How do news aggregators choose source feeds and what impact does that have on ideological balance?
Executive summary
News aggregators assemble feeds through a mix of technical plumbing (RSS/subdomains), human curation, and algorithmic selection, and those choices — driven by product goals, business incentives and data constraints — materially shape what users see [1] [2] [3]. Empirical work shows aggregated or platform-distributed access can increase exposure to cross-cutting outlets compared with direct self-selection, but audits and bias ratings reveal many dominant aggregators still concentrate sources on one side of the spectrum, producing systematic ideological skews [4] [5] [6].
1. How source pools are built: RSS, subdomains and editorial lists
At the technical base, aggregators ingest publisher RSS/XML feeds and often map content by subdomain or feed ID rather than by root domain — a key step that determines whether different brands or verticals inside one publisher are treated as separate ideological sources — and many early and current aggregators simply pull whatever public feeds are available unless editors intervene [1] [3] [2].
2. Human curation vs. automated selection: tradeoffs that shape balance
Some aggregators rely on editorial teams or community voting to choose and weight sources — a model that foregrounds human judgments and clearer selection criteria — while others rely on automated recommenders and machine-learned rankers that optimize for engagement, relevance or novelty; each approach carries distinct biases, with human curation reflecting explicit editorial priorities and algorithms reflecting implicit incentives encoded in training data and objectives [1] [7] [8].
3. Algorithmic mechanics: what recommender systems optimize for
Algorithmic selection is rarely neutral: systems use signals such as click-through rates, dwell time, personalization histories and source-level features to rank items, and those objectives can unintentionally favor polarizing or attention-grabbing outlets and amplify popular viewpoints unless constrained by deliberate diversity metrics — a dynamic documented in academic audits of search and recommender behavior [3] [8].
4. Business and policy incentives that bias source choices
Commercial aggregators operate under advertiser and retention incentives that shape source mixes: platforms seek content that keeps users engaged and returning, and as independent auditors such as AllSides have argued, that often translates into heavier curation from outlets rated on one side of the political spectrum (AllSides’ audits find many aggregators skew left), meaning balance promises can mask commercial selection criteria [5] [6] [9].
5. Measured impacts on ideological diversity: mixed empirical evidence
Large-scale observational studies find a nuanced picture: “distributed” access through aggregators, search engines and social feeds can boost cross-cutting exposure relative to direct, self-selected visits, implying algorithmic mediation is not inevitably polarizing; at the same time, other work shows algorithmic systems can raise ideological concentration for certain communities and that recommender loops can harden preferences over time absent interventions [4] [10] [8].
6. The role of self-selection and audience sorting
Even where aggregators present a diverse source pool, users exercise agency: many people self-select like-minded outlets and tune feeds, so observed ideological balance in an aggregator’s index does not guarantee balanced consumption — studies spotlight the enduring role of audience choice and cognitive comfort in shaping what people click and read [11] [4].
7. Transparency, audits and competing framings of “balance”
A recurring problem is asymmetric transparency: researchers argue platform curation algorithms are opaque and editorial criteria (subdomain aggregation, source whitelists, weighting rules) are rarely public, which allows audits and partisan critiques to reach divergent conclusions — for example, AllSides publishes percentage tallies and bias ratings that frame many aggregators as left-leaning, while other services emphasize breadth or neutral ranking policies [5] [6] [3].
8. Practical consequences and the hidden agendas to watch for
The combined effect of technical choices, business incentives and user behavior is that aggregator source selection can create emphases, blind spots or feedback loops: publishers rewarded by engagement get more distribution, political actors can exploit weak vetting, and platforms that tout neutrality may be pursuing retention metrics that implicitly privilege particular outlets — a set of incentives that auditors, academics and consumer-facing tools all try to surface [8] [5] [7].