How to create a data-driven product strategy in 2021
Understanding how to create a data-driven strategy is key to driving and building success. We spoke to Stu Neilson our Product Strategist to discuss the key areas you need to be aware of when developing a strategy.
Firstly, what is a Product Strategy?
Product strategy defines what you want to achieve, provides context around the market you are operating in, and guides a large theme of work that will help you accomplish your goals and vision. It is a system of achievable goals and visions that work together to align the team around desirable outcomes, both for the business and for your customers.
The key to starting a product strategy is to have a clear overarching vision of what you want for the business and create a proactive long-term plan, instead of reactive marketing strategies.
Where do you start?
At Rawnet we start every product strategy with discovery which is where we identify the business objectives, current pain points for the business and its customers, and the opportunities for growth. The discovery phase entails diving deep into our client’s customer’s journey, figuring out who they are, what makes them tick, what stakeholders know about their customers, and forming in-depth detailed personas (find our more about our process here). We can identify areas where we need to conduct further research into customers’ pains and needs, enabling us to create opportunities for growth within the business and marketing strategy.
Where does data come into it?
While defining the purpose and challenges, we ask the obvious questions; what does success look like and how do we measure that success? Whether it’s an increase in sales, increasing repeat spend, or generating more MQLs, we map them all out and (if available) include any comparison points we can refer back to, to see if the delivery has driven the growth expected.
A combination of opportunities gathered from the customer, stakeholder, and competitor research and the KPIs mapped out allows us to start formulating more detailed objectives. We use OKRs to do this (Objectives & Key Results). The principle of OKRs is to ensure there is a clear goal, along with a defined way of measuring the performance of that goal. OKRs inform everything from UX, design, and development through to the ongoing marketing acquisition work and ensures that all deliverables are mapped out within a way that will allow all involved to identify 2 things: That all deliverables have a purpose and that we can clearly see what is and isn’t working with the product.
There is also a need to identify the systems which allow us to report and analyse performance. Starting with the basics such as Google Analytics and the required events and goals needed, moving up to a Data Studio with more customised reporting and collating different data points, and topping it up with site engagement monitoring such as Hotjar. There are a lot of tools available and ways of tracking insights and data and which are most effective would be determined by what we’re looking to measure. That’s something we identify during the process. Building your data insights and analytical knowledge is an integral first step but it’s also critical that there’s a home for it all and a strategy for what to do with data. This is where MarTech (marketing technologies) comes in which is necessary to stay relevant in 2021. When you consider the customer journey, you need to have automated efficiency to adopt and scale marketing efforts faster and smarter. But that’s a topic for another day!
A more specific example is the data our CRO (conversion rate optimisation) team found with a client’s site’s product page views recently. For numerous categories, the product page views were much lower than we’d normally expect to see with an eCommerce site. We found this was a unique challenge posed by the unique products they sell. Customers are visiting the listings but if that one particular item isn’t there, the rest of the site isn’t of interest to them (not straight away anyway). This led to new product releases such as saved search updates and stock notifications based on preferences. This was where we can notify customers when that particular item is back in stock and we’re hoping they return on their own.
What are goal loops?
Goal loops are key to a data-driven product strategy. By regularly checking success, having all involved agree and aligning tasks directly against them provides insights and loops back increasing successful performance.
This stage is key in having a successful project, by sharing your findings, results and understanding their implications on the project will improve the decision-making process.
By talking to the internal project team and the stakeholders and ensuring they are happy with the progress and results keep everyone on the same page. It also gives stakeholders an opportunity to understand in-depth how the project is going to benefit the business and leverages feedback from all areas of the business. We handle this through regular insights sessions to pick out recent performance including top-level stats and also granular.
What if there isn’t enough data to inform decisions?
Sometimes releasing an MVP (minimum viable product) in the most efficient way is needed, purely driven by the experience of those in the business and intuition. An MVP provides an opportunity to release something quickly, gather some insights, and fine-tune from there. Using customer and stakeholder feedback and competitor insights initially before fine tuning the strategy as the insights start rolling in. Once insights start to open a little, changes can be made - whether it’s something simple like introducing a book a demo workflow, or something more involved like a new product category. But in short, having no data to support decisions shouldn’t put a complete stop to continuous improvement - it just changes the path forward.
Begin testing the success of the new strategy. There is no set of guidelines for strategy success; every business is different, and so is the audience. Testing is the best way to gain a concrete understanding of what is working and what is not through data.