Pursway was a real marketing-analytics company, and its 2014 Pursway Connect launch was described as a way to turn inferred social connections into marketing prospects. Clients supplied customer information; Pursway said it used public and open-web clues to identify people connected to those customers and returned prospects’ names and street addresses. The “Israeli intelligence” description came from the company’s CEO, as reported at the time—not from independent verification that an intelligence agency supplied the software. Pursway was later acquired by IPG in September 2015; the available records do not establish that Connect still operates in 2026.
What Pursway was
Pursway was a business-to-business marketing-analytics company founded in 2005, associated with Herzliya, Israel, and also reported to have had a Waltham, Massachusetts presence. It was not a consumer social network or simply an advertising-cookie service. Its products focused on customer analytics, social influence and expanding the audiences marketers could reach. Startup Nation Central’s company profile lists Pursway as acquired and inactive.
The VentureBeat article that prompted the headline was published on August 12, 2014, alongside reporting on the launch of Pursway Connect. It described a system intended to identify people socially connected to a marketer’s existing customers, then make those people available as prospects. VentureBeat’s 2014 report is a historical account, not evidence that the product remains available today.
How Pursway Connect reportedly worked
- A client provided customer data. The reported example was a customer list containing names and ZIP codes.
- Pursway matched those customers against its relationship database. The company said that database was assembled from public and open-web information.
- The system inferred connections. It looked for people who might be socially connected to the customers, rather than relying only on a declared online friendship.
- The client received an expanded prospect list. VentureBeat reported that the output included names and street addresses. It was a “flat” list, not necessarily a map telling the client exactly which person was connected to which customer.
- The client could market to those prospects. The point was to reach beyond existing customers using social affinity as a targeting signal.
That distinction between an internal inference and a client-facing list matters. The client reportedly did not receive a simple “Joe is friends with Mary” relationship map, but the audience expansion still depended on Pursway making relationship judgments behind the scenes.
Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchWindows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstall#1 Best Overall
What counted as a “friend”
Pursway’s use of “friend” was broader than a confirmed Facebook friendship or a person’s own description of a close relationship. The reported examples and descriptions encompassed family, acquaintances, neighbors, people whose children attended the same school, alumni, club members, coworkers and other associations that might appear in public records or online traces.
- Declared relationship: One person explicitly identifies another as a friend or connection.
- Observed association: Two people appear in the same directory, event list, workplace history or online community.
- Statistical inference: A system combines signals and estimates that a relationship or influence exists.
- Marketing prediction: A marketer treats that estimated connection as a reason one person may affect another’s purchasing behavior.
These are not equivalent. A person classified as a “friend” by a marketing model might not be considered a friend by either person involved. The sources do not provide Pursway’s model architecture, feature list, error rate or an independent validation study.
What data it used—and what the company said it did not use
Pursway said it relied mostly on public or open-web material, including social profiles, alumni information, employment history, event attendance, clubs, blog responses and other scattered sources. Its 2013 investor announcement described a database built from thousands of open-web sources. These examples are not a complete technical inventory, and the company’s account was not an independent audit. The October 2013 investor announcement also described the company’s audience and relationship data.
Pursway’s CEO told VentureBeat that the system was not based on browsing behavior. That is the company’s statement as reported in 2014, not a separately audited finding about every data source or processing step.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
“Public” does not mean harmless, accurate or expected to be used for advertising. A person might share information for one purpose or audience; a data company can combine it with unrelated records, infer a relationship and make the result useful to a marketer. The privacy issue is therefore not limited to secret access. Aggregation, inference and reuse can create a new profile from fragments that seemed ordinary on their own.
Why marketers wanted social-relationship data
Pursway’s commercial premise was that people can influence one another’s purchases. Its system was intended to identify influential customers and extend a campaign to people thought to be socially close to them. The reporting named Comcast, Orange and Sony as users or clients of Pursway services generally; it does not establish that all three used Pursway Connect specifically.
