Alexander Beletsky's development blog

My profession is engineering

AgileBaseCamp Kiev 2011: Sergey Dmitriev: Setting up priorities with Kano analysis

Disclaimer: text below is compilation of notes I made on AgileBaseCamp 2011 conference, listening to different speakers. I do it to keep knowledge I got on conference, share it with my colleagues and anyone else who interested. It is only about how I heard, interpret, write down the original speech. It also includes my subjective opinion on some topics. So it could not 100% reflects author opinion and original ideas.

Sergey Dmitriev is a first russian-speaking Certified Scrum Trainer. On AgileBaseCamp he presented scientific based approach for prioritizing the backlog. I liked his speech since it was very new for me and could be done on practice.

The approach is based on Kano analysis with weighted factor matrix. It might sound a little bit complex, but whole process is rather simple and splitted up on 2 stages.

Kano analysis of backlog items

In very simplified version it is basically 2 questions, with 5 answers. Questions are:

  • How do you feel that feature is PRESENT in product?
  • How do you feel that feature is ABSENT in product?

And answers are:

  • I like that
  • I expect that
  • I don’t care
  • I can live with that
  • I don’t like that

Each item in backlog are put through that questions. It is important that real customers to participate that survey. Due to statistics laws, as biggest sample you have as more precise results you got. Sergey has created a web application that helps to perform such surveys, KanoSurvey.

Classification of results

Based on users answers and using some complex math behind KanoSurvey would split up all user stories up to 5 classes:

  • Have to be done (mandatory) class - features that customers sees as very important/expected and have to be done fast
  • Linear class - linear features are something that you might accomplish but still do not fully satisfy customers (like battery life of mobile phone - if it works 3 days without re-charge.. it is nice, but I would like to have 4 days. Or, if I have 1Gb of storage on new service.. it is nice, but I would like to have 2Gb etc)
  • Amazing class - leadership function, something that makes you unique and produce “wow”-effect
  • Inversed class - features that you you might have, but customers are feeling bad about it
  • Don’t care class - features that customers not interested in
  • Questionable class - these are probably mistakes, then users answering “I want, I don’t want” in the same time

The value only persist in first 3 classes of features. So, we use that in further analysis, rest of features are just thrown away are useless.

It is done. After analysis it is time for weighing.

Weighing

With Kano analysis we understood that features are important for our auditorium. Now we need to plan the feature thought releases, to do that we have to prioritize them correctly. Weighing of features helps to do right prioritization.

To perform a weighing business criteria’s have to be defined. It is very up to the product what the business criteria’s are. It is recommended to select not more than 5, that makes the process simpler. The example of criteria’s could be:

  • Uniqueness - weight 3
  • Sales - weight 2
  • Low costs - weight 1

Each pre-selected backlog item is assigned with weight.

Combining the results

Sergey has created an Excel spreadsheet (that is available here). In the Excel spreadsheet we specify backlog items, estimations in points, class of Kano and it’s weight. Based on this input spreadsheet would automatically propose desired priority. It is also possible to see value and cost of particular item. This is a priceless information for any Product Owner.

The presentation slides are located here.