Fundamentals

Understanding The Power Of Recommendations

In today’s digital marketplace, standing out is not just about visibility; it’s about relevance. Small to medium businesses (SMBs) are constantly seeking effective strategies to connect with customers on a personal level, driving sales and building lasting relationships. AI-powered product recommendations are no longer a futuristic concept reserved for large corporations. They are now accessible and, more importantly, essential tools for SMBs aiming to compete effectively.

These systems analyze customer data Meaning ● Customer Data, in the sphere of SMB growth, automation, and implementation, represents the total collection of information pertaining to a business's customers; it is gathered, structured, and leveraged to gain deeper insights into customer behavior, preferences, and needs to inform strategic business decisions. to suggest products they are most likely to purchase, enhancing the shopping experience and boosting revenue. Think of it as having a highly knowledgeable sales assistant for every online visitor, guiding them towards items that genuinely match their interests and needs.

AI-powered product recommendations act as personalized sales assistants, guiding customers to relevant products and boosting SMB revenue.

Why Recommendations Matter For Small Businesses

For SMBs, every customer interaction is valuable. Recommendations provide a unique opportunity to maximize the value of each visit to your online store. They move beyond generic marketing by offering tailored suggestions that resonate with individual shoppers. This personalization is critical because it directly addresses the common challenge of customer acquisition and retention.

By showing customers that you understand their preferences, you create a more engaging and satisfying shopping experience. This leads to increased customer loyalty Meaning ● Customer loyalty for SMBs is the ongoing commitment of customers to repeatedly choose your business, fostering growth and stability. and repeat purchases, which are vital for sustainable growth. Furthermore, effective recommendations can significantly increase average order value as customers discover and purchase items they might not have found otherwise. In essence, recommendations transform passive browsing into active purchasing.

Essential First Steps Setting Up Recommendations

Getting started with AI-powered recommendations doesn’t require a massive overhaul of your existing systems. The initial steps are surprisingly straightforward and focus on leveraging readily available tools and data. The first crucial step is to understand your customer data. This involves looking at past purchase history, browsing behavior on your website, and any demographic information you collect.

Even basic data can be powerful when used correctly. Next, explore platforms that offer easy-to-integrate recommendation features. Many e-commerce platforms like Shopify and WooCommerce have built-in recommendation engines Meaning ● Recommendation Engines, in the sphere of SMB growth, represent a strategic automation tool leveraging data analysis to predict customer preferences and guide purchasing decisions. or readily available plugins. These options often require minimal technical expertise and can be set up quickly.

Start with simple recommendation types, such as ‘customers who bought this item also bought’ or ‘recommended for you’ based on browsing history. The key is to begin implementing recommendations in a manageable way and gradually refine your approach as you gather more data and insights.

Avoiding Common Pitfalls In Early Implementation

While the benefits of AI recommendations are clear, there are common pitfalls SMBs should avoid when starting out. One frequent mistake is overcomplication. Resist the urge to immediately implement highly complex algorithms. Start simple and focus on getting the basics right.

Another pitfall is neglecting data quality. Recommendations are only as good as the data they are based on. Ensure your product data is accurate, well-categorized, and consistently updated. Poor data quality Meaning ● Data Quality, within the realm of SMB operations, fundamentally addresses the fitness of data for its intended uses in business decision-making, automation initiatives, and successful project implementations. can lead to irrelevant or even misleading recommendations, damaging the customer experience.

Additionally, avoid a ‘set it and forget it’ mentality. Recommendations require ongoing monitoring and adjustment. Track the performance of your recommendations, analyze customer feedback, and be prepared to tweak your strategies to optimize results. Finally, don’t underestimate the importance of user experience. Ensure recommendations are presented in a clear, non-intrusive way that enhances, rather than detracts from, the shopping experience.

Foundational Tools And Platforms For Beginners

For SMBs just beginning their journey with AI-powered recommendations, several user-friendly tools and platforms are available. These options are designed to be accessible, affordable, and require minimal technical expertise. For businesses using Shopify, the platform’s built-in product recommendation engine is a great starting point. It’s easy to activate and offers basic recommendation types.

