How to create a data strategy for a marketplace

elisa garrido
9/4/2021

In this post we share with you the steps we followed when articulating and implementing the data strategy of one of the clients we have recently worked with. The client's name is My Spalist. It is a vertical search engine and directory of services in which people (or users) who want to have a massage look for professionals (freelancers or companies) to contact them. In turn, these professionals will register themselves in order to appear in the users' searches.

Myspalist home

Most of the projects that fall into Minimum's hands are developed with Webflow. We consider this tool as the best NoCode tool for developing webs. In this case, and before starting to develop the data strategy, we will implement in Webflow the Google Analytics pixel through Google Tag Manager.

Preliminary considerations

When introducing and defining a project, we usually take two things into account; the first is the type of business model we are talking about and the second is the stage in which the project is at.

1. Business Model

In this case, in terms of the business model, we are dealing with a marketplace, i.e. a digital marketplace where both suppliers and end users meet and interact. In our case, the dynamics of interaction between suppliers (hereinafter customers) and those who receive the service (hereinafter users), will be carried out through:

  1. A subscription by the customers who will then appear in-
  2. A directory that end users will be able to access to choose the service they need

2. Project phase

As for the project phase, given that the project started from scratch, having completed the ideation phase but without any configuration yet -neither design nor construction-, in this case it was an initial phase.

Defining the data strategy of the marketplace

Once we have defined these two aspects of the project and with both in mind at all times, we can begin to design our data tracking strategy. Having a data strategy will help us extract learnings to improve the project performance. In other words: they will help us track and improve any data informed project.

These are the steps we follow:

  1. Measurement plan: in which we define main and secondary objectives and determine how we are going to measure whether they are being met or not.
  2. Tagging and configuration: where we detail all the implementations that have to be carried out in order to get the data we need. We will also explain the configuration we have performed in Google Analytics.
  3. Data visualization: what we deliver to the client so that he can read and understand the data.

Phase 1: Measurement Plan

In the measurement plan we define and seek the objectives that will orchestrate our entire data strategy. Once we have defined them, we have to specify the metrics we will need to be able to measure the achievement of the objectives. We will call these indicators KPIs or Key Performance Indicators.

1. Objectives

We distinguish two types of objectives: Firstly, the main objective, the one that "defines" the project and its raison d'être. Secondly, the objectives of the digital asset, which in this case is the website.

1.1. Main objective

Given that we are in the first phase of the business, our main objective will be to validate the business idea. That is to say: to guarantee that the idea is of interest to the users, to the people.

Objective

1.2. Digital asset objectives

The secondary or digital asset objectives will be those that "unravel" the main objective at a more concrete level - referring to the digital asset, which in our case is the website.

The objectives of the digital asset will be divided into two paths; those related to customers and those related to users.

These objectives will be:

  1. Acquisition: the attraction of customer and user traffic to our website.
  2. Retention: the recurrence of those customers who have already signed up for our subscription model to your account.
  3. Conversion: customer registration in our subscription model and end-user contact to our customers.

It is important to visualize the objectives in detail, with numbers and dates. For example, one of the acquisition objectives could be translated as "at least 30,000 users enter our website every week", the retention objective could be translated as "50% of our customers enter their account twice a week" and a conversion objective could be translated as "to have 10,000 subscribers in our database by the end of the first quarter".

2. Key Performance Indicators

The KPIs or Key Performance Indicators are the data that tell us whether the objectives we want to achieve are being met or not.

For a better understanding, we will divide the main KPIs into two sections. On the one hand, the KPIs related to the user and on the other hand, those related to the customer.

To properly get all these data we have combined different types of tools:

- On the one hand, the Google analytics suite tools (Google Analytics, Google Tag Manager). With these we obtain data such as the number of users who come to our website or the devices and sources from which they do it,...

- On the other hand, the data from our database -Memberstack-. We sent the data from Memberstack to a dynamic Google Sheets document and connected it to Google Data Studio. From Memberstack we obtain data such as the type of customer plan or the customer profiles.

We will also collect additional data that we consider interesting but not mandatory to determine the achievement or failure of our objectives.

Phase 2: Tagging and configuration

The tagging guide is basically a document that lists what tags should be implemented and where on the web page. It can be understood as the "instructions" or the modeling of the data to be implemented on the web page.

