Driving an AI Leader's Global Expansion
We helped a leading AI company uncover the true enterprise opportunity across LATAM and MENA by mapping company revenue, paid marketing spend, and media mix to identify the highest-value accounts in each priority market.
- Published
- 14 Apr 2026
- Reading time
- 3 min read
- Author
- Stride
- Category
- Case Study
Net-new accounts
30%
of the highest-priority accounts were uncovered as effectively net new once local operating teams were tied back to the right financial structures.
Priority targets
Top 100
spenders identified in each priority country to give commercial teams a defensible ranked account view.
Vendor miss rate
50%+
of reviewed vendor revenue estimates were more than 2x the true value.
The goal
As a leading AI company expanded into emerging markets, it needed a more reliable view of where enterprise demand actually sat. The brief was clear: identify the top 100 spending companies in each priority country across LATAM and MENA, estimate their revenue, understand paid marketing investment, and map the relative share of wallet across the key partners in the region.
The data was far less clear than the brief. Commercial activity often sat in one market while the financial entity sat in another, which distorted traditional location mapping and obscured the real buying centre.
Existing vendors were also of limited use. Most offered little on marketing spend or media mix, and in more than 50% of the cases we reviewed, existing CRM enrichment providers revenue estimates were wrong more than 2x the true value. Given these were often large, publicly traded companies, this was particularly surprising.
Finding the Real Buying Centre
Legal structure can hide where buying power actually sits. Stride agents help reveal where budget and commercial activity are really concentrated.
Enrichment estimates more than 2x off
50%+
In more than half of the reviewed cases, existing CRM enrichment provider revenue estimates were wrong by more than 2x, which made ranked expansion planning hard to trust.
Approach
We used Stride's company universe to map the enterprise market across LATAM and MENA, then applied our revenue, paid marketing, entity resolution, and media mix agents across the full population to isolate the highest-priority enterprise targets in each country.
To improve reliability, we evaluated the outputs against known reference points using bespoke ML models, recalibrating the system through repeated testing. The result was a clearer, more decision-ready view of which companies were truly active in each market, where commercial opportunity was concentrated, and where the client's teams should focus first.
Results
This materially changed the target universe. Approximately 30% of the highest-priority accounts were effectively net new, hidden by earlier location-entity mapping that failed to connect local operating teams with offshore financial structures.
The client could now prioritise the highest-value enterprise accounts with greater confidence, sharpen its view of the addressable market, and focus expansion effort where commercial potential was real, not merely visible on paper. We have changed identifying details, but the underlying results are real.
Conclusion
From fragmented signals to a ranked expansion plan
The case shows what changes when entity resolution, spend intelligence, and market context are treated as one problem rather than three separate datasets.
- 30% of the highest-priority accounts emerged as effectively net new.
- The top 100 spenders were identified in each priority country across LATAM and MENA.
- Internal validation suggested roughly 80% accuracy on overall spend identification and 60% on partner-level media allocation.
Benchmarking vs Frontier Models
We benchmarked the work against deep research outputs from leading frontier models, including GPT Pro and Gemini Pro. Those tools were directionally useful, but in our internal validation they identified roughly 50% of the true top 100 companies by overall media spend and around 30% of partner-level media spend.
Our estimate for Stride's performance was materially stronger, at roughly 80% on overall spend identification and 60% on partner-level media allocation. That estimate should still be read with caution because it is based on internal validation.
The gap was most visible in B2B, manufacturing, and other traditional sectors, where general-purpose models tended to overestimate marketing spend and miss important accounts altogether.