Bicycle.ai had an amazing time at eTail 2026, meeting with dozens of companies across retail, ecommerce, ocean cruises, and specialty verticals. Our conversations surfaced a clear pattern: revenue teams have plenty of dashboards, reports, and data, but it still takes too long to move from a metric change to the right business action.
When a KPI drops, inventory disappears, ad spend underperforms, or checkout conversion slips, teams often have to pull signals from disconnected systems before they can understand what happened and decide what to do next. That gap – the one between signal, root cause, and action – is where revenue gets lost.
Bicycle’s conversations at eTail showed strong demand for AI analytics that can go beyond static dashboards and help teams detect issues, explain what’s driving them, and recommend the next step. Here are the five pain points prospects raised most often, the use cases that resonated most strongly, and the questions buyers asked as they evaluated what AI-powered revenue intelligence could look like inside their organizations.
1. Manual reporting is slowing down decision-making
The most common pain point was manual reporting and delayed insights, For many retail and ecommerce teams, reporting is still too labor-intensive. Analysts pull data from multiple systems, business teams wait for recurring updates, and executives get the story after the best window for action has already passed.
This came through clearly in conversations about automated business review reporting. Several prospects were interested in replacing manually assembled monthly reports with AI-generated narratives that explain what changed, why it matters, and where the business should focus next.
The underlying need is faster decision-making: teams need to spend less time assembling the report and more time acting on what the data is telling them. Doing so is crucial for lean teams. When analysts are tied up preparing reports, they have less capacity for deeper investigation, strategic analysis, and revenue recovery.
Use case that resonated: automated business review reports that turn performance data into AI-generated data stories for business users.
2. Data fragmentation is making root cause analysis harder
Several raised data fragmentation across systems as a major challenge.
Retail and ecommerce organizations are running increasingly complex stacks. Prospects mentioned tools and systems including Snowflake, BigQuery, Google Analytics 4, Tableau, Looker, Shopify Analytics, SAP, Access, legacy ERPs, and internal systems.
The result is familiar: every team has part of the picture, and no single team has the full story.
Marketing may see ad performance. Ecommerce may see conversion drops. Merchandising may see product availability. Data teams may see warehouse metrics. Engineering may see technical logs. But when revenue moves unexpectedly, the business still has to connect those signals manually.
That’s why root cause analysis was one of the strongest use cases from the event. Prospects wanted help explaining why performance changed, not another view showing that it had changed. In practice, that means connecting a KPI drop to the drivers behind it: inventory, checkout performance, product availability, traffic quality, campaign spend, pricing, or another operational factor.
Use case that resonated: root cause analysis that explains why performance changed, not just what changed.
3. Teams lack real-time anomaly detection
We heard a lot about the lack of real-time anomaly detection as a key gap.
This may be the clearest sign that the market is ready for a different analytics experience. Prospects weren’t asking for another dashboard to monitor, but rather for proactive alerts when something important in the business had shifted.
This didn’t surprise our team, because many revenue leaks don’t look like major outages. They show up as small but meaningful changes: a conversion drop in one device segment, a payment issue in one region, a product availability problem in one category, or a checkout issue affecting a narrow slice of traffic.
Traditional reporting often catches these issues after the damage is done. Real-time anomaly detection gives teams a chance to respond while the issue is still unfolding.
Use case that resonated: real-time alerts for KPI drops, stockouts, checkout failures, and other revenue-impacting anomalies.
4. Inventory issues are wasting ad spend and hurting conversion
We also heard that inventory and ad spend waste was a major challenge for teams, a pain point that came through especially strongly in retail and ecommerce conversations. Prospects were concerned about discovering too late that campaigns were driving traffic to products that were unavailable, understocked, or not purchasable in a customer’s location.
Problems related to inventory and out-of-stock monitoring resonated because they’re tied directly to revenue protection. Teams need proactive alerts before spend is wasted, not a post-campaign explanation of why performance underwhelmed.
For retailers, this is where AI analytics can become more operational. The value isn’t simply showing that conversion dropped, but rather identifying that the drop is connected to product availability. Ai can then prompt the team to suppress products, adjust campaigns, throttle spend, or investigate replenishment.
Use case that resonated: proactive inventory and out-of-stock monitoring to prevent wasted ad spend and missed revenue.
5. Build-vs-buy pressure is colliding with limited analyst resources
Build versus buy came up quite a bit. Many organizations recognize the need for better AI-powered analytics, but they’re still evaluating whether to build internally or buy a specialized solution. In fact, several prospects asked directly why they shouldn’t build this themselves.
Our response? The strategic issue is time-to-value. Internal teams may be capable of building pieces of the solution, but they’re also managing competing priorities, data integration work, model governance questions, and ongoing maintenance. The cost isn’t only the build, but rather the time the business spends waiting for something usable.
That’s especially important for lean analytics teams. When a business unit is already overloaded with reporting, dashboard maintenance, ad hoc requests, and cross-functional investigations, building a full AI analytics layer can delay the outcomes the business needs most.
Use case that resonated: natural language chat and self-serve analytics that allow business users to ask questions without waiting on data scientists or analysts.
The five questions prospects asked most often
The frequently asked questions from eTail were just as revealing as the pain points. They show what buyers need to believe before moving forward with AI analytics.
1. How is data security and compliance handled?
Security came up in nearly every meeting, including questions around SOC 2 Type 2, cloud versus on-premise deployment, and how data is grounded. We told them that Bicycle has public enterprise organizations and we treat security as a first-class citizen.
2. How long does integration and onboarding take?
Prospects wanted to understand time-to-value, especially when multiple data sources were involved. We explained that onboarding can be done quickly and that Bicycle helps every step of the way. We’ve done this many times before, and we can do it quickly and efficiently.
3. Why not build this internally?
Build versus buy was a recurring theme, with buyers weighing internal development timelines against speed, specialization, and resource constraints. We heard that prospects had done their best with Claude, but quickly realized they needed a third-party solution to get to the finish line.
4. How is pricing structured?
Prospects asked about overall cost. We explained that we weren’t replacing existing tools but rather a layer on top of them, and explained how our focus on stopping revenue leaks and revenue recovery led to quick ROI and payback.
5. Can Bicycle connect to our specific data sources?
Connectivity questions covered systems including Access, Shopify, Snowflake, GA4, and observability tools. We explained that there’s almost no major data source or tool that Bicycle can’t connect to in order to present the entire picture – that’s a major part of our value proposition.
Together, these questions point to a market that’s evaluating whether AI analytics can be trusted, connected, implemented quickly, priced clearly, and justified against internal alternatives.
What eTail revealed about the next phase of AI analytics
The strongest takeaway from eTail is that retail and ecommerce teams are ready for analytics to become more operational.
They require a faster path from business signal to root cause to action. That means earlier detection, clearer explanations, proactive alerts, automated reporting, connected inventory and campaign signals, and self-serve answers for business users.
For Bicycle.ai, eTail 2026 validated a clear market need. Prospects are actively looking for ways to detect revenue leaks earlier, investigate them faster, and operationalize the next best action across teams. In doing so, they’ll protect revenue in the moments when action matters most.