Similarly, WooCommerce users can explore plugins like ‘Product Recommendations’ or ‘Recommendation Engine,’ which provide straightforward integration and customization options. For businesses using other e-commerce platforms or custom websites, consider platforms like Nosto or Recombee, which offer more advanced features but still cater to SMBs with varying technical capabilities. These platforms often provide drag-and-drop interfaces and pre-built recommendation templates, simplifying the setup process. The selection of the right tool depends on your platform, technical comfort level, and budget. Starting with platform-native features or simple plugins is often the most efficient way for SMBs to gain initial traction and experience with recommendations.

Data Collection Basics For Effective Recommendations

Data is the fuel that powers AI-driven recommendations. For SMBs, understanding what data to collect and how to collect it is fundamental. The most valuable data points for product recommendations include:

• Purchase History ● What products has a customer bought in the past? This is a strong indicator of future interests.
• Browsing Behavior ● Which products and categories has a customer viewed on your website? This reveals current interests and preferences.
• Demographics ● Age, location, gender, and other demographic data can help personalize recommendations, especially when combined with purchase and browsing history.
• Product Interactions ● Clicks, adds to cart, items saved to wishlists ● these actions signal customer interest in specific products.
• Search Queries ● What keywords are customers using to search on your site? This provides direct insight into their needs.

Collecting this data can be done through your e-commerce platform’s built-in analytics, Google Analytics, or specialized customer data platforms (CDPs). Ensure you have proper data collection mechanisms in place from the outset. It’s also crucial to be transparent with customers about data collection practices and comply with privacy regulations like GDPR and CCPA. Building trust through data transparency is as important as collecting the data itself.

Quick Implementation Strategies For Immediate Impact

SMBs often need to see results quickly. Fortunately, several implementation strategies can deliver immediate impact with AI-powered recommendations. Start by focusing on high-traffic areas of your website, such as product pages and the homepage. Implementing ‘related products’ or ‘you might also like’ recommendations on product pages can immediately increase product discovery and average order value.

On the homepage, ‘personalized recommendations’ based on browsing history can engage returning visitors from the moment they land on your site. Email marketing is another area for quick wins. Include personalized product recommendations Meaning ● Personalized Product Recommendations utilize data analysis and machine learning to forecast individual customer preferences, thereby enabling Small and Medium-sized Businesses (SMBs) to offer pertinent product suggestions. in your promotional emails or transactional emails (like order confirmations). This re-engages customers and drives repeat purchases.

Social media can also be leveraged for recommendations. Platforms like Facebook and Instagram allow for product tagging and shoppable posts, where you can highlight recommended products directly within your social content. By focusing on these key touchpoints, SMBs can quickly see the positive effects of AI-powered recommendations on their sales and customer engagement.

Measuring Success And Iterating On Fundamentals

Implementing recommendations is just the first step. Continuously measuring success and iterating on your fundamental strategies is crucial for long-term effectiveness. Key metrics to track include:

1. Click-Through Rate (CTR) on Recommendations ● How often are customers clicking on recommended products? A low CTR might indicate irrelevant recommendations or poor placement.
2. Conversion Rate of Recommended Products ● What percentage of clicks on recommendations result in purchases? This measures the effectiveness of recommendations in driving sales.
3. Average Order Value (AOV) ● Does the inclusion of recommendations increase the average amount customers spend per order? This shows the impact on revenue.
4. Customer Engagement Metrics ● Are customers spending more time on your site and viewing more products after recommendations are implemented? This indicates improved user experience.
5. Feedback and Reviews ● Pay attention to customer feedback Meaning ● Customer Feedback, within the landscape of SMBs, represents the vital information conduit channeling insights, opinions, and reactions from customers pertaining to products, services, or the overall brand experience; it is strategically used to inform and refine business decisions related to growth, automation initiatives, and operational implementations. and reviews related to recommendations. Are they finding them helpful? Are there any complaints about irrelevant or repetitive recommendations?

Regularly analyze these metrics to identify areas for improvement. A/B test different recommendation types, placements, and algorithms to see what works best for your audience. Iterate based on data-driven insights.

For example, if you notice low CTR on ‘related products,’ try refining your product tagging or category structure to improve relevance. The process of continuous measurement and iteration is what transforms basic recommendations into a powerful, revenue-generating engine for your SMB.

