Greeen) score of each Option against those difficulties
Boils each Option down to its most appropriate use

The Cost Estimate section:

Runs over 2 years (to allow for a more fair comparison)
Lists the assumptions for Hardware, Development, and Operating Costs
Computes the total costs for each Option

Upon Reflection

Once I set up my Interface Options, it was easy to use the same criteria for a second requirement (e.g. To Pull Cost Information from an External System into @task Twice Per Day). Having the two options on two tabs in Excel let me flick between the two scenarios. While doing so, I noticed some interesting comparisons, and it forced me to rethink a few assumptions.

Now that it’s set up, I think that my Interface Options matrix will help me provide better estimates more quickly. You are welcome to download it and change it for your own use.

0 Comments »

Going the Extra Milestone

22 September 2010 - POSTED by Doug Den Hoed

Choose Your Path

Most Project Management Software has the ability to mark certain Tasks as Milestones. Some PPM softwares also offer a set of customizable Milestones — a "Milestone Path" – to encourage standardization across Projects. A few sophisticated Project Portfolio Management Software packages even let you pre-assign a Milestone Path to Tasks in a Template, ensuring consistency across Projects.

However, what if the detailed Tasks required to support the work plan don’t align well with the Milestones the Project Management Office (PMO) wants to see? Or what if Projects encompass multiple, similar Milestones across large parts of work? Even in some of the most advanced Project Management Tools, the built-in Milestone features simply can’t accommodate these requirements.

Fortunately, there is a different technique that not only addresses these limitations, but opens the door to a whole new way of presenting Project Information in a Gantt Chart.

Trial Run

The screen-shot below represents a complex Project. There are lots of detailed Tasks required for execution, but in a way that doesn’t align very well to the Milestones the PMO needs to track. Furthermore, the Project plan Branches at Stage 2 into Parts 1, 2, and 3: similar sub-Projects with different timing and effort, but the same Milestones across them. A high resolution version and sample MS Project file are also available for download.

Going The Extra Milestone Example

The shaded portion at the top represents the normal Project Details. The bottom half is the new concept: a Project Summary of what people outside the Project most care about. In this case, it contains a summary "Highlights Task" for Part 1 (red), 2 (yellow), and 3 (purple). Beneath each Part are three "Milestone Tasks": PSum Stage 1, 2, and 3. Each has a token Duration of 0.1 Hours for percent complete calculations, and zero Work required.

The vertical lines highlight the trick. The PM drags constraints from the appropriate Tasks in the Project Details — regardless of where they are buried — to the corresponding Milestone Tasks. PSum Stage 1 and 2 are Start-to-Start constraints, and PSum Stage 3 is a Finish-to-Finish constraint. As the constraints snap the Milestone Tasks in place, they roll up and drive out the footprint of their Highlights Task.

This Project Summary technique offers several interesting effects and opportunities for you to:

Plan the Project Based Work as needed, decoupling it from the PMO Milestone requirements
Meet the need of some complex Projects to show repeating Milestones across sub-Projects
Quickly "collapse down" to a simplified view of the Project Highlights
Search for the Milestone Tasks alone across multiple Projects, and view them in Gantt format, removing the noise of the Project Details
Leverage the resulting predictable Percent Complete values (0%, 33%, 67%, and 100% in this example) to perform additional logic — in @task, for example, you could put Custom Data behind the Highlights Task and add a Calculation Parameter to compute which Gate each Part is at over time
Create additional Project Summaries based on other Tasks for different viewers

Crossing the Finish Line

I hope the Project Summary technique inspires you to rethink how you structure Projects in your Project Management Software. I should note that, in the case of @task, there is no built-in way for the Milestone Tasks to automatically complete when their controlling Predecessor completes. I’m working on it…but that’s another article.

1 Comment »

Bar Tending 101 (no table)

8 September 2010 - POSTED by Doug Den Hoed

A Toast. "To our new guests: The Business"

Those of us who have worked with Business Project Management Software have sampled from its vineyard of features. Over time, we acquired a taste for what we like, learned to avoid those we don’t, and ultimately settled in on our favorites.

As PPM Software has matured, The Business has come to appreciate how they too can enjoy these Project Management Tools. As Project Management professionals, wouldn’t it be decadent if we had time to play the role of connoisseur, sharing the subtleties of our experience as we tour The Business around?

