What you get

Data that has already been tested and confirmed as efficient for finding the people you need.

It is data that anyone can use to "hack" these ad targeting tools:

  • Facebook Interests (Detailed Targeting)
  • Google (Display Network Keywords)
  • Twitter (Add followers)

It is data you use to avoid excessive waste and find the people you need in these “spaces”:

  • TV shows.
  • Social media influencers.
  • Locations (zip codes, specific venues).

It is data that can be turned into actionable insights of any target market: from consumers to voters.

Why our approach is different

Segmentation is a 20th century way of thinking:

Meet the Unifying Property approach: a way to define the people you really need

As an example, let’s take a Beer Brand. The traditional approach for them would be to try to target a “profile” such as “beer drinkers: male, 21-42 years old, football fan, visits sports websites”; instead, the Unifying Property approach indicates: if you need “beer drinkers” then let’s find beer drinkers; nowadays you don’t really need “proxies” for that.


  • A Unifying Property for an online English learning service may be “people interested in learning English in the next year”.
  • A Unifying Property for a political party may be “people who are going to vote in upcoming elections”.
  • A Unifying Property for a pharma company may be “people affected by diabetes”.

You just have to provide a brief description of your market or the specific problem you’re trying to solve, and we’ll define the right Unifying Property for it.

Find the people you need,

not the audience they sell you

Predictvia delivers targeting parameters that point to “virtual spaces” visited by the people you need. Unifying Property-based targeting is systematic; targeting based on “audiences” or “inventory” is opportunistic.

Ready-to-use data,

will be the new standard

Unifying Property data is ready to be used immediately as targeting parameters in multiple ad platforms, including digital and TV. It doesn’t have to be “analyzed” by a human specialist for targeting decisions to be made.

Marketing & Research

boundaries erased

It is easy to obtain sophisticated research about your market; but those insights are not immediately actionable. Predictvia’s are.

Artificial Intelligence

for finding the needle in the haystack

The current structure of ad platforms supplies millions of targeting parameters options. Predictvia’s algorithms select those that are efficient for reaching people who have a Unifying Property

Know your waste

Be certain about your marketing spend

Data-driven decisions control waste of resources. Unifying Property-based data lets you know, beforehand, how much of your ad-spend will be a waste, allowing you to make decisions based on calculated risk.


Data that points only to those “virtual spaces” where you will find
a high proportion of the people you need for your business goals:


Highly specific clusters in digital platforms.


Specific geographic locations (zip codes, venues, etc.).


Social media influencers.

TV Shows

Specific shows/periods of time.


Actionable insights that allow you to better predict this group of people.

How our platform works

Show Predictvia Summary Diagram
Predictvia Summary Diagram

Unifying Property: “Beer Drinkers”

Region: Florida, US.






(90’s Bjork Records,

(zip code: 33132, Marlins

(Alexis Ren, Kim Garst,



Netmums, Lifehacker,…)

Stadium, Starbucks,…)

Sarah Arrow,…)

The Flash, GoT…)

Validation and Quality




Alexis Ren


Control Process



Dynamic Adaptive

40% Beer Drinkers

75% Beer Drinkers

40% Beer Drinkers

50% Beer Drinkers









Base Count


Google Consumer

Social Media

TV Spots


Sponsored Message

Targeting Tools

Targeting Tools




Marketing & Research



Get access to our product demo


One of our Core Technologies

Seenatra is a predictive analysis, machine learning-based method fed with large datasets gathered from social media and directly from users. This method accomplishes two major results:

  1. It trains a model that classifies any given user as positive or negative for having the Unifying Property.
  2. As a result of the modeling process, it identifies certain variables, mostly “human interests”, that are predictive of the Unifying Property. Essentially, a few “human interests” are “machine-picked” from thousand of options.

So then, we obtain the first stage outcome: Tags, Locations, Influencers and TV Shows, that are predictive of the Unifying Property

- Read more -


Since we are able to apply advanced modeling techniques on data from Internet Users, either from Social Media or directly from them, we can find the precise “virtual spaces” where people with the Unifying Property are “located”. These “virtual spaces” are validated social media targetable interests.

- Read more -

Validation and Quality Control Process

Predictvia measures how good the Seenatra selected interests are, via Hypothesis Test run through Dynamic Adaptive Sampling-based surveys of people who have them. The “good” interests are the Unifying Property Finders.

Dynamic Adaptive Sampling

One of our Core Technologies

Dynamic Adaptive Sampling is an intelligent sampling technique that optimizes a sample design as the data is being collected. The method analyzes data from every stratum of a sample design and then decides if it needs more sample, according to how “uniform” is such stratum.

Because this occurs simultaneously to the surveying process, DAS is able to “ask” for more completed questionnaires on an as-needed basis. Therefore, this technique improves the sampling process by reducing collection time, sample size and cost. But more importantly, it guarantees a sample that efficiently represents the universe.

- Read more -

Base Count

Survey that means counting how many individuals have the Unifying Property (Beer Drinkers) within a region (Florida, US). This survey is representative of population of the region (country, state, county, etc.).

- Read more -


How does Predictvia know if a Unifying Property Finder is good?

First, we use the metric obtained by the Base Count Survey, which indicates the proportion of people who have the Unifying Property among the region general population.

Second, we run a survey among the group of people who have any interest (Predictor, Location, Influencer, TV Show) that Seenatra has previously identified as a potential Unifying Property Finder. Through this survey, we obtain the exact proportion of people who have the Unifying Property among those who have the interest.

Finally, we compare these two proportions: if the second (Unifying Property frequency among those who have the interest) is larger than the first (Unifying Property frequency among region general population), then we know this interest is a good Unifying Property Finder.

This comparison yields the Boost: statistical measure of how many times a specific interest (Predictor, Location, Influencer or TV Show) is better at finding people who have the Unifying Property, in comparison to randomness (region general population).

Boost your Marketing & Research efforts:

Validated Tags, Locations, Influencers and TV Shows are targetable by definition. So you can precisely reach the people you need (people with the Unifying Property) by available digital and TV targeting tools, for persuading them by marketing campaigns or getting insights via surveying campaigns.

Tags: applicable for setting Facebook, Twitter and Google targeting campaigns for marketing and research purposes.

Locations: applicable for setting Facebook, Twitter and Google targeting campaigns for marketing and research purposes. And specifically via Google Consumer Surveys for obtaining insights on people who have been at a specific location.

Influencers: applicable for Influencer Sponsored Message campaigns for marketing and research purposes.

TV Shows: applicable for TV Ad campaign on specific TV Spots for marketing and research (via hashtag based Ads) purposes.

Campaigns for marketing or research purposes can be conducted by Predictvia upon request.

Contact Us