Standard marketing dashboards were not built for companies where a single safety event can move demand by 20 percent overnight. Winston Francois builds AV marketing analytics that account for the variables that actually drive your numbers.
The Problem
Ride Data and Marketing Data Live in Separate Systems
AV companies generate rich operational data – trips completed, routes served, dwell times, service reliability metrics – that is directly relevant to marketing performance but almost never connected to marketing analytics systems. When ride data and marketing data are siloed, you cannot see how operational performance affects acquisition cost, retention rates, or brand health. Marketing decisions get made without the most important context for interpreting them.
Safety Events Are Not in the Attribution Model
A disengagement event, a fender bender, or a permit investigation does not just affect operations – it affects how your paid and organic marketing performs over the following days and weeks. App installs drop, CPMs spike, and organic search interest shifts in ways that cannot be explained by looking at your marketing stack alone. Without a structured way to incorporate safety event data into your analytics, your team will repeatedly misread performance and draw the wrong conclusions.
PR Coverage and Earned Media Are Invisible in the Funnel
AV companies often drive significant awareness and demand through press coverage – a feature in a major publication, a TV segment, or a viral social media moment. Most marketing analytics setups cannot attribute any of this to business outcomes. When a wave of coverage drives a spike in app downloads or fleet inquiry volume, it looks like organic demand in the dashboard. You cannot optimize what you cannot see, and you cannot make the case for PR investment without attribution data.
Regulatory Approval Events Create Demand Spikes That Break Forecasts
When a permit is approved, a new service area opens, or a regulatory milestone is reached, demand typically spikes in ways that have nothing to do with marketing spend. Standard marketing analytics frameworks treat these as external noise rather than inputs. The result is that forecasts are consistently wrong around regulatory events, and the team cannot distinguish between demand driven by their marketing programs and demand driven by external factors they do not control.
How We Help
Winston Francois starts AV marketing analytics engagements by auditing the full data landscape: what operational data exists, where it lives, what marketing data you are already collecting, and what PR and earned media tracking is in place. Most AV companies have more relevant data than they realize – the gap is in the integration and interpretation layer.
We design a unified marketing analytics architecture that connects ride data, safety event logs, PR coverage tracking, and standard marketing performance data into a single measurement environment. This does not require replacing your existing marketing stack. It means defining the integration points, the shared identifiers, and the event taxonomy that lets data from different systems be analyzed together.
Channel attribution is built around the reality of AV marketing spend patterns. Paid search, app store optimization, out-of-home in service areas, and fleet partner co-marketing all require different attribution approaches. We design a multi-touch attribution model that accounts for the role of PR coverage and earned media as demand drivers alongside paid channels.
Safety event modeling is one of the most distinctive elements of AV marketing analytics. We work with your safety and operations teams to define the event taxonomy that is relevant to marketing performance and build a structured process for incorporating safety events into your analytics interpretation. The goal is not to create a public-facing safety dashboard – it is to give your marketing team the context they need to correctly interpret their own performance data.
Forecasting models are designed to incorporate regulatory calendar events – permit applications, scheduled review hearings, anticipated approval announcements – alongside standard seasonality and trend factors. This dramatically improves forecast accuracy around the events that most consistently break conventional marketing forecasts.
Reporting is designed for the AV executive team, not just the marketing team. We build dashboards that connect marketing performance to operational metrics and business outcomes in a way that is legible to your CEO, CFO, and board.
What we deliver
- Marketing analytics architecture: data source map, integration design, and event taxonomy
- Unified data environment setup with ride data, safety events, PR coverage, and marketing data connected
- Multi-touch attribution model calibrated for AV marketing channel mix
- Safety event impact model for interpreting marketing performance during and after incidents
- Regulatory event forecasting adjustments integrated into the standard forecasting process
- Executive reporting dashboard with marketing-to-operations connection and quarterly business review templates
In the AV category, marketing performance cannot be correctly interpreted without operational and safety context. Teams that analyze their marketing data in isolation will consistently draw wrong conclusions and make poor spend decisions.
Our Methodology
The first 30 days of a Winston Francois AV analytics engagement are diagnostic and architectural. We audit your current marketing stack, your operational data infrastructure, and any existing analytics or reporting you have. We interview your marketing, operations, and safety teams to understand the data they generate and the decisions they need analytics to support. The output is a detailed analytics architecture document that specifies what gets built, how it connects, and why.
