3 simple ideas for selecting the best product features

No matter how big and developed your application or product is, as a Startup Owner, Product Owner, or Product Manager, you need to face similar dilemmas all the time:

• how to pick the right next features,
• how to decide which ideas are worth pursuing, and which are not.

From my own experience I know how extremely hard it is to shape a good MVP, set appropriate priorities for your product’s backlog items or even decide which project we should invest in to achieve a positive ROI (Return on Investment) indicator.

Business decision making is almost always accompanied by conditions of uncertainty, at the same time the more information the decision maker has, the better the decision will be. Today’s article contains a description of three methods, that might facilitate your decision-making process and will help you evaluate and find ideas worth pursuing.

Make a registry of your ideas

The first and most important action to begin the decision-making process is to write down all your ideas. In agile software development you call this a product backlog which is a prioritized features list containing short descriptions of all functionality desired in the product. However, no matter whether or not you create list of features, ideas for a new product, or to-do list, the process of writing down your backlog alone can free up your memory, and allows you to be less stressed and more concentrated on crucial activities.

Usually, there are far more ideas than capability ready to implement them. In our startup the list of features that we would like to deliver for our users has more than 30 items. Our pace and ability to add new features usually gives us more or less one item per month, so we need to carefully make decisions, based on clear criteria, which item is the one that we want to focus on. Below you can see the list of features that we keep in CayenneApps as a registry of features to deliver in the future.

When we’ve built our registry, besides naming the idea, we’ve also added a short summary of the concept and reasons why we think that this idea is worth pursuing. It might seem superfluous, but we’ve noticed that when the item has been in our list for a longer period, sometimes we can’t even remember what was hiding under that specific title and why we thought it was a good idea to implement. Needless to say, usually such ideas weren’t worth pursuing, but that’s a separate story.

Moreover, we always try to add to each feature a description of expected results after delivering it. This type of information has at least two advantages. The first, when we look for a specific feature that gives us the desired results (like increasing frequency of user returns), this information is the first filter, which narrows down the possible choices of features. The second, when we finally deliver each feature and then measure its results, we have important data to compare our estimation with real values. This is great material to analyze and learn more about our assumptions and avoid the same mistakes in the future.

When we come to prioritization, first we always specify which goal we need to achieve in our product and then we set objective parameters that will give us a chance to compare each feature with others, using direct comparison and asking questions like:

• Is this feature more important/useful etc. than that feature?
• If I could choose only one of these two features which one would I choose?
• Are the results that we can receive from one feature compared to the another more promising?

Usually this helps us to open the discussion and then make a decision whether we should deliver that feature rather than another one.

In the end, we need to pick one feature, make even small improvements and measure the results. If the idea looks promising we implement the next features, but if the outcome is far from our expectation, we right down our conclusions and with new knowledge we once again look for the feature that might give more promising results.

When simple comparison is not enough

Sometimes when decisions are bigger and more complicated, and the comparisons aren’t easy, we use a SWOT analysis process to organize our thoughts and put Strengths, Weaknesses and Opportunities and Threats in one place.

The analysis allows us to understand our features better — both in terms of the known facts (Strengths and Weaknesses - “here and now”) and expected results after delivering a particular component (Opportunities and Threats - “there and then”). These four categories focus our minds on searching the advantages and disadvantages of each potential solution. On the one hand, we try to estimate how long it will take to deliver a feature and how big and positive an impact it will make on the user. On the other hand, we look for potential threats that might show up after delivering a product.

As we wrote previously: “The idea of listing features, evaluating the most important ones and identifying connections ignites the fire of creativity. People begin doing market research, look closer at their current competitors or notice annoying liabilities within their companies which were unnoticed at first.”

To make the process easier and more data-driven, we can often take advantage of different tools available online. We collect the information from our users using applications such as SurveyMonkey or UserVoice. Additionally, in the background, we verify the current patterns of users’ behavior using tools such as Google Analytics or Kissmetrics. We can also try to predict how the feature being currently examined fits into current trends. In order to do this, we use simple applications such as Google Trends or more sophisticated products focused on social media monitoring.

For instance, Brand24, a popular social media monitoring tool created in Poland, gathers all public mentions regarding chosen keywords. This tool can be used to show what and where people write about different brands, products, and services.

It also significantly extends the ability to collect the information from the previously unknown users by monitoring customer complaints or suggestions for improvement in posts and comments outside of the previously defined communication channels.

Brand24 can also give us helpful insights about the scale and popularity of each feature that we want to deliver. Moreover, social monitoring tools help to verify, if there are already products on the market which are similar to ours and allow us to monitor feedback about them.

Many times, from opinions available online we can find out what the most popular feature is, what people love about it or what they literally can’t stand. This type of data can not only enable more discussion during the SWOT brainstorming session but also enhance the final outcome by finding new inspiration or noticing the lack of something essential that we have missed during the discussion. The ideas derived from real data are always more powerful than the ideas created out of thin air.

Select, evaluate, connect

Of course, as we have written many times, the sole identification and examination of different attributes of features or products is not enough for the Product Manager. We need to evaluate and prioritize.

Even though the primary form of SWOT analysis does not include prioritization, in the extended analysis the prioritization is essential. It allows us to compare different features and eventually pick out the most important ones. In this phase, we also come back to the previously defined goal which shapes the further development of our application. This goal facilitates the process of selection and evaluation of the features that we have written in each category (strengths, weaknesses and so on). We also check whether e.g. strengths and weaknesses have a connection with opportunities and threats or whether the connection between them is relatively weak. These pieces of information are presented in the Summary Step in the SWOT process which gives us helpful suggestions to make the final decision.

Furthermore, the SWOT analysis that is written down has also one more significant advantage. When we’ve already delivered the particular feature, and we’ve gathered feedback from our users and analytical tools, we can come back to the previously completed analysis and compare the collected results with the expected Opportunities and Threats that we previously filled out. The differences are sometimes surprising to us, but nevertheless, we have great material for further analysis. Sometimes we even expand the existing SWOT analysis with new data and repeat the process of prioritization and evaluation for the new reality.

The simplicity and flexibility of SWOT analysis lets us do it without extra effort and many times this is the opening for brainstorming discussions about our new ideas.

Final thoughts

Whether we use a simple prioritization based on the registry or more powerful tools such as SWOT, we should focus on removing the randomness from the process.

Features must be designed, developed and promoted, and the resources, especially in smaller businesses, are always finite. Therefore, we must make sure that we are not only delivering random features, but that we actually deliver the right features with the biggest value to our users and companies, and that we do it all at the right time.