Not all the data we want to extract is given to us by Google Analytics by default. Let's take a concrete example: in this project, Google Analytics tells us when a user enters a specific page and we can even know the name of that page. However, to understand the name that appears in Google Analytics we must give that name to the page externally. Another example would be that of clicks: if we want to know which specific element a user has clicked on (for example, to know how many users access a form from a certain CTA or to know which CTA is working best).

1. Tagging Guide

In this particular labeling guide we will define two types of variables:

1.1. Page variables

The page variables are used to group the pages by their nature. For example, we can group all the pages that refer to legal issues, or all those that are informational.

If we group them together, we can filter the records in our subscription system and know if the information we give has any effect on conversion. If not, perhaps it would be a section of our website that we could improve.

1.2. Event variables

Interaction event variables refer to those actions that users take on our website and we are interested in identifying. For example, when they click on a certain button, or if they open any of the FAQs.

Since events in Google Analytics can contain three different data layers - the event category, the event action, and the event tag - we will be interested in specifying at this point what information we want each of these data layers to bring us.

In the example you will see in the picture below, we are talking about a therapist contact button. Since we have other types of contact buttons, we will give as a category to these contact events with the therapist the name Contact Request Clicked.

Event Variables

In addition, as you can see, this button has three variants: contact via call, via text message and via email. These three variants will be included in the event action. Finally, since we are interested in knowing whether the user has tried to contact an individual therapist or a SPA, we will collect the type of page in the event label -which, as we pointed out when explaining the page variables, will indicate whether it is an individual therapist page or a massage center or SPA-:

Configuring the events

2. Google Analytics Configuration

As we have mentioned before Google Analytics gives us only some data by default, but not everything we are interested in. This is why we at Minimum usually configure Google Analytics in different ways for each project.

Previously, we will have introduced in our Webflow website the Google Analytics pixel through Google Tag Manager.

2.1. Configuration of objectives or goals in Google Analytics

Since we are interested in knowing how many conversions we have (customers who sign up for our subscription model), we will have to configure Google Analytics so that it considers what we want as a conversion.

In our case, we want Google Analytics to consider a conversion every time a customer registers on the website and confirms their registration after receiving the account confirmation email.

The registration process for this project is as follows:

  • First, a user enters the page to choose his account
  • Then, he selects whether to create an independent therapist or SPA center account.
  • After, he clicks on the button to register and then completes a small form.
  • Finally, the user receives a confirmation email in his/her inbox through which he/she can access the registration confirmation page.

Google Analytics configuration

To set up an objective with a sales funnel, this option must be selected when defining the goal in Google Analytics.

We will set the goal up as a destination type. The final destination will be a thank you page. We also created the different and consecutive steps as shown in the screenshot above.

Thanks to this, we will be able to visualize in Google Analytics our objective in funnel mode and we will know where users are dropping:

Funnel Analytics

Phase 3: Data visualization

Finally, to facilitate the understanding of data, we have created different dashboards in Google Data Studio with different topics to group the results in a logical way.

The distinction of dashboards we have made is as follows:

3.1 A general traffic dashboard that will collect all the visits to the website and their details.
3.2 A general traffic dashboard that will collect all the visits to the website and their details.
3.3 Two customer dashboards (as we mentioned before, customers are those who pay, in this case SPAs or independent therapists.)
  • One that collects leads: leads are requests for "something" that we give on our site. In the case of our clients, a lead will be the visitor who completes the goals we have set for him. In this case, we will consider as leads those customers who register, verify their account and/or activate their profile.
  • Another one that collects the interaction events: there are basically three steps or interactions that our customers can follow, 1. register, 2. verify their account and 3. activate their profile.
3.4 Two user dashboards (we define users as end customers, i.e., those who use our website to contact therapists or SPAs).
  • One that collects leads: a lead will be a user who contacts a therapist.
  • Another one that collects interaction events: there are basically two interactions that end-clients or users can do: 1. save a therapist in their favorites and 2. contact a therapist.

Did you find this post interesting? Do you want to hear about more projects we developed in Minimum data team?

If so, feel free to leave us a comment or ask us any questions or opinions you may have and we will be happy to answer them!

Like this article? Spread the word.
You have got questions, we have got answers‍
Who we are?
How long does it take to build my digital product or MVP?
Can I work with Minimum.run on an ongoing basis once my MVP or product is released?
What technologies and integrations does Minimum.run work with?
What's the working process with Minimum.run?
What are the typical deliverables?
Will I be able to manage my custom product once it goes live?

Let’s grow together

Reach a new level of growth

View
project