“I aspire for societies to be governed by people again, not algorithms.”
— Juan Carlos Blanco, journalist and author of The Tyranny of Screen Nations
In his recent interview with El País, Juan Carlos Blanco warns about the growing influence of algorithms in our lives: from the news we read to the products we buy online, “automated systems prioritize engagement and economic profit, often at the expense of truth and democratic coexistence.” Although Blanco analyzes this phenomenon from a social and political perspective, his reflections have direct implications for any digital business, especially in the field of e-commerce, web design, and online marketing.
Next, we explore the dangers of leaving algorithms unchecked in your online store, how to identify the main risks, and what practices any company can implement to regain control of the user experience and brand reputation.
1. The rise of algorithms: help or digital cage?
From recommendation engines to programmatic advertising systems, algorithms are designed to maximize metrics (clicks, time on page, conversions). Ultimately, they pursue a commercial goal: to increase revenue. However, this iterative approach generates:
- Content bubbles: users only see what matches their previous behavior patterns, limiting exposure to new or niche products.
- Amplification of errors: if an algorithm mistakenly identifies content or a product as valuable, it will repeat and spread it, even if it doesn’t meet expectations.
- Emotional disconnection: extreme personalization can become invasive or mechanical, removing the human touch that builds trust.
Blanco points out that “algorithms end up building parallel realities” and influence individual and collective decisions. In e-commerce, this translates into shopping carts with suggestions that don’t really meet needs, insistent pop-ups that frustrate users, and a dehumanized brand experience.
2. Specific risks for your online store
2.1 Loss of diversity in the catalog
A good recommendation algorithm boosts short-term sales, but if it always shows the same “top-selling” products, it misses the opportunity to promote new or seasonal items. E-commerce sites that rely solely on these recommendations risk stagnating in trends and losing freshness.
2.2 Incorrect or misleading content
Automatic product description generation systems can hallucinate (invent non-existent features or guarantees). This phenomenon, called “hallucination,” results in product listings with erroneous data, leading to negative reviews and complaints.
2.3 Advertising saturation
Programmatic advertising adjusts offers in real-time. However, a poor setup can cause your ad to follow the user for days across different websites, creating a sense of harassment instead of a gentle reminder. This damages brand image and leads to “banner blindness,” where the user directly ignores your messages.
2.4 Technological dependency
If your e-commerce platform integrates third-party recommenders or chatbots without validation, you are exposed to changes in their policies, price increases, or service failures. Your business ceases to own its user experience.
3. Regaining control: fundamental principles
A. Active human oversight
No automatic system should operate without periodic reviews. Schedule quarterly audits of:
- Recommendations: What are the criteria?
- Generated descriptions: Do they correspond to real data?
- Advertising campaigns: Are they not intrusive?
These audits allow for corrections before causing reputational damage.
B. Transparency with the customer
Inform your users that you use artificial intelligence and explain how they can benefit from it. A simple message like “Personalized recommendations based on your interests” builds trust and helps them understand the system’s logic.
C. Mix of algorithms and human creativity
To launch new products or collections, combine sales data with your team’s intuition: make manual selections of featured products, edit generated descriptions to give them a unique tone, and create original multimedia content (explanatory videos, interactive guides).
D. Diversification of channels
Don’t concentrate all interaction on a single algorithmic channel. Complement automatic recommendations with:
- Segmented newsletters by human profiles.
- Social media with curated content.
- Corporate blog where the team shares news, interviews, and tutorials.
This hybrid strategy reinforces your brand’s personality.
4. Good technological practices: concrete solutions
- Version control of algorithms
Keep a history of your recommendation engine’s configurations. If a change causes a drop in sales or complaints, you can quickly revert it. - Training with clean data
Filter your databases to remove duplicate records, anomalous transactions, or discontinued products before feeding the AI. - Regular A/B testing
Compare the 100% algorithmic experience with hybrid versions (mixing human selections) and measure conversion metrics, session time, and customer satisfaction. - Alerts for sensitive content
Implement systems that detect “hallucinations” in descriptions: words like “guaranteed,” “certified,” or “exclusive,” if not backed by verified data, should be manually reviewed.
5. Human value as a competitive advantage
While many stores choose to automate every process, the human factor remains the differentiating element. Micro-influencers or brand ambassadors (employees, founders, or loyal customers) generate authentic content that algorithms cannot replicate. These testimonials, stories, and close reviews connect with the buyer on a deep level.
Additionally, having a real customer service team, capable of empathizing and resolving complex doubts, strengthens reputation and fosters “word of mouth” recommendations that machines cannot achieve.
6. How EHERO Consulting supports your transformation
At EHERO Consulting, we have been helping digital businesses balance technological innovation with human experience for years.
Our goal is not to demonize algorithms but to place them in their rightful place: tools at the service of human decisions, not uncontrolled shortcuts. In this way, you will protect your brand, retain your customers, and maintain the ability to surprise them with unique products and experiences.
7. Final reflection: governing with responsibility
Juan Carlos Blanco’s warning challenges us all: algorithms are only as good as the human guidance they receive. In e-commerce, AI can multiply sales and efficiency, but without an ethical and supervised framework, it can lead to monotonous experiences, misinformation, and loss of trust.
Regaining control means:
- Putting the customer at the center, not the metric.
- Demanding transparency from the systems we use.
- Investing in the creativity and judgment of people.
Only then will we keep alive the promise of digital commerce that combines the best of technology with the best of humanity.
Don’t forget to follow EHERO Consulting on our social networks to combat misinformation with reliable sources. Follow us on Instagram, X, Facebook, LinkedIn, YouTube and TikTok. We look forward to seeing you!