Performance figures attached to Pursway should be read as promotional claims, not independently verified benchmarks. VentureBeat reported a company claim of a 50% or better improvement in return on investment in many cases. The 2013 investor announcement cited one campaign with a 200% increase in results and conversions, while a Crunchbase description associated the company with claims of 2–5× ROI lifts. Those measures are not interchangeable, and the available descriptions do not establish consistent control groups, industries, time periods or independent verification. Crunchbase’s Pursway profile summarizes company-associated information.
To assess any such result, a reader would need to know what was compared, whether the figure meant response rate, conversions, revenue or ROI, whether other targeting changed at the same time, and whether results were replicated. The cited claims alone do not answer those questions.
Quick wins for a faster PC:
Scan for outdated or missing drivers - takes under a minuteDriver Scan →Repair Windows errors before they cause bigger problemsFix Now →What PIVO meant
PIVO stood for Purchase Influence Value Order. Contemporary descriptions used it for Pursway’s broader approach to finding influential customers and social relationships that could shape purchasing. In broad terms, it appears to have combined relationship inference, customer or transaction information, influence scoring, audience expansion and campaign optimization. The public descriptions do not provide enough technical detail to reproduce the method or judge its accuracy.
PIVO and Pursway Connect should not be treated as interchangeable names. Connect was described as the product that expanded a customer base into a prospect list; PIVO referred to the wider influence-analysis approach or technology.
How large was the relationship database?
Published figures differ in both geography and apparent scope. VentureBeat reported more than 150 million North Americans in the relationship database, while the October 2013 investor announcement described more than 100 million U.S. consumers and their social relationships. Startup Nation Central later repeated a figure above 150 million North American consumers. These are source-reported estimates, not a single independently verified count; different dates, geographies or counting methods may account for the difference.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Why the privacy concern went beyond a prospect list
The system’s usefulness and its risks came from the same process: combining weak clues to produce a stronger-seeming conclusion about who was connected to whom. Potential errors include stale employment or address information, treating former coworkers as current influencers, mistaking event co-attendance for a meaningful relationship, or incorrectly linking people because of overlapping names or household details. The available reporting does not quantify how often such errors occurred.
Best Value
A returned list of names and addresses is not the same as disclosing a visible relationship map to the client. But the lack of a displayed map does not remove the inference: the system still had to estimate relationships to build the audience. Nor does the fact that data was described as public answer whether people expected those records to be joined and used for a different purpose.
What the 2014 opt-out report said
VentureBeat reported that individuals could not opt out directly through Pursway at the time. The company’s stated position was that an opt-out would have to be handled by the marketing client receiving the list. This is a description of the reported 2014 policy, not a statement about current data-broker law or any present-day procedure.
A client’s campaign suppression list is not necessarily the same as removal from a data source or an industry-wide opt-out. Suppressing a record for one campaign does not by itself show that it was deleted from every database or that another company will not hold similar information. Current rights and available remedies depend on jurisdiction, which company still holds the data and applicable law; the historical account does not settle those questions.
What “software from Israeli intelligence” actually establishes
Pursway’s CEO described the company’s machine-learning methods as drawing on “anti-terrorism” techniques originally developed by Israeli intelligence, according to VentureBeat. The reporting does not independently identify an intelligence organization, specify the algorithms, or establish that software was directly transferred from an agency. The headline’s phrasing should therefore be understood as an attributed company-origin claim, not proof that Pursway used an intelligence agency’s surveillance database or had access to private accounts.
What happened to Pursway?
Transaction and company-history records report that The Interpublic Group of Companies (IPG) acquired Pursway in September 2015. Startup Nation Central now lists the company as acquired and inactive, while a MarketScreener transaction listing reports the acquisition. The available records do not establish whether the Pursway brand or database continued, whether technology was integrated into another product, or whether the original data was retained, deleted or repurposed.
For a person investigating a current marketing database, the practical questions are specific: what categories of data are held, where they came from, what inferences or scores are attached, who received the information, and whether correction, suppression or deletion applies to the whole database or only one campaign. The historical Pursway reporting does not establish a current contact or consumer-request process.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