Well, sober up. There’s never that much time.

That said, may I offer to at least fill your cup with one popular Project Management concept I’ve uncorked?

From Sippage to Slippage

Let’s switch from Wine Country to Oil Country. On Oil and Gas projects, somewhere in the middle of the Project Plan, there’s a Task where the well itself gets drilled; what’s called the Spud Date. It’s a key event in the entire project, and since lots can go wrong either side of it, The Business often watches the Spud Date to quickly gauge how the Project is going, typically in a Gaant Chart view.

1. If all Tasks are ASAP, a Project’s Start date determines where the Spud Well (S) will occur. This is typically what it looks like in the future, while we’re planning ahead, with no Tasks started yet (0%).

2. As time moves on, real work occurs and some Tasks get completed (100%). However, until a Task is marked as totally complete, @task pushes everything that’s ASAP past it further into the future, assuming that the earliest you would mark it as complete is "Today". The dotted yellow bar shows what’s CAUSING the delay, and the other shading shows the EFFECT of the delay.

3. If you realize you forgot to enter an end date, you can backdate a Task with the real actual end date (A). When you do, @task "snaps" the schedule back to that date, but then assumes (again) that any late Tasks can’t finish any sooner than "Today".

4. Since it’s the Actual Tasks, the fact that is "Today", and the ASAP relationships that drive where the Spud Well occurs (S), there is NO POINT in adjusting the Start Date of the Project — it would simply make the overall Project start earlier, without moving (S).

Last Call

I hope you can use this analogy to understand or explain why ASAP Gaant Charts move the way they do once a Project gets underway.

You are also welcome to download the original Excel version. It includes a similar example for a Project with a Start No Earlier Than restriction. For Oil and Gas, that is often the case when waiting for ground to freeze before the drilling rig can get into position.

For a Winery….hmm…please leave a comment with your own analogy.

0 Comments »

Can your PPM software defuse a bomb?

10 August 2010 - POSTED by Doug Den Hoed

Spy vs Spy When Companies Merge

Synergies…Efficiencies…Economies of scale….that all sounds great! Why wouldn’t two big companies want to merge?

Well, one reason might be that if they’re in the same industry, they probably use a lot of the same software, and sorting out which packages to Retain and what to Decommission is an enormous undertaking.

It’s a good thing that one of my clients knew about AtTask when they needed to sort it all out.

The AppRat Team

Our Applications Rationalization Team (or AppRat, as it’s more commonly referred to) was charged with sifting through 4000+ software applications as two major companies merged. Initially, we tried to use a combination of a small website database and spreadsheets stored in Livelink to keep track of things. It didn’t work, as this diagram shows, so we converted to @task.


Before: Web DB/Livelink/Excel - Limited access to Web DB - Restricted to 99 custom columns - Daily extract to Livelink - Repetitive data reformatting - Data staleness caused confusion		After: @task - Multi user environment - Unlimited custom parameters - Common view to all users - Reusable custom reporting - Real time reporting for all audiences

Ownership vs Usage

The AppRat team set up a single Project with a Task for each of the 4000+ apps they needed to either Retain or Decommission. Behind each such app Task, they also tracked over 100 custom data parameters to help guide the Work Management. Often, apps were used by more than just one department. And sometimes, the Owner making the Decision to Decommission an app wasn’t the same as the User(s) wanting to Retain it.

Hence the bomb scare that lead to this post.

Always cut the Red Wire

With all that data flying around, it’s easy to see how certain scenarios could be overlooked. Recently, our executives challenged us to prove that we were only Retaining what was Used, and that we were Using what we Retained. Huh. Sounds obvious…but we really weren’t sure how to show it. So we decided to build a dashboard to look for exceptions.

It worked. We can now spot and defuse explosive exceptions right away. There’s a screen shot below with a real example.

The moral, though, is that by leveraging @task’s flexible custom data and exceptional custom reporting, we used our IT Project Portfolio Management platform to solve one of the more challenging aspects of the merger.

That’s dynamite.

