Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Story-focused Reading in Online News and its Potential for User Engagement

We study the news reading behaviour of several hundred thousand users on 65 highly-visited news sites. We focus on a specific phenomenon: users reading several articles related to a particular news development, which we call story-focused reading. Our goal is to understand the effect of story-focused reading on user engagement and how news sites can support this phenomenon. We found that most users focus on stories that interest them and that even casual news readers engage in story-focused reading. During story-focused reading, users spend more time reading and a larger number of news sites are involved. In addition, readers employ different strategies to find articles related to a story.

We also analyse how news sites promote story-focused reading, by looking at how they link their articles to related content published by them, or by other sources. The results show that providing links to related content leads to a higher engagement of the users, and that this is the case even for links to external sites. We also show that the performance of links can be affected by their type, their position, and how many of them are present within an article.

This work co-authored with J. Lehmann, C. Castillo and R. Baeza-Yates has been published in the Journal of The Association For Information Science And Technology (JASIST), available online in May 2016. The work was presented at the Yahoo TechPulse Annual conference in December 2016.

  • Be the first to comment

  • Be the first to like this

Story-focused Reading in Online News and its Potential for User Engagement

  1. 1. 1 Story-focused Reading in Online News and its Potential for User Engagement Janette Lehmann, Carlos Castillo, Mounia Lalmas, Ricardo Baeza-Yates
  2. 2. Story-focused news reading 2 •  While reading news, users may become interested in a particular news item, and want to find more about it: – to obtain various angles on the story – to overcome media bias – to confirm the veracity of what they are reading – … Users read multiple articles about a particular news development or event.
  3. 3. Story-focused news reading 3 •  Previous studies found that – many users visit established news outlets to confirm a story, no matter from which source the information initially came from (The Poynter Institute) – some users find links to related information on a news article page important (Pew Research Center) Users read multiple articles about a particular news development or event.
  4. 4. Story-focused news reading 4 •  News sites recognise that users want to further inform themselves so they: – provide information on different aspects or components of a story – link to other articles published by them and other sites Our goal is to understand the effect of story-focused news reading on user engagement and provide insights into how news sites can support it. Users read multiple articles about a particular news development or event.
  5. 5. Questions 5 •  Does story-focused news reading exist and to what extent? •  What are the characteristics of story-focused news reading? •  How do links on news articles promote story-focused news reading? •  How does story-focused news reading impact user engagement? Users read multiple articles about a particular news development or event.
  6. 6. One-month Yahoo Toolbar data 6 •  A news story is a collection of articles related to same news event (TD-IDF term weight in article and cosine similarity with threshold 0.4) •  On average, each story has 14 articles (median 8), is covered by 7 news sites (median 5), and has 2,482 users per story (median 758) 65 popular news sites publishing articles in English; 4.9M news reading sessions; covering 2,536 stories comprising 25,703 news articles •  Percentage of users engaging at least once in story-focused reading is 16% •  Avid news readers more likely to engage in story-focused reading: – 64% of users with at least 15 reading sessions have at least one story-focused session – But being an active news reader does not imply engaging in story-focused reading
  7. 7. Questions 7 •  Does story-focused news reading exist and to what extent? •  What are the characteristics of story-focused news reading? •  How do links on news articles promote story-focused news reading? •  How does story-focused news reading impact user engagement? Users read multiple articles about a particular news development or event.
  8. 8. Story-focus news reading exist •  shuffle test: alternative dataset with same distribution of session length, but with random articles •  probability of story-focused reading is x4 larger and probability of multi-site is x2 larger in actual dataset than shuffle dataset 8 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 0.0 0.3 0.6 0.9 10−4 10−3 10−2 10−1 Prob. of story-focused reading Density Actual data Shuffled data ● ● ● ● ● ● ● ● ● ● ● 0 1 2 3 0.25 0.50 0.75 Prob. of multi-provider reading Density Actual data Shuffled data ● story-focused reading does not depend on popularity of a story, its number of articles, or number of news sites that cover it
  9. 9. Questions 9 •  Does story-focused news reading exist and to what extent? •  What are the characteristics of story-focused news reading? •  How do links on news articles promote story-focused news reading? •  How does story-focused news reading impact user engagement? Users read multiple articles about a particular news development or event.