Days 31 through 60 focus on implementation. We work with your engineering and data teams to establish the integrations between operational and marketing data systems. We build the attribution model, configure the safety event tracking process, and set up the forecasting model with regulatory calendar inputs. We run the first wave of analysis using historical data to validate that the integrated model is producing interpretations that make sense to people with operational context.
The final 30 days establish reporting cadence and transfer capability. We build the executive dashboard, train the analysts who will own ongoing reporting, and document the interpretation guidelines that help your team apply the right context when reading performance data. We run the first full monthly and quarterly reporting cycle with your team so they are comfortable owning the process before the engagement closes.
How We Work
Winston Francois analytics engagements are scoped as 90-day projects with fixed deliverables at each 30-day milestone. We do not bill by the hour. The engagement fee covers the full sprint from audit through implementation and handoff.
The first 30 days produce the analytics architecture document and the data audit findings. This is a concrete deliverable that your engineering team can act on immediately, regardless of whether you continue the engagement.
Days 31 through 60 produce the integrated analytics environment and the first set of calibrated models. We present findings from the first historical analysis run at the 60-day review, which gives you a concrete demonstration of what the integrated analytics approach reveals that your current setup does not show.
The final 30 days deliver the executive reporting dashboard, the trained internal team, and the documented processes. After the engagement closes, your team owns every element of the system we built. We remain available for questions for 30 days post-engagement at no additional charge.
- A unified marketing analytics environment that connects ride data, safety events, PR coverage, and paid marketing performance
- Attribution model calibrated for the AV channel mix including earned media and regulatory event demand
- Forecasting capability that accounts for regulatory calendar events and reduces model error around permit milestones
- Executive reporting that connects marketing performance to operational metrics in a boardroom-ready format
If your autonomous vehicles company needs marketing analytics leadership, we should talk.
Expand your marketing team output with our experts
Let us take a custom approach to your growth goals by assembling and leading the best-in-class marketing team to support your next stage.
Frequently asked questions
How do you connect operational ride data to marketing analytics without creating data privacy issues?
We design the integration at the aggregate level, not at the individual rider level. The connection between ride data and marketing analytics is built around market-level and segment-level aggregations – trip volume by zone, service reliability by time of day, rider satisfaction scores by cohort – rather than individual trip records.
What does safety event tracking in a marketing analytics context actually look like?
We build a structured log of safety events that are relevant to marketing performance interpretation. This includes disengagement events above a defined threshold, incidents with media coverage, permit suspensions or regulatory actions, and proactive safety communications from your company. Each event is tagged with a severity level, a media coverage score, and a geographic scope. When analysts review performance data, the safety event log provides the context needed to distinguish between marketing-driven performance changes and externally driven ones.
Can you attribute demand from press coverage and earned media to specific business outcomes?
Attribution for earned media is never as clean as attribution for paid search, but it is far better than ignoring it entirely. We build an earned media tracking system that captures coverage volume, publication authority, sentiment, and estimated reach. We correlate coverage events with demand signals – app installs, fleet inquiry volume, direct traffic – using time-series analysis and holdout testing where possible. The result is not a perfect attribution number but a calibrated range that lets you make investment decisions about PR and communications with evidence behind them.
How do you handle attribution when a competitor has a safety incident that affects our demand?
This is one of the scenarios where AV marketing analytics diverges most sharply from conventional approaches. We build a competitive event tracking layer that logs significant incidents involving your named competitors. When a competitor event occurs, we flag it in the analytics environment and apply an interpretation note to all performance data in the surrounding period. This prevents your team from attributing competitor-driven demand changes to your own marketing programs and helps you understand your true baseline performance.
What is the typical investment for a Winston Francois AV marketing analytics engagement?
Full 90-day analytics engagements for AV companies typically range from $25K-$50K depending on the complexity of your data environment, the number of data systems being integrated, and the sophistication of the attribution and forecasting models required. Data engineering work to build integrations may be additional if your team does not have the capacity to implement based on our specifications. We provide a detailed scope and fee estimate after an initial audit of your current analytics environment.
Is this analytics work something our data team can maintain after the engagement ends?
Yes, and that is explicitly the goal. Every integration, model, and dashboard we build is documented in enough detail that your data team can maintain, modify, and extend it without our involvement. We train your analysts during the final 30 days of the engagement and conduct a structured knowledge transfer session before the engagement closes. If your team has specific capability gaps that would make maintenance difficult, we identify those early and either adjust the technical approach or recommend targeted hiring before we build systems that require skills you do not yet have.