0 Comments »

PPM Performance Testing For PMs

27 July 2010 - POSTED by Doug Den Hoed

Food for Thought

It was over 20 years ago, around 8 pm. Three Business Analysts and two fellow coders crowded around my desk with the sinking realization that it was going to take an all-nighter to pull off our deliverable. Just then, the senior PM poked her head in with a concerned look and a "What can I do to help?" We shot a round of blank looks at each other, but before we could even formulate a polite reply, she nodded quickly, and said: "Right: I’ll go get some food."

Surely, figuring out what the team needs to keep them going efficiently is one of the most important skills a PM can offer.

But what happens when your team’s Online Project Management itself isn’t keeping up? The prospect of assessing the performance of sophisticated Web-based PPM Software is pretty daunting, even if you are a technical PM.

This spreadsheet can help you performance test your On-Demand Project Management environment and keep your team going:

Peek Performance

The spreadsheet has several practical features:

a cell (B2) of the maximum allowable time for any test (e.g. 120 seconds)
a cell (B5) to define the Date and Reason for the tests
a column (C) to define and describe three tests
a column (D) to define two physical locations for each tests (e.g. Office, Offsite)
a column (E) to define four different types of connections (e.g. LAN, Proxy, etc.)
a grid (F3:K16) to record the results of the tests across two environments and three browsers
three graphs — one per test — with fixed Y-axis scales that illustrate the test results
two lines across all graphs dividing acceptable, borderline, and unacceptable performance

You too can take your own Performance Testing. Download the spreadsheet, update the ranges above to suit your tests, and then each time you want to test:

duplicate the tab and date it
update the Date and Reason (B5)
clear the results grid (F3:K16)
run the tests and enter the new results (F3:K16)
use the tabs to "flip" between the tests, effectively animating the graphs

The performance tracking spreadsheet is simple to use and easy to understand. If certain aspects of your Project Management Software are not running as quickly as your team needs them to, you — as PM — can take the initiative to quantify what’s really happening, initiate a solution with your vendor, and keep your team going efficiently.

But food’s always appreciated, too.

0 Comments »

Sufferin’ SaaS-afras! Where can I Test?

6 July 2010 - POSTED by Doug Den Hoed

ExSaaSperation

Online Project Management is a great fit for Software as a Service (SaaS). Teams can perform their project based work in different locations and timezones, but still collaborate in a loosely coupled way. And as vendors improve the underlying PPM Software, everyone instantly benefits when the production website is upgraded. What more could you want?

Well, when a major new release that changes the user interface is coming down the pipe…how about a place to test?

DiSaaSter

About a year ago, AtTask wisely decided to give its @task ondemand customers a beta site so they could test new features against known data. The weekly beta refresh worked so well that some customers also use it for training purposes. Views, reports and other objects copy across automatically, and because the data is in a separate environment, it’s a safe place for new users to get some practise.

However, the next version of @task (R14) is a major look-and-feel release. Suddenly, those customers have few options for training new users:

There is no Test environment as such, where the current release points to a copy of their data
Beta is different enough that it could cause confusion
Even if they’d agree to pay for a data restore, they can’t copy their data back into the ondemand environment, since the primary keys would collide
They could create what’s called a Testdrive with sample canned data, but then none of their dashboards, their reports, their data — their business, really – would be included

But fortunately, there’s an easy solution.

SaaSisfaction

The trick is to build a Test environment within Production. Seriously!

Create a special Group called Training Data, and a special Access Level called Trainee, with rights to see Home Group data only
Set all new users’ Home Group to Training Data, and their Access Level to Trainee
Create several Projects, Tasks, and Issues that belong to the Training Data Group; or better yet, keep several side-by-side versions; or best of all, prepare them in Kickstarts, so you can delete and reload them each time a fresh class of new users shows up
Have the new users log in: all they will see is the Training Data
As new views, reports, dashboards and so on are added to the Production environment, they can be instantly made available to the Trainees — no migration required
Because it’s really Production, the Test environment will get the same care and attention as real data does as far as uptime, performance, support and backups (whereas beta can be down unexpectedly or intentionally for days…true story…)
When training is complete, simply upgrade their Access Level and change their Home Group to graduate them into Production

It sounds a bit daunting at first, but by leveraging @task’s security model, you can provide a safe, refreshable Test environment within Production that stays current with your business as it evolves. Who knows? If enough of us adopt the approach, it could become the defacto standard.

If you decide to try it, please let me know how it goes.

0 Comments »