# precise.ai · full text This file is the complete text of https://precise.ai/ and its pages, as plain Markdown for retrieval. Charts on the site are drawn from a synthetic fixture; their captions and readouts are included here as text. Last updated 2026-09-21. --- # Home Source: https://precise.ai/ ## Make more of every media dollar. Improve returns while your campaign is in flight. The purple line shows the predicted return if you act on our recommendation. The shaded area shows the range around that prediction. The black line shows actual results. The gray holdout line shows results for the group that kept the original settings. The dashed gray line projects the return if you keep the current plan. Connects to your data and your stack; lands in your dashboard, or an agent of your own. Chart: Our recommendation: predicted and measured returns. Return per $1 of media and fees, Aug 1 to Aug 29, as known Aug 15. A recommendation was acted on Aug 6. Since then the measured line is read against a holdout that kept the old setting. Illustrative campaign. Return means revenue per $1 of media spend and fees. Predictions are estimates. ## One campaign. Same budget, more return. See where to move the next dollar while your campaign is live, and the return that change could deliver. You decide what to put into action. Three ways to improve (illustrative): - Buy through a better-priced platform. Move Hulu from DSP A to DSP B to lower costs and improve expected return. - Move budget toward the stronger publisher. Shift $600 a week from Stream TV to Daily Living, where the same budget is expected to deliver more. - Pay less to reach the same audience. Replace the Hulu private deal with the lower-priced open auction. Selecting one shows the expected revenue per $1 spent at the same spend, with its range and the scored history it stands on. Illustrative estimates for the selected change, at the same spend. ## Know what earns its cost. Precise evaluates campaign attributes individually and together to find what earns its cost and where spend could work harder, while the campaign is live. Attributes evaluated: audience segments, data layers, creative, publishers, supply paths, DSPs, frequency, and their combinations. ### See how your spending aligns with your revenue. Spot the gaps while your campaign is live. Here, Stream TV receives 24% of spend but accounts for 12% of revenue. Daily Living receives 21% of spend and accounts for 33% of revenue. These gaps flag allocations worth examining. Precise looks at the underlying attributes and recommends where to move the next dollar while the campaign is still running. Chart: Campaign so far, share of spend vs. share of revenue (illustrative, results to date). | Placement | Share of spend | Share of revenue | |-|-|-| | Hulu · DSP A | 31% | 22% | | Hulu · DSP B | 6% | 9% | | Daily Living | 21% | 33% | | Neighborhood News | 18% | 24% | | Stream TV | 24% | 12% | Illustrative results to date. Revenue totals may change as more sales are recorded. ## Measure. Learn. Improve. Predict. Act. Check the result. Keep improving while the campaign is live. 1. Understand the recommendation. See the finding, the recommended action, and the expected improvement against your objective. Each recommendation explains where the budget would move and why. 2. Track what was implemented. Record which changes were made, when they happened, and how spend moved. Review the action taken alongside the original recommendation. 3. Measure the return. Compare expected and observed results over the agreed outcome window. Precise uses that evidence to refine the next recommendation while campaigns remain in flight. One reviewable record, from recommendation to result. ## Better returns. Built into your workflow. Precise connects to your data and your stack, and lands where your team already decides: a dashboard, or an agent of your own. Same methods, same record, same controls either way. ### Your data: connects wherever your logs live. Delivery, cost and outcome logs stay where they are: warehouse, clean room, platform exports. Precise reads at whatever grain the source has, and says what one more connection would add. ### Dashboard: open it and act. What changed since you left, the moves worth making, what each stands on, and the one motion that puts a change on record. Hosted by Precise; your team reviews and decides. ### Your agent: an agent that works for you. Precise provisions your advocate agent, or answers inside the agent you already have. It watches the campaign and recommends. For the kinds of change you have cleared, one rung at a time, it makes the change and puts it on record. It answers to your rules, not ours. ### Your workflow: built in with you. We work with your team to put in-flight measurement where decisions already happen: where a recommendation lands, who confirms it, how the result is read back. ### Reach and overlap analysis in live campaigns. Agency teams have used Precise to analyze reach and audience overlap across DSPs and media partners, informing reallocations within live campaigns. Client and partner identities remain confidential. This describes live use; the product example above uses illustrative estimates. ### Your objectives. Your controls. Your team sets the business objective, budget, permissions, and approval rules. Precise works within those controls. Agencies retain their client relationships. ### Review when you need to. Precise works in the background through your existing data, platforms, and agents. Review recommendations, changes, and results in the same workflow, so the next decision can build on what happened. ## Applied science. Better decisions. Value every attribute. Improve the allocation. Measure the outcome. Precise is an applied research lab in New York. We build intelligence for better economic decisions. Our media application helps brands and agencies understand what earns its cost, improve allocations, and learn from measured results. Our team brings experience from Madhive, AdLedger, BlockScience, and Cognitiv: building media infrastructure, studying complex systems, and turning research into working products. Team: - Spencer Potts, Co-founder & Chief Executive Officer - Adam Helfgott, Co-founder & Chief Technology Officer - Matt Barlin, Chief Science Officer - Dave Antonelli, Chief Revenue Officer ## Put more of your budget to work. Start with one active campaign and the outcome you want to improve. Bring the people responsible for media performance and data so recommendations can move into action. - Connect. Bring campaign data, costs, and outcomes from your existing systems. Choose the access and deployment model that fits your operation. - Make it yours. Set the objective, budget, permissions, and approval rules. Agree the integration, rollout, and software fee. - Keep improving. Put recommendations to work. Track implemented changes and measured results, then extend across campaigns and accounts as value is established. ### A few practical questions. **How will we know whether returns are improving?** Agree the business outcome, relevant costs, comparison, and measurement window up front. Review what Precise recommended, what was implemented, and the measured result against the expectation. Those results inform the next recommendation. **How does Precise fit our existing setup?