Intermediate

Moving Beyond Basic Recommendation Strategies

Once SMBs have established a foundation with basic AI-powered recommendations, the next step is to move beyond simple strategies and explore more sophisticated techniques. This intermediate phase focuses on enhancing personalization, improving recommendation accuracy, and integrating recommendations more deeply into the customer journey. While fundamental recommendations might rely on simple rules like ‘frequently bought together,’ intermediate strategies leverage more nuanced data analysis and algorithms to deliver truly personalized experiences. This shift is crucial for SMBs looking to gain a competitive edge by providing a shopping experience that feels uniquely tailored to each customer.

Intermediate recommendation strategies focus on enhanced personalization and deeper integration into the customer journey for SMBs.

Data Segmentation For Enhanced Personalization

Personalization is key to effective recommendations, and data segmentation Meaning ● Data segmentation, in the context of SMBs, is the process of dividing customer and prospect data into distinct groups based on shared attributes, behaviors, or needs. is the engine that drives it. Instead of treating all customers the same, segmentation involves dividing your customer base into distinct groups based on shared characteristics and behaviors. This allows you to tailor recommendations to the specific needs and preferences of each segment. Common segmentation criteria include:

• Demographic Segmentation ● Grouping customers by age, gender, location, income, etc. This can be useful for products with demographic-specific appeal.
• Behavioral Segmentation ● Grouping customers based on their actions on your website, such as browsing history, purchase frequency, items added to cart, and website interactions. This is highly effective for personalized recommendations.
• Value-Based Segmentation ● Segmenting customers based on their purchase value (e.g., high-value customers, repeat customers, new customers). This allows you to offer different types of recommendations and incentives to different customer groups.
• Psychographic Segmentation ● Grouping customers based on their lifestyle, values, interests, and opinions. While more challenging to collect, this data can lead to highly relevant and engaging recommendations if available.

By segmenting your audience, you can move from generic ‘top sellers’ recommendations to more targeted suggestions like ‘recommendations for fashion-conscious millennials’ or ‘products you might like based on your past outdoor gear purchases.’ This level of personalization significantly increases the relevance and effectiveness of your recommendations.

Advanced Recommendation Algorithms For Accuracy

While simple rule-based recommendations are a good starting point, advanced algorithms are needed to achieve higher levels of accuracy and personalization. Several algorithms are particularly effective for intermediate-level implementations:

1. Collaborative Filtering ● This algorithm recommends products based on the preferences of similar users. It identifies customers who have similar purchase histories or browsing patterns and recommends products that those similar users have liked or purchased. This is effective even with limited data on individual users.
2. Content-Based Filtering ● This algorithm recommends products that are similar to those a user has liked or purchased in the past. It analyzes product attributes (e.g., category, tags, descriptions) to find items with similar characteristics. This works well when you have rich product data.
3. Hybrid Recommendation Systems ● These systems combine collaborative and content-based filtering to leverage the strengths of both approaches. They can provide more robust and accurate recommendations, especially when dealing with diverse customer behavior and product types.
4. Association Rule Mining ● This technique identifies relationships between products that are frequently purchased together. It can be used to generate ‘frequently bought together’ or ‘customers who bought this also bought’ recommendations. Algorithms like Apriori and FP-Growth are commonly used for this purpose.

Implementing these algorithms often requires using recommendation platforms or plugins that offer these capabilities. Many intermediate-level platforms provide options to choose and customize different algorithms based on your specific needs and data.

Integrating Recommendations With Marketing Automation

To maximize the impact of AI-powered recommendations, integrate them with your marketing automation Meaning ● Marketing Automation for SMBs: Strategically automating marketing tasks to enhance efficiency, personalize customer experiences, and drive sustainable business growth. efforts. This creates a seamless and personalized customer experience Meaning ● Customer Experience for SMBs: Holistic, subjective customer perception across all interactions, driving loyalty and growth. across different touchpoints. Key integration points include:

• Email Marketing ● Personalize email campaigns with product recommendations based on customer segments, past purchases, and browsing behavior. Include recommendations in promotional emails, abandoned cart emails, and post-purchase follow-ups.
• Website Personalization ● Dynamically display personalized product recommendations on different website pages, such as the homepage, product pages, category pages, and cart page. Tailor recommendations based on real-time browsing behavior and customer profiles.
• Retargeting Ads ● Use product recommendations in retargeting ads to re-engage customers who have previously shown interest in specific products or categories. Show them personalized ads featuring items they have viewed or added to their cart.
• Chatbots and Customer Service ● Integrate recommendations into chatbot interactions and customer service communications. When a customer asks for product advice or information, the chatbot can provide personalized recommendations Meaning ● Personalized Recommendations, within the realm of SMB growth, constitute a strategy employing data analysis to predict and offer tailored product or service suggestions to individual customers. based on their query and past interactions.

