Thursday, 20 November 2014

From ADDIE to SAM to Evolve in Learning Technologies

I recently heard someone wanting to roll back to a nine step or even eleven step process for producing learning materials - it wasn’t elearning, but actually that shouldn’t really make a huge difference should it as elearning is 90% learning (give or take a letter).  I was actually fairly shocked by this, as an agile proponent I’m always amazed by how failing and struggling organisations blame process and try to solve their issues by adding more process and more complexity to already overly complex situations.  The problem with being very process driven is that you can’t actually have a process that encompasses every possible situation at the most minute detail; what you in fact end up with is the old operations manuals that end up stuck on shelves gathering dust whilst relying on individuals and people doing things the way they always have.  Now I’m not saying that you should ditch all processes (although the anarchist in me would sometimes like to see it) but you need to reduce down your process and reduce the steps required to get things done if you want to make a more efficient and more flow.

A few years back the ADDIE process was considered the king of instructional design.  It was a no-nonsense common-sense approach to designing learning where we started by Analysing the problem, then Designed and Developed the learning before Implementing and then Evaluating it.  It really does make a lot of sense to have this sort of ideal behind making training - I mean can you imagine the outcome if you didn’t ever analyse or evaluate?  Yes, it was pretty much all training 20 years ago (and I bet a fair bit of it still around today) it was called ‘this is the way we’ve always done it’ training without anyone questioning the why, let alone the need for it.  The thing is though, that ADDIE is still a bit of a cumbersome beast with many hand-off stages that can cause issues.  If you apply the process in the way it is here the iterations take a long time to come about.  It’s great that you’re analysing the problem, but just like front-loading training, if you front-load your analysis things can get missed and forgotten - not to mention the likelihood that things actually change whilst your going through the process.  For example, you could be analysing some legislative piece that you need to include in training, have that all set, finish the design and start the development and the legislation changes (as legislations like to do).  What do you do know?  According to the model you continue through and this will get picked up in the evaluation - but that’s no good if we want to deliver effective training that complies with the new legislation.  Alternatively we groan and go back to the start of the project - or what often happens in reality is we do a bit of reverse engineering, chuck in the new changes and go from there.  Whatever we do, the process isn’t going to work as well as we’d like in an ideal world.  And that’s where the problem with ADDIE type processes lie for me; they rely on an ideal which seldom matches the reality of the modern world.  I’ve talked about the technological acceleration we’re currently living under, but there’s always a constant change in any business in any organisation that we need to be aware of and to tap in to.

That brings us to a more modern and agile form of instructional design SAM.  SAM isn’t just a more modern and friendly sounding buddy, but it’s a streamlined process for our ID that is a bit more responsive.  SAM stands for Successive Approximation Model which as the name suggests means that rather than design the perfect fit in the ideal world we hit it running and then we continually improve what we’ve done to get it right.  If ADDIE is a linear approach with a single cycle, SAM is an iterative approach that relies on as many iterations as necessary to achieve your aim.

My problem with SAM is two-fold, firstly it’s a model and all models are inherently relying on ideal situations to work and secondly it’s focus is heavily on the iterations and in reality two many iterations cause many projects to fail and run out of budget.  I’m not anti-SAM, I just believe that rather than having discrete iterations we should concentrate on a more analogue approach of evolving what we’re doing. For me rather than cycling around a model or working in a linear fashion we’re still working in a pre-formed solution type environment. 

All that sounds great but this blog isn’t called the Nth Degree just because my name starts with N (although… yeah… partly) it’s because we take things a little further (and sometimes a little too far!).  For me SAM ideals are great, but its still based around ideals and essentially I think we can boil that down to a single thing; evolving.  I’ve been discussing this model a little with like-minded individuals around the world (yes the beauty of Twitter) and I’m still not convinced if my own model is a great idea or if I’m just a bit more lazy when it comes to ID (… and project management and work as a whole).  Anyway, let me test it out on you and I’m grateful for any comments.  The theory for Evolve is this; you don’t waste inordinate amounts of time in complex design and analysis phases that always seem out of date by the time you complete them.  You don’t write reams of documentation about the design or indeed about much of anything.  At the same time forget about making a solution up that you have to keep going backwards and forwards over in a gazillion iterations.  Forget complex storyboarding and design docs, forget days and hours of formal training needs analysis and interviewing every man and his dog before you get going or going through complex evaluation scripts or numerous re-designs.  Instead you work together.  Collaboratively.  If it sounds a bit like SAM it is, but there’s a limit to it.  In SAM the stakeholder kind of knows what they want and then you keep doing it till you get it right in a process that is akin to trial and error and your ultimate success is measured by the stakeholder.  In Evolve the very aims are developed collaboratively so the success is shared and you reach that point together.  In my theory that means less iterations and more about minor tweaks as you work towards the goal.  It’s not just developing a storyboard together like you might in ADDIE but actually taking it back to the very essence of the story and coming up with it together.


  Yeah I know, now that I read my own stuff it comes out a bit hammy.  Maybe it’s just SAM without the catchy name, but I think it has principles of both ADDIE and SAM rolled in there - it’s not really the model that matters so much but the way we apply and work together.  If people are truly invested in a project then they have a greater interest in its outcomes - if they actually feel they’re responsible for the concept and idea then you’re already half way there.  I’ve worked in a creative team where we’ve come up with an idea together and it’s always been accepted and successful with the team involved; if you can include your ‘clients’ in that team then buy-in is a given and the iterations will reduce regardless of how you define the ‘project’.

Of course there’s a big flaw in trying to make all learning technologies projects evolve; simply put some people just don’t want to work that way with you.  Some clients like you to go away and do the work they think their money should be paying for and that their job is purely to evaluate what you do.  With those type of clients you’re probably best with the ADDIE style.  If they’re really keen to put more into the project look at SAM and if they’re really invested and want to work with you then I think Evolve is the model-less model that could really pay dividends.  Of course you can apply an Evolve mindset to an ADDIE or SAM project too - as always it’s about knowing who you’re working with and adjusting your style to match.

Evolve.  A wry grin at myself for defining a way of removing process and a more free and productive environment without having to do the boring bits of work!  Somebody is bound to shoot this down - maybe I will on one of my more process heavy days :)