# Juliana J.

Juliana Jackson is a data and marketing professional who writes and teaches about analytics, measurement, and turning data into business decisions.

- Profile: https://b2binfluencer.marketing/creators/juliana-j/
- LinkedIn: https://www.linkedin.com/in/juliana-jackson
- LinkedIn followers: 14,908
- Topics: artificial-intelligence, ai-in-marketing, attribution, first-party-data
- Based on 18 dated posts, latest 2026-10-09

## Summary

Juliana Jackson is most useful for teaching teams to connect analytics and measurement to business decisions.

- Best for: Analytics, measurement or data storytelling workshops for mixed analyst, marketer and business stakeholder teams.
- Not for: Hands-on SQL, GA4 configuration, tool certification or short-form video consumer launches.
- Check first: Confirm audience includes training buyers or budget holders and ask for recent audience data.

### Takeaways

- Numbers from different platforms need not match; each system measures different things.
- Defining a new customer is a business decision, not a tool default.
- Analytics and AI work should be judged by whether a decision changed.

## Main ideas

- **Numbers that disagree are usually not wrong; they are measuring different things**: Jackson argues that the familiar argument over whose number is right targets the wrong problem. Google Analytics, Google Ads and an A/B testing tool measure different things, collect data at different moments, treat identity differently and apply their own unseen attribution and deduplication rules, so their figures were never going to agree. In Jackson’s view, chasing parity outsources confidence to the data instead of building judgement about what the numbers mean. The people most insistent on matching numbers are often the ones who trust the data least. Jackson credits Simo Ahava with the framing they now use: it is fine for data to be different as long as it is consistently different. The productive question is whether the relationship between two sources has changed, and why—not why they do not match.
- **The definition of “new customer” is a business decision that platforms make for you if you do not**: Jackson describes acquisition campaigns that look healthy on cost per acquisition until a granular report or a finance question reveals that a meaningful share of the people acquired already had purchase history with the brand. Tracking usually gets blamed first, but Jackson argues that nobody had formally agreed what a new customer is before anything was built. Instead, teams inherited the definition used by whichever system did the measuring, typically a session-level, cookie-scoped, device-scoped default shipped by the ad or analytics platform. Different teams have genuinely competing interests in the answer. Jackson breaks “new” into at least four distinct ideas, including new to the brand and new to the category. The fix is to write down a business definition before building audiences.
- **In dealerships and franchises, the attribution gap is an ownership problem, not a tracking problem**: The standard recipe is to capture the click ID, store it in the CRM next to the lead, and send the closed deal back to the ad platform through a server-side setup. Jackson argues that this assumes every step of the pipeline belongs to the same company. In a franchise model, the franchisor owns the brand and funds national marketing, each franchisee is a separate legal entity that holds customer records, and a third vendor owns the software holding the sale. Access to a sales record becomes a commercial product with its own pricing. The gap survives every new measurement technology because the data is not missing: it sits complete and accurate in thousands of systems owned by parties who have weighed what sharing it gets them and mostly concluded it is not enough.
- **Channel and measurement strategy has to follow the business model, not the trade press**: Reporting from d3con in Hamburg, Jackson contrasts the direct-to-consumer playbook the industry generalises from with what brand-side marketers at large, traditional businesses described on stage. Jackson argues that a brand’s buyers and the way money physically reaches the company should determine its channel and measurement approach, not the newest tactic. The industry treats these established businesses as edge cases while generalising from a narrow slice of venture-backed commerce. What looks like slowness in large consumer goods companies, Jackson argues, is a highly evolved form of risk management: mistakes in product formulation or supply chain carry consequences across dozens of jurisdictions.
- **The test of analytics and AI work is whether a decision changed, not how efficient it became**: Jackson questions AI mandates driven by fear of missing out and the wish to look innovative, and applies the same test to analytics tooling. Their standard is effectiveness rather than efficiency: if you cannot point to a decision that changed, you have spent a lot of money learning to send emails faster. Jackson argues that distribution was never the problem: analytics data has never been more distributed. Putting an agent on top of a dashboard changes nothing; a stakeholder who ignores the dashboard will ignore the agent too. Jackson’s constructive claim is that the creative asset is the last digital touchpoint a brand still controls, so it is worth testing before launch.
- **The warehouse is the starting line, not the destination**: Jackson treats connecting GA4 to BigQuery and joining it with first-party data as the non-negotiable first step, but argues that the work that actually affects advertising happens further into the cloud stack. The strategic shift, in Jackson’s view, is moving from using data for reporting to using it to power advertising. The browser session is fading as a reliable source of commercial value as generative search handles intent before anyone reaches a site and more transactions happen server to server. With automation inside ad platforms, the quality of the data fed to them is the main lever a team still controls.
- **A data analysis process starts with business objectives, not the dashboard**: Jackson wrote a free step-by-step guide to their approach to analysis after looking for a resource for the data storytelling and visualization course they teach. The guide separates three jobs often lumped together as data analysis: engineering and modelling; business intelligence dashboarding; and the business analyst’s work of interrogating prepared data, enriching it with context and turning it into a narrative about the current state, opportunity and recommendations. Jackson places themself in the third group and insists the work begins with business objectives and market context before anyone opens a dashboard. They also note that their process is shaped by their background in business, product and conversion rate optimization, and credit Marco Giordano with enriching the search approach in the piece.
- **Technical people need commercial literacy to make their work count**: In a yearly LinkedIn reminder, Jackson urges people in conversion rate optimization, digital analytics, CRM, data science, development and engineering to take a product marketing course. Technical skill alone, they argue, doesn’t show how a business makes money or how to contribute value; the farther you are from the source of revenue, the harder it is to prove the value of your work. Jackson credits positioning, messaging, product launches, narrative, go-to-market principles and competitive intelligence with improving the questions they ask, and calls product marketing the best extra skill set they’ve invested in.