By integrating recommendations into your marketing automation workflows, you create a consistent and personalized experience that guides customers through the purchase funnel and encourages conversions. This also allows for automated and scalable personalization efforts.

Measuring Performance Beyond Basic Metrics

While CTR and conversion rate are important metrics, intermediate-level performance measurement requires looking beyond these basic indicators. To gain a deeper understanding of the impact of your recommendations, track the following metrics:

1. Revenue Per Recommendation ● Calculate the average revenue generated by each recommended product. This provides a direct measure of the financial impact of your recommendations.
2. Incremental Revenue Lift ● Measure the increase in revenue specifically attributed to recommendations, compared to a control group or baseline. This isolates the impact of your recommendation efforts.
3. Customer Lifetime Value (CLTV) Improvement ● Analyze whether customers who interact with recommendations have a higher CLTV over time. This assesses the long-term impact on customer loyalty and repeat purchases.
4. Recommendation Coverage Rate ● Measure the percentage of website visitors or customers who are exposed to recommendations. Increasing coverage ensures that more customers benefit from personalization.
5. Fall-Back Rate ● Monitor how often your recommendation system fails to provide relevant recommendations and resorts to generic or irrelevant suggestions. A high fall-back rate indicates areas for algorithm improvement.

By tracking these advanced metrics, you gain a more comprehensive view of the effectiveness of your intermediate recommendation strategies and identify specific areas for optimization and further refinement.

A/B Testing Advanced Recommendation Strategies

A/B testing is essential for optimizing intermediate recommendation strategies. Instead of simply implementing changes and hoping for the best, A/B testing Meaning ● A/B testing for SMBs: strategic experimentation to learn, adapt, and grow, not just optimize metrics. allows you to compare different approaches and make data-driven decisions. Effective A/B testing strategies for recommendations include:

• Algorithm Comparison ● Test different recommendation algorithms (e.g., collaborative filtering vs. content-based filtering) to see which performs better for your audience and product types.
• Placement Optimization ● Experiment with different placements of recommendations on your website (e.g., product page, homepage, cart page) to determine the most effective locations for engagement and conversions.
• Personalization Level Testing ● Test different levels of personalization (e.g., basic segmentation vs. advanced behavioral segmentation) to find the optimal balance between personalization and data complexity.
• Recommendation Type Variation ● A/B test different types of recommendations (e.g., ‘related products,’ ‘frequently bought together,’ ‘personalized recommendations’) to identify which types resonate most with your customers.
• User Interface (UI) and Design Testing ● Experiment with different UI elements and design layouts for your recommendation widgets to improve visibility, click-through rates, and user experience.

When conducting A/B tests, ensure you have clear hypotheses, well-defined control and variation groups, and statistically significant sample sizes. Analyze the results carefully and use the insights to iterate and refine your recommendation strategies continuously.

Case Study ● Online Boutique Personalization Boost

Consider an online boutique specializing in handcrafted jewelry and accessories. Initially, they implemented basic ‘related products’ recommendations on product pages. To move to the intermediate level, they decided to focus on enhanced personalization using data segmentation and advanced algorithms. They segmented their customer base into three groups ● ‘Fashion-Forward Trendsetters,’ ‘Classic Elegance Seekers,’ and ‘Budget-Conscious Shoppers,’ based on purchase history and browsing behavior.

For ‘Fashion-Forward Trendsetters,’ they implemented recommendations based on collaborative filtering, suggesting items popular among customers with similar style preferences. For ‘Classic Elegance Seekers,’ they used content-based filtering, recommending jewelry with similar materials and designs to their past purchases. For ‘Budget-Conscious Shoppers,’ they focused on ‘frequently bought together’ recommendations, highlighting value sets and discounted items. They also integrated recommendations into their email marketing, sending personalized newsletters to each segment featuring curated product selections.

