Project description

This Turkish election portfolio covers the collaboration work between the BBC Near East visual journalism team and the BBC Turkish service ahead of the 2018 poll. This was a snap election, called with just six weeks notice, requiring the team of 3 journalists and a developer, designer and data specialist, to create a suite of content with little time for prior preparation.

In the lead up to the poll we created animations, still graphics and interactive features to explain the election process and the stated policies of parties and Presidential candidates taking part. We also created a dashboard to show the results of both the presidential and parliamentary contests.

Playing with the maths: We set out to explain the complex voting system used for the parliamentary poll in, using a simulator game that allows users to play with hypothetical results data to create their own vote outcomes. This was designed to explain in a playful, interactive way how the mathematics of proportional representation works, using the D’Hondt seat allocation system. BBC Turkce promoted this in the run up to the election alongside other explainer content, including a video showing how to use the new multiple choices ballot paper.

Cutting through the confusion: In the 2018 elections, people could vote for a party-alliance rather than a single party for the first time, leading to a complex re-allocation of ‘second alliance votes’. Out interactive product explained this too, using a worked example of vote counts being re-allocated to the parties inside an alliance.

Menu of manifestos: We also offered digestible formats of party manifestos through a clickable guide to key policies published ahead of the polls. In addition a searchable election glossary, offered an alternative to producing fairly dry content in a regular text article.

Result! After the election we produced a dashboard showing the 2018 results, comparing them to the 2015 parliamentary and 2014 presidential polls. We also dug into the 2018 data to compare presidential results with the vote share of each candidate’s party.

What makes this project innovative?

Our aim was to cut through the considerable complexities of the election system. Apart from the video explainers of aspects of the voting process, the interactive guide to the proportional representation system was a truly innovative approach, taking the user through the complex calculations made to determine various methods of vote allocations. We did this by giving the user a tool that ‘does the maths’ and sliders to play with the numbers and try out various results outcomes. We also offered personalisation, giving users the choice of their own region, and to calculate possible results for the seats available in that area; in that way we were putting the audience right into the story, allowing them to play with the data. For the election results, other news agencies in Turkey visualised the 2018 outcomes, mainly through using a map. We decided to lead with data visualisations and offer a set of comparative graphics - showing both the difference between the presidential and parliamentary performances of parties in 2018 and also the difference between the party performance in 2014 and 2018. This allowed people to see the effect of the newly formed ‘Good’ Party, who were not present in the previous election.

What was the impact of your project? How did you measure it?

Turkish elections:

All our interactive and associated digital products performed way above average consumption and interaction rates (measured at between 3,000 and 5,000 page views). The results data for 5 of our featured products is below, expressed in website page views and associated engagement times plus Instagram Story views - the main social platform we focused on during the election.

(Figures compiled July 4th 2018, based on internal data anlytics tools, using Chartbeat/Comscore)

Seçim 2018: 24 Haziran seçimleri sözlüğü: Jargon Buster published June 12th),

Website page views: 103,752, 1:06 min average engagement time

IG stories: 15,600

Oy pusulası nasıl geçerli ya da geçersiz sayılacak?: Voting card explainer, (published Jun 19th) Website page views: 95,577, 0:47 secs average engagement time (68% of video viewed on average)

IG Stories: 15,300

Parti ve ittifakların oy oranına göre sandalye dağılımı nasıl yapılacak?: Result simulator,

(published June 20th)

Website page views: 82,821, 0:26 secs average engagament time

IG Stories: 14,000

Seçim 2018: Siyasi partilerin beyanname ve manifestolarında neler var?: Manifesto guide (published June 1st)

Website page views : 44888, 0:31 secs average engagement (No IG)

İnteraktif: Karşılaştırmalı seçim 2018 sonuçları: Results Dashboard,

Website page views: 30,514, 0:36 secs average engagement

IG Stories: 10,400

The broader impact of the project was manifold:

In user experience terms we offered our audience some new story telling formats, and fresh angles on a story obviously widely covered by local and international competitors. It helped us to establish a distinct voice in a crowded digital field, and underlining our role as an independent news source at a time when impartial media are under immense pressure in Turkey. Products like the jargon buster, the manifesto guide, the voting system explainers and data-driven story formats gave depth and credibility to our wider analysis.

Internally the election allowed us to try out new techniques and formats that were re-used for other elections by other teams, for example the manifesto guide, which went on to perform very well in the Brazilian elections in October and was turned into a reusable component that now can be quickly deployed.

The pre-election content performed above average with a high engagement times. Additionally, the content and graphics performed very well on social media, and were picked up by other news agencies and ultimately provided assets for the live coverage of the election night.

Source and methodology

Data and details about the voting system were collected from the Turkish electoral commission; their data bases were also scraped for historical/comparative results data. 2018 voting data came from the Anadolu Agency, the only officially accredited source. However we obtained reference vote data we could use for social graphics on election night, sourced from independent monitoring project, Adil Seçim ( Data on party manifestos was sourced from the political parties themselves.

Technologies Used

We used different tools such as HTML/CSS/JS - vanilla JS, Illustrator for graphics, Python/pandas for data analysis or Sketch and Marvel for prototyping. We also test all our products to make them accessible. Great teamwork was at the core of our success. We couldn’t have made it if our project management wasn’t solid. Slack was a great tool to communicate. But we also made sure we sat together closely for periods of time, embedding our visual journalism team members with the Turkish service team and catching up in brief update sessions on a regular basis.

Project members

Josh Rayman, developer Alice Grenié, UX and data designer Leoni Robertson, data journalist İrem Köker, journalist, data journalist Osman Kaytazoğlu, journalist, video Journalist


Additional links


Click Follow to keep up with the evolution of this project:
you will receive a notification anytime the project leader updates the project page.