5 techniques for prioritising features (or user stories) to try out

Prioritising features

“OK, so which of the features are ‘must haves’, which are ‘should haves’ and which are ‘nice to haves’?”

“All of them are must haves.”

“All of them?”

“Yep, we absolutely MUST HAVE all these features.”

“Really? All of them?”

“Yep, all of them are MUST HAVES.”

[Blood curdling scream…] (But only in your head)

Does this scenario sound familiar to you? Hopefully like me you only screamed in your head, rather than to your product owner / business owner / tormentor.

Prioritising features is a pretty important step when designing and building any product, from websites and mobile apps to toasters and lawn mowers. You need to decide exactly what the product will do and which features to include and focus on. You need to ask difficult questions, like does the toaster really need a crumpet setting? (Hell yeah it does)

Prioritising features is pretty damn important, but it isn’t easy, at least not if you’re doing it properly. Many teams use the classic MoSCow method (‘Must haves’, ‘Should have’s and ‘Could haves’) but in my experience MoSCow doesn’t force you to work hard enough. Even if you do identify the ‘Must have’, ‘Should have’ and ‘Could have’ features what are the ‘Must have’, ‘Must haves’? Which features should you build first? Why is everything always a MUST HAVE (aaaarrrrrhhhhhh)? Whether you’re prioritising features or user stories, here are 5 prioritisation techniques that you should try out.

Buy a feature

Buy a feature

Buy a feature is a great prioritisation technique because it gets people thinking about the cost of a feature verses the benefits. The idea is really very simple. Take your features and then give them all a cost which relates to the complexity of delivering that feature. For example, 50 pounds / dollars / points for one feature, 25 for another and so on. Try to ensure that features are costed relative to one another so a feature that costs twice as much is twice as hard to deliver. If you’re working with user stories and story points then you can simply use story points as your costs.

Having costed all your features you then give your product owner / users / business owner a set budget and ask him or her to go on a feature shopping spree. You’ll want to capture the features bought and ask that all important ‘Why?’ question so that you can fully understand the buyer’s rationale.

Find out more about ‘Buy a feature’

Priority poker

Priority poker

Priority poker is not only a great prioritisation technique, but also gives you a legitimate reason to play cards at work and to generally pretend that you’re a hot shot poker player in a Vegas Casino. It’s a variation on planning poker and is played in a very similar way.

Each player is given a set of priority cards which he or she can use against a feature. Cards are typically 1 to 5, with a ‘Don’t know’ joker for when someone can’t make up their mind, or needs to know more information. You then take each feature in turn and describe it so that everyone knows what they’re being asked to prioritise. Each player must then choose a priority card for that feature and place it face down on the table. Once all the players have played a card everyone reveals their choice at the same time and discusses and agrees a priority. The real power of this technique is not only the agreed priority, but more importantly the discussion surrounding that agreement (or disagreement). By placing cards face down you also avoid the HiPPo effect – the highest paid person (or most vocal) significantly skewing the prioritises.

Find out more about priority poker

Pairwise comparison

Pairwise comparison

What are your top 10 action movies of all time? (Terminator 2 better be in there). It’s not any easy question to answer because prioritising a long list of stuff can be hard. It’s hard because you have to consider a lot of things at the same time. Which is the best film? What about the next best? What about the next, next best? The longer the list, the harder it becomes until your head hurts and you need to lie down in a dark room.

Rather than trying to size up lots of things at the same time it’s much easier to compare just two items at a time. For example, is feature A higher priority than feature B? This in a nutshell is pairwise comparison. Rather than trying to prioritise the entire list you instead split it up into lots and lots of pairs (hence the name). You compare feature A with feature B, then with feature C, feature D and so on until you’ve compared every pair of items. It sounds like a long and laborious process, but you’d be surprised at how quickly each comparison can take. Each time a feature beats it’s foe you give it a point. Tot up the points for all your features and voila, one prioritised list.

Find out more about pairwise comparison

WOW, HOW, NOW, POW prioritisation technique

Batman POW

A feature prioritisation session is often rather one dimensional. The only question asked is “What is the business value?” or “What is most important for users?”. This is a problem because features are certainly not one dimensional (apart from that one-dimension conversion feature). When comparing features you should be considering not just the benefits that a feature can bring (to both the business and users) but also the cost of delivering that feature. After all, it’s all too tempting to prioritise the game changer features that are going to cost more than the Space Shuttle to build and neglect those that are going to be relatively easy to deliver, but provide only incremental benefits. This is where the ‘Wow’, ‘How’, Now’, ‘Pow’ or (or WHNP for short) prioritisation technique comes into play.

For the ‘Wow’, ‘How’, ‘Now’, ‘Pow’ prioritisation technique you map features against a vertical axis for impact (for both users and the business), and horizontal axis for cost of delivery. You can then see which features fit into the following groups:

  • WOW features! (High impact, low cost) – These are the high impact, low cost features you should be focusing on
  • HOW features (High impact, High cost) – You should be asking how to reduce the cost of delivering these high impact but high cost features
  • NOW features (Low impact, Low cost) – You should be delivering these low cost features now to incrementally improve things
  • POW features (Low impact, High cost) – Get rid of these features with your best Batmanesque ‘Pow’ as they are high cost but low impact

Find out more about ‘WOW, HOW, NOW, POW’ prioritisation technique

Feature Top Trumps

Top Trumps card game

Do you remember playing Top Trumps card games as a kid? The thrill of shoving your higher scoring card in your opponent’s face to show them who’s boss. Top Trumps is not only a great game for children, but it’s also a great game for prioritising features. Here’s how you play it.

First you need to decide your feature Top Trump categories. This will be how you score and judge features. Feature Top Trump categories might include:

  • Business goals and desired outcomes (e.g. User retention, Usage)
  • User value (e.g. Love, Like, Loath)
  • Ease of delivery (The higher the score, the easier to deliver)
  • Design principles (e.g. Simplicity, Delightful)

You then create a Top Trump card for each feature. Each card should include the feature name, description, optional picture or sketch, scores against your Top Trump categories (out of 10) and a total score. You can then not only prioritise your features by taking the total scores, but also recreate your Top Trump playground battles by directly pitting features off against each other. Let the best feature win!

Find out more about feature Top Trumps

See also

Image credits

British Women’s Open leaderboard by Wojciech Migda
Top Trumps Prototypes by Graham Holliday