Calibrate your plan based on real effort

At the start, I mentioned that you need to base your plans on something sustainable. You start out ethusuaitac and over time that will wane. Life just gets in the way, especially for those with families and deadlines at work that encroach into your study time.

That’s why capturing your time spent and reviewing your projections is so vital. The certification isn’t insanely difficult, but it is a fair reflection on your study effort. I know that there aren’t shortcuts because I tried by just looking at exam papers to start with – without mapping out a training plan I’d have no way of knowing where to spend the breadth and depth of my time.

Having said that bear in mind that the training plan is MY training plan, based on my experience and speed of learning. So the correlation between Goals and Weeks is 1-1. Some of you may be more adept at picking up the concepts than I was. Some of your may have busier workloads so cannot commit the time. My takeaway point is this:

Treat each goal section in the training plan as a distinct block of learning, but adjust the “Weeks” to reflect your own learning speed. If you can fit more in to each week, great. If you need longer, no problem, just move out your projected finish date.

Which takes us on to the estimated vs actual part of our training plans.

Estimate your hours in advance, and record what actually happened

In the training plan, I’ve included my first week of study as a record of what happened.

Read AWS Overview to understand what AWS is12AWS Overview FlashcardsAt least 2 hours to create flashcards from notes – completely underestimated this
Create flashcards to cover what core services there are23AWS Overview Flashcards
Create flashcards to cover architectural patterns12Best Practice Flashcards
Read Architecting for the Cloud AWS Best Practices whitepaper, February, 201611.5Cloud Best Practice Flashcards
Flashcards for regions and Availablity Zones13Region and Availibility FlashcardsInclude time for flashcards from previous weeks
Stretch – read security whitepaper10Security FlashcardsDidn’t get time for this – was a stretch goal but good to see how I’m over committing

So in week 1 I thought I’d only spend 7 hours but I ended up spending 11.5. So I need to be a bit more conservative with my efforts. The notes were useful reminders of what ended up happening, not just for this but for any future study projects I might have.

Ok so once we have the estimated vs actual for Week 1, then what? Then you can do either of both of the following:

Estimate the entire training plan (Waterfall-ish)

You have a rough idea on your margin of error for your first week. So you could now put your more conservative estimates into each part of the training plan, and this would tell you how many estimated hours of study you have before you’re ready to start doing practice papers. The advantage of this approach is that you can see in one one go how many hours you’re likely going to have to commit. Divide these hours by the average hours per week you’ll commit and you have the time until you’re ready for practice papers.

Estimated weeks till ready for papers = hours estimated / weekly study commitment

Let’s say you’re already fairly experienced in AWS, so your estimates are lower than mine were at the time. Your first week backed you up too. Your already understood most of the concepts so the flashcards were easy for you to remember.

You estimated week 1 at 7 hours of study, and that’s how it came in. And since you’re doing AWS on the job, some of your study time you’re effectively getting for free. With that in mind, you reckon you’re getting around 10 hours a week of study in when you combine it with dedicated some of your spare time.

For argument’s sake, we’ll say that the estimated hours across all the tasks in the plan added up to 9 hours a week on average. That gives us 9 x 16 = 144 hours / 10 hours per week.

Estimated weeks till ready for papers = hours estimated / weekly study commitment

So we’re looking at 15 weeks till we’re ready for practice papers (144 hours total/study commitment of 10 hours per week)

Giving us 14.4 weeks, which we’ll round up to 15.

This is good for a rough estimate, but like anything with waterfall, long term projections can be hopelessly inaccurate. Life gets in the way, enthusiasm wanes, some tasks are harder than others and they’re not completely offset by easier ones. There are plenty of reasons why you might want to be a lit more flexible in your planning.

Monthly review of progress (more agile)

Another approach you might want to consider is to fill out your estimates for the first month only. Then at the end of the month review the accuracy of your estimates, and see how you are tracking, and whether any different problems to previously are cropping up that are steering you off course.

This has the advantage of being a more agile – you are getting quicker feedback on how you’re tracking and can constantly review whether you think your commitment is realistic enough for you to finish the certification. It might keep you more motivated to be reviewing your plans and feel that your plans are more realistic as they are tracking your pace more regularly.

The disadvantage of this approach is that you don’t have an end-date upfront though. Each month might be seen as another month without a definite end in sight. If that’s the case for you, maybe a hybrid approach would help.

Plan the initial end date, and review monthly

This is the approach that worked best for me. I originally set out 16 weeks of study planning upfront but got interrupted a few times as I didn’t really appreciate the commitment involved. When I started reviewing monthly it was frequent enough for me to remember some of the life events that had led to me underestimating the commitment.

With the monthly review I and the original end date in mind, I found it easier to visualise the consequences of the events of the week. For example, within the first month it was easy for me to see that unless I found some quiet focused time at the office, 16 weeks wasn’t going to happen. It was a case of either “be ok with it slipping out 2 weeks” or “find a way to commit the extra time if I want to stick to the original deadline”.

Having that feedback loop between recent events and their consequences was very empowering for me. Incidentally, in all of this, I don’t think anyone should beat themselves up about their plan slipping, that’s why they’re called estimates.

Now you have your forecasting, and techniques for calibration. The last part is to set up your spaced repetition reminder system so that we can keep interleaving previous material with new material and build our knowledge.