## Work to read

- [How to do data analysis: a step-by-step guide](https://julianajackson.substack.com/p/how-to-do-data-analysis)
- [The business model is the strategy](https://julianajackson.substack.com/p/the-business-model-is-the-strategy)
- [Why is my data different? GA, Google Ads, and the mismatch question](https://julianajackson.substack.com/p/why-is-my-data-different-ga)
- [Offline sales data integration in dealerships and franchises](https://julianajackson.substack.com/p/offline-sales-data-integration-dealerships-franchises)
- [How to define a new customer](https://julianajackson.substack.com/p/how-to-define-a-new-customer)
- [Google Cloud, AI, and marketing creative performance measurement](https://julianajackson.substack.com/p/google-cloud-ai-marketing-creative-performance-measurement)
- [Take a product marketing course](https://www.linkedin.com/posts/activity-7427072801506758658-DoFc)
- [Podcast appearance on AI mandates, analytics distribution and creative testing](https://www.linkedin.com/posts/activity-7485605220417208320-MmJH)
- [The browser session is fading: server-side, signal quality and new channels](https://www.linkedin.com/posts/activity-7470734240855437312-EXLJ)

## Brand relationships

- **Jellyfish** (employment)
- **Stape** (disclosed_partnership)
- **Standard Deviation Podcast** (mention)

## Partnerships

Juliana Jackson’s brand partnerships appear in their Beyond the Mean newsletter and related live programming. Three newsletter editions disclose a partnership with Stape: one examines why analytics platforms report different numbers, another explores offline sales data across dealerships and franchises, and a third questions how brands define a “new customer.” A LinkedIn announcement also promotes a live Substack session with Stape’s CEO on data pipelines, useful signals and server-side tracking. The examples show subject-led collaborations integrated into practical measurement discussions, rather than standalone ad placements.

## Alternatives

- Elena Dyulgerova: Choose Elena Dyulgerova over Juliana J. if the training company's clients are mainly SaaS SEO and content teams, and the pain in the room is slow execution rather than measurement interpretation. Dyulgerova's posts argue that branded search is an underrated trust signal and that struggling SaaS businesses usually lose to faster competitors because too many departments review a single blog post or a strategy takes months to approve. That gives a workshop a clear operational spine: how to cut review queues, ship SEO and content work faster, and treat branded search as a cross-channel outcome, not just a keyword metric. It is the stronger choice when the buyer wants a blunt execution and accountability session with explicit actions for marketing teams.
- Hasnat Ali: Choose Hasnat Ali over Juliana J. when the workshop problem is ecommerce paid media and first-party data activation, especially for clients selling in markets where cash on delivery changes buying behavior. Ali's Bonkers.corner case explains why a drop model that worked in India did not transfer to Pakistan because 65-70% of ecommerce transactions were cash on delivery, so scarcity lost force when customers had nothing at stake. Ali's first-party data post gives a three-step practical sequence for Meta and Google: upload customer lists as custom audiences, connect CRM data to ad platforms where possible, and exclude existing customers from cold campaigns. For a training company whose clients need immediate paid media actions rather than a narrative workshop, this is a more direct operational fit.
- Bivas Kumar Roy: Choose Bivas Kumar Roy over Juliana J. when the client problem is a specific ecommerce tracking discrepancy and the audience needs a diagnostic method. Roy’s posts lay out a concrete investigation for Meta Ads reporting more purchases than Shopify or GA4: compare attribution windows and click-through versus view-through credit, check Meta Pixel and Conversions API deduplication with shared event IDs, audit duplicate or test purchase events, inspect the payment checkout journey, and review consent and measurement limitations. Roy separates tracking discrepancies from attribution discrepancies and gives an audit order: tracking, deduplication, data quality, consent, attribution. That’s more precise than a general data storytelling workshop when the room is made up of tracking specialists, media buyers and analytics engineers.

---
AI-assisted summary of public posts. Audience demographics are not available. Engagement appears only where we checked recent posts. Method: https://b2binfluencer.marketing/methodology/