The results were significant. Click-through rates on recommendations increased by 45%, conversion rates for recommended products rose by 30%, and average order value saw a 20% uplift. The boutique demonstrated how intermediate strategies, focusing on data segmentation and algorithm sophistication, can dramatically improve the performance of AI-powered recommendations for SMBs.

Advanced

Cutting Edge Recommendation System Approaches

For SMBs aiming to achieve a significant competitive advantage, advanced AI-powered recommendation systems represent the cutting edge. Moving beyond intermediate strategies involves exploring sophisticated techniques that leverage the latest advancements in artificial intelligence and machine learning. These advanced approaches focus on hyper-personalization, real-time adaptability, and predictive capabilities.

They are designed to create truly dynamic and anticipatory shopping experiences that not only meet current customer needs but also predict and fulfill future desires. This level of sophistication requires a deeper understanding of AI principles and often involves leveraging specialized tools and platforms.

Advanced recommendation systems utilize hyper-personalization, real-time adaptability, and predictive capabilities for a dynamic customer experience.

AI Powered Hyper Personalization At Scale

Hyper-personalization takes individualization to the next level. It moves beyond broad segmentation to create a truly one-to-one experience for each customer. This involves leveraging AI to analyze a vast array of data points to understand individual preferences, contexts, and even momentary needs. Key aspects of hyper-personalization include:

• Contextual Recommendations ● Recommendations that adapt to the current context of the customer’s interaction. This includes factors like time of day, day of the week, location, device being used, and even current weather conditions. For example, recommending weather-appropriate clothing or time-sensitive promotions.
• Behavioral Micro-Segmentation ● Moving beyond static segments to create dynamic, real-time micro-segments based on immediate browsing behavior and interactions. This allows for highly responsive and relevant recommendations based on the customer’s current session.
• Personalized Content Recommendations ● Extending recommendations beyond products to include personalized content like blog posts, articles, videos, and user-generated content. This enhances engagement and builds a deeper connection with the customer.
• Predictive Recommendations ● Using AI to predict future customer needs and proactively recommend products before the customer even realizes they need them. This involves analyzing historical data and patterns to anticipate future purchases.

Achieving hyper-personalization at scale requires robust AI infrastructure and sophisticated algorithms capable of processing and analyzing vast amounts of real-time data to deliver truly individualized experiences.

Predictive Analytics And Anticipatory Recommendations

Predictive analytics plays a crucial role in advanced recommendation systems. By leveraging machine learning Meaning ● Machine Learning (ML), in the context of Small and Medium-sized Businesses (SMBs), represents a suite of algorithms that enable computer systems to learn from data without explicit programming, driving automation and enhancing decision-making. models, SMBs can move from reactive recommendations to anticipatory suggestions. This involves:

1. Demand Forecasting ● Using historical sales data, seasonal trends, and external factors to predict future product demand. This allows for proactive inventory management and recommendation of products likely to be in high demand.
2. Next Best Action Recommendations ● Predicting the next best action Meaning ● Next Best Action, in the realm of SMB growth, automation, and implementation, represents the optimal, data-driven recommendation for the next step a business should take to achieve its strategic objectives. a customer is likely to take and recommending products or content that aligns with that predicted action. This could involve anticipating a customer’s move to purchase, subscribe, or engage with specific content.
3. Churn Prediction and Prevention ● Identifying customers who are at risk of churn and proactively recommending products or offers to re-engage them and prevent them from leaving. This leverages customer behavior patterns to predict churn risk.
4. Personalized Promotions and Offers ● Predicting the types of promotions and offers that are most likely to resonate with individual customers and delivering personalized incentives to drive conversions and loyalty.

Predictive analytics empowers SMBs to be proactive in their recommendation strategies, anticipating customer needs and delivering timely and relevant suggestions that drive sales and build stronger customer relationships.

Real Time Recommendation Engines For Dynamic Experiences

In today’s fast-paced digital environment, real-time recommendations are essential for creating truly dynamic and engaging customer experiences. Real-time recommendation engines process and analyze data instantaneously to deliver recommendations that are relevant to the customer’s current interaction. Key features of real-time recommendation engines include:

• In-Session Personalization ● Recommendations that adapt in real-time to the customer’s browsing behavior within a single website session. As a customer navigates your site, the recommendations dynamically update based on their clicks, views, and searches.
• Trigger-Based Recommendations ● Recommendations that are triggered by specific customer actions or events, such as adding an item to cart, viewing a specific category, or spending a certain amount of time on a page. These triggers signal specific interests and needs.
• Dynamic Content Adjustment ● Beyond product recommendations, real-time engines can dynamically adjust website content, layout, and messaging based on individual customer profiles and real-time behavior. This creates a fully personalized website experience.
• Fast Response Time ● Real-time engines are designed for speed, delivering recommendations with minimal latency to ensure a seamless and responsive user experience. This is crucial for maintaining customer engagement.

