March 2020
Experiment: the switch to a subscription model.
After four years of the product's existence, the team had the idea of moving the service from a completely free model to a freemium one. The motivation was to test the hypothesis of the value of the product for the audience.
The idea involved extending and simplifying the functionality, as well as working out the tariffs, analysing the user experience and rethinking the logic.
In simple words, a subscription is a periodic collection of money from the user for additional services, certain goods, as well as extending the capabilities on information and entertainment services.
When there is enough product functionality to allocate limits, paid services usually start with two or more plans. Such an approach distinguishes them from competitors that do not give users flexible choice, customisation and transitions (Upgrade, Downgrade).
In addition, companies often resort to marketing tricks to create the appearance of benefit to users, such as offering discounts or using odd prices that create the illusion of lower fees.
Odd prices are part of the psychology of pricing. Psychological pricing, in turn, aims to keep the product within the «Mental boundaries» of customers by slightly lowering the price.
We started our work on the update with a discussion. The main issues we intended to address included: Working out the interfaces with the existing functionality; Elaboration of tariffs based on our value in the market as well as that of our competitors; Generating audience interest in the upcoming update; Defining a timeline; Implementation.
We decided to start by studying direct and indirect competitors in order to identify recurring patterns that would form the basis of our own functionality.
In addition to the analysis, we needed to determine the capabilities of our payment system in order to close most of the obvious problems with users in the first version. Such as «Changing the payment data when the card is blocked or lost». In addition, we needed to find customer interaction points with the subscription and establish the connection between the subscription and the rest of the service's functionality.
When the minimum task description was formed, including system behaviour and user actions, we had all the necessary data on hand to outline the task pool.
In the process of implementing a paid subscription, we couldn't pass up the opportunity to test the impact of the name of the tariff on the purchase decision.
The decision to call the minimum tariff «Amateur» was not coincidental. According to our hypothesis, this name should have motivated users, who were mostly actors and models, to buy the «Premium» tariff, which in their mind would correspond to their level of professionalism.
Thus, by dividing users according to their professional level, we distinguished 3 tariffs: «Amateur» for beginners; «Standard» for specialists; «Premium» for professionals.
The initial prices were calculated based on monthly traffic, income, the value of the services provided and service fees. The monthly rates for Standard and Premium were RUB 199.00 and RUB 499.00 respectively. The «Premium» tariff in turn was divided into several periods: RUB 499.00 for a one-month subscription; RUB 2,990.00 for a six-month subscription (RUB 498.00 per month); RUB 4,990.00 for an annual subscription (RUB 415.00 per month).
In order for the user to see more benefit in the «Premium» tariff for six months and a year, we decided to display the monthly rate and the discount percentage. However, in reality the discount was so small that we decided to revise the prices to increase the discounts and consequently the user's average bill.
As a result, the prices for the «Premium» tariff were reformed: RUB 499.00 for a one-month subscription; RUB 2,694.00 for the six-month subscription (RUB 449.00 per month); RUB 4,788.00 was for an annual subscription (RUB 399.00 per month).
We started the full work on the subscription by working out the logic and scripting, which helped with finding and refining the states.
To make it easier to select and pay for a plan we liked, we used pop-ups which made any action falling under the free plan restrictions a trigger for purchase. We didn't redirect the user to another section; the choice was left to the user to renew the plan or to close the window and take an action that didn't require an investment.
When we started designing, we proceeded from the principle of consistency and reusing existing patterns so that people wouldn't get lost in the new functionality. After refining the sections and notifications, we assembled a general flow and conducted a corridor test for error rates and task success.
We allocated separate time to copywriting edits. In order to reduce the number of potential appeals to technical support in the future, we worked through the information in the section on payment confirmation, highlighting items such as: Plan cost, Payment period, Discount, Totals, Hints and Features received from upgrade.
We didn't forget to give time to informative popups that told you about the limitations and features of the tariff.
And we have worked on a mobile version, for easy access to subscriptions on the user side.
To significantly reduce negative feedback from users when launching subscriptions, we told the audience how useful they are and how they work everywhere.
We haven't forgotten about people who may be coming to us for the first time and people who may be questioning the usefulness of the plan upgrade, so we've added a promo code for 7 free days of the Premium tariff.
During the first days of the test, when we monitored conversion rates, service rejections and feedback received, more than 100 plans were purchased. For the available traffic, available benefits and set prices, this was an unexpectedly high result.
There was no clear gradation, one bought standard for a month, then premium for a year, then used promo code for a week, so we will be able to give a prediction about further purchases only after some time. For now, we need to test new hypotheses and not dwell on the result we have achieved.
You can see all the screens and flows that our team has developed in the Figma preview or by clicking on this link. The screens were developed before the design team switched to Auto-Layouts, which were added to Figma in November 2020.
The list of persons reflects those who provided support and cooperation in the process of developing the feature and the article. Also the list indicates all those who are responsible for the content of the article.