  10. 10. Story-focused vs non story-focused reading 10 % sessions num. of articles % focused duration (minutes) number of sites non-foc. foc. non-foc. foc. 74.93 17.31 4.95 1.66 0.63 0.27 0.12 0.13 1 2 3 4 5 6 7 7> 15.16 29.00 41.33 51.02 60.85 67.13 78.15 3.09 5.77 8.57 11.24 13.60 15.92 18.61 21.29 6.66 9.23 12.92 15.96 18.35 21.59 27.91 1.00 1.20 1.37 1.52 1.65 1.81 1.91 2.12 1.26 1.48 1.67 1.83 1.98 2.14 2.55 •  users spend more time reading news when focusing on a specific story •  more sites are accessed when users engage in story-focused reading when session length increases (more articles are read) à probability that session is story-focused increases
  11. 11. Depth of story-focused reading 11 session % number of articles duration (minutes) number of sites in-story out-story total per-article 85.03 11.48 2.43 0.69 0.23 0.08 0.06 2 3 4 5 6 7 >7 0.89 1.09 1.45 1.67 1.80 2.67 3.05 6.67 10.48 14.29 18.23 20.09 23.09 25.03 3.34 3.49 3.57 3.65 3.35 3.30 2.79 1.27 1.53 1.79 2.05 2.31 2.36 3.19 Per-article dwell time increases for sessions with <= 5 articles: users spend time reading articles they are accessing Per-article dwell time decreases for sessions > 5 articles: users are skimming articles, as articles contain increasingly more redundant information number out-story articles increases as session depth increases deeper story- focused sessions are longer à involve a larger number of sites + and mostly made of in-story articles
  12. 12. Questions 12 •  Does story-focused news reading exist and to what extent? •  What are the characteristics of story-focused news reading? •  How do links on news articles promote story-focused news reading? •  How does story-focused news reading impact user engagement? Users read multiple articles about a particular news development or event.
  13. 13. Links •  Type of link, Position of link, Number of link •  Performance metrics: •  Popularity: percentage of links of a particular group •  Performance: how often the links in the group are clicked (percentage) 13 2 3 4 5 6 7 >7 0 20 40 60 80 100 Number of in-story news articles %ofupstreamtraffic Internal Article (Int) Non-Article (Int) External •  75.45% of articles have inline links •  6.4% of links in article are inline links inline links
  14. 14. Link performance: Type of links •  Internal links within articles promotes story-focused reading •  Users tend to click on links that bring them to other news articles within same site 14 -20 0 20 40 60 80 100 Internal Article (Int) Non-Article (Int) External % Articles with this type of links Popularity (%) of this type of links Performance (%) of this type of links apart for multimedia, no external links with performance comparable to internal links to news articles
  15. 15. Link performance: Position of links 15 −0.2 0.0 0.2 [0,0.1[ [0.3,0.4[ [0.6,0.7[ [0.9,1] Position in article text Linkperformance Linkpopularity Link popularity● Link performance ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 10% 20% 30% best performance with links at end of articles good performance for links in upper 20% and 40% of article (multimedia) low performance for links at top articles
  16. 16. Link performance: Number of links •  < 10 inline links may be wasting an opportunity, as users may be enticed to click to access related content •  > 10 and < 29 links does not result in more clicks, but spreads the clicks more •  > 29 links may actually harm user experience 16 0 5 10 15 20 [0,2] [9,11] [18,20] [27,29] [36,38] Numberofclicks Number of inline links in article 0.0 0.5 1.0 1.5 [0,2] [9,11] [18,20] [27,29] [36,38] Number of inline links in article Avg.clicksperlink is 10 links per-article the “sweet spot”?
  17. 17. Questions 17 •  Does story-focused news reading exist and to what extent? •  What are the characteristics of story-focused news reading? •  How do links on news articles promote story-focused news reading? •  How does story-focused news reading impact user engagement? Users read multiple articles about a particular news development or event.
  18. 18. User engagement •  50 news sites, 57K users,1M sessions with users with at least one non-story- focused reading session and one story-focused reading session •  Engagement metrics: – Dwell time: time spend on news site – Absence time: % of sessions with absence time < 12 hours on news site 18 •  story-focused reading vs non story-focused reading •  internal story-focused reading vs external story-focused reading
  19. 19. 19 User engagement: Dwell Time internal story-focused sessions: •  dwell time higher compared to non story-focused sessions •  average increase in dwell time is 50% external story-focused sessions: •  no observable effect on dwell time News provider Dwelltimepersession Non-focused Focused Ext-focused
  20. 20. User engagement: Absence Time 20 internal story-focused sessions: •  for 78% of sites, probability that users return within 12 hours increases by 68% external story-focused sessions: •  for 70% of sites, probability that users return within 12 hours increases by 76% News provider p(absence12h) Non-focused Focused Ext-focused
  21. 21. Story-focused news reading exist …………and 21 •  Internal inks within article promotes story-focused reading and keeps users engaged –  longer period of engagement (reading sessions are longer) and earlier re- engagement (shorter absence time). •  External links does not have negative effect on user engagement –  period of engagement remains the same (reading sessions are the same), and re-engagement begins even sooner (shorter absence time). •  Important to use links properly, in terms of quantity, type and position.
  22. 22. Future work 22 •  We need to account for the novelty and more importantly the quality of the related content •  The online news ecosystem has changed, both in terms of how news sites provides news and how users inform themselves

×