Implementing real-time recommendation engines requires robust infrastructure and algorithms capable of handling high volumes of data and delivering recommendations in milliseconds. However, the payoff is a significantly more engaging and personalized customer experience.

Advanced Automation For Recommendation Workflows

To effectively manage and scale advanced recommendation systems, automation is paramount. Advanced automation techniques streamline recommendation workflows and reduce the need for manual intervention. Key areas for automation include:

1. Automated Algorithm Selection and Optimization ● AI-powered systems that automatically select the best recommendation algorithms for different customer segments and product categories, and continuously optimize algorithm parameters based on performance data.
2. Dynamic Segmentation and Profile Updates ● Automated systems that dynamically segment customers in real-time and continuously update customer profiles based on new data and interactions. This ensures that personalization is always based on the most current information.
3. Automated A/B Testing and Experimentation ● Automated A/B testing platforms that continuously run experiments on different recommendation strategies, algorithms, and placements, and automatically implement the winning variations.
4. Performance Monitoring and Alerting ● Automated monitoring systems that track key performance metrics for recommendations and trigger alerts when performance drops below predefined thresholds, allowing for proactive intervention.

Automation is essential for SMBs to effectively leverage advanced recommendation systems without overwhelming their resources. It enables scalability, efficiency, and continuous optimization of recommendation efforts.

Case Study ● Subscription Box Service Predictive Recommendations

Consider a subscription box service that delivers curated boxes of gourmet food items monthly. To move to an advanced level of recommendation sophistication, they implemented a predictive recommendation system. They leveraged machine learning to analyze customer subscription history, feedback, dietary preferences, and even external data like seasonal food trends and local events. Using this data, they built a predictive model that could anticipate the types of food items each subscriber would enjoy in their next box.

Before each month’s box curation, the system generated personalized recommendations for each subscriber, suggesting specific items to include based on their predicted preferences. The curators then used these recommendations to assemble boxes that were highly tailored to individual tastes. They also integrated real-time recommendations into their website and app, suggesting add-on items based on the predicted contents of the upcoming box. The results were remarkable.

Customer satisfaction scores increased by 60%, subscriber retention rates improved by 35%, and average order value for add-on items saw a 50% jump. This case study demonstrates how advanced predictive recommendations can transform a subscription-based SMB, creating highly personalized experiences that drive customer loyalty and revenue growth.

Future Trends In AI Driven Recommendations

The field of AI-powered recommendations is constantly evolving. SMBs looking to stay ahead should be aware of emerging trends that will shape the future of personalization:

• Explainable AI (XAI) in Recommendations ● Increasing focus on making recommendation algorithms more transparent and explainable. Customers are increasingly interested in understanding why certain products are recommended to them. XAI will enhance trust and transparency.
• Voice-Based and Conversational Recommendations ● With the rise of voice assistants and conversational commerce, recommendations will increasingly be delivered through voice interfaces and natural language interactions. This will require adapting recommendation systems to understand and respond to voice queries.
• Ethical AI and Responsible Recommendations ● Growing awareness of ethical considerations in AI, including bias in algorithms, data privacy, and responsible use of personalization. Future recommendation systems will need to be designed with ethical principles in mind.
• Integration with Metaverse and Immersive Experiences ● As the metaverse and immersive technologies evolve, recommendations will extend into these virtual environments, creating personalized shopping experiences in virtual and augmented reality.

Staying informed about these future trends will enable SMBs to proactively adapt their recommendation strategies and maintain a competitive edge in the evolving landscape of AI-powered personalization.

References

• Aggarwal, Charu C. Recommender Systems ● The Textbook. Springer, 2016.
• Jannach, Dietmar, et al. Recommender Systems ● An Introduction. Cambridge University Press, 2010.
• Ricci, Francesco, et al., editors. Recommender Systems Handbook. Springer, 2011.