** It connects to your data where it already lives and lands where your team already decides: a dashboard hosted by Precise, or an agent of your own. That can be your advocate agent, provisioned by Precise, or the one you already run. It recommends and, for the kinds of change you have cleared, acts and records. Teams that want it deeper have us build it into their workflows with them. Your team reviews the reasoning, changes and results whenever needed. **What does our team control?** Your team sets objectives, budgets, permissions, and approval rules. You choose the data, access, and deployment environment. Data handling, retention, and deletion follow your agreements with Precise. See Security and data. ## Start improving your media returns. Let's find where your budget can do more, and agree how to measure the difference. Contact: hello@precise.ai. The contact form prepares an email in your own mail app; nothing is sent until you send it. Your objectives. Your controls. A clear record. New York · Working wherever you do. Intelligence for better media returns. Independent economic valuation. © 2026 Precise.ai, Inc. --- # Methods Source: https://precise.ai/methods ## How a number gets onto the page. Three layers, each reviewable: contribution methods value every attribute, decision models turn values into a recommendation, and the record checks the recommendation against what was measured. This page describes method, not results. Every figure on this site is a synthetic example unless it says otherwise. ### 1 · Contribution methods An impression carries many attributes at once: the creative, the audience segment, the publisher, the supply path, the DSP, the daypart, the frequency. Each can contribute value on its own and change value in the company of the others. Precise assigns each attribute the share of the outcome it can fairly claim, individually and in combination, against the cost it added. The unit is the same everywhere: outcome value per $1 of media spend and fees, so a publisher and a data layer can be read on one scale. Where two attributes reached the same people, their contributions are not summed. The overlap is named and, where only platform reports exist, the number is labeled gross. ### 2 · Decision models A value is not yet a decision. Each combination can absorb only so much spending before the next dollar returns less; the response curve says how much. Precise reads the curve for the cell that would give and the cell that would receive, and prices the move: how much, and what is expected at the horizon. Every expected figure carries a range. The range comes from scored history: moves of the same kind whose outcomes have already been measured. Fewer scored moves, wider range. Where there is not enough evidence, that is the answer, said in words. ### 3 · The record A recommendation is written down with what was known, what was chosen, what was expected, and, later, what happened and what it taught. Sent, reported and observed are three separate facts. Scoring happens at the agreed horizon against a holdout that kept the old setting, so the next range stands on more evidence than the last. Nothing on the page is typed by a model. Words come from the assistant; every figure comes from these methods with its basis attached. Chart: Scored history for one kind of move, what a predicted line stands on. Each point is one scored move: the gain in return per $1 at the 7-day horizon against its holdout. Synthetic fixture, n = 14. --- # Security and data Source: https://precise.ai/security ## Your data stays under your control. Your team chooses the data, deployment environment, and people with access. We build the integration around those choices. Last updated September 7, 2026. ### Where Precise runs. Deployment and integration are agreed with your team, including the environment, systems to connect, and permissions for moving data and results. ### Access and control. Your agency sets the objectives, permissions, and approval rules. Recommendations, reasoning, and results remain available for your team to review. Access follows the controls agreed for your deployment. ### Data handling. Data handling, access, retention, deletion, and audit rights are defined in the data processing agreement and master services agreement. Security documentation and subprocessor information are available for your team's review. ### A clear counterpart. Your counterpart is Precise.ai, Inc. Your agency retains its client relationships. Precise owns its software and research. Our website terms and privacy policy are public. ### Start the security review. Connect your security, data, and business owners with our team. We will work through the required access, deployment, and agreements together. Contact: hello@precise.ai, subject "Precise security review". --- # Privacy policy Source: https://precise.ai/privacy What this website collects and how we use it. Last updated September 7, 2026. ### What this site collects. The contact links and the form open your email app with a message ready to send. We receive only what you choose to send us by email. This website has no analytics scripts, advertising trackers, or tracking cookies. Fonts, images, scripts, and styles load from this site's own domain. Vercel hosts the site and keeps standard infrastructure access logs. ### How we use your message. We use your message to respond to your request. We do not sell it or share it for marketing. Email hello@precise.ai to ask us to delete your message. Opening an email link does not send a message. Nothing reaches us unless you send it. ### Customer data. Data processed by Precise software is governed by your signed agreements, including the data processing agreement and master services agreement. Your team chooses the data, deployment environment, and people with access. See Security and data. ### Changes and contact. We may update this policy and will change the date shown on this page. For privacy questions, contact hello@precise.ai. --- # Website terms Source: https://precise.ai/terms Terms for using this website. Use of Precise software is covered by your signed agreements. Last updated September 7, 2026. ### This website. This website is operated by Precise.ai, Inc. It describes our software and lets you request a briefing. Its content is provided for information. Contacting us does not create a contract for the software or services. ### Your agreement with Precise. Use of Precise software is governed by signed agreements, with terms your counsel can review. Nothing on this site creates a software license or warranty. Where these website terms and a signed agreement differ, the signed agreement governs. ### Content and marks. The content, design, and marks on this site belong to Precise.ai, Inc. You may quote the site with attribution. ### The site is provided as is. We keep the site accurate and current. Product examples are labeled illustrative. The site itself is provided as is, and Precise.ai, Inc. accepts no liability arising from reliance on the site alone. Product commitments are defined in the signed agreements. ### Changes and contact. We may update these terms and will change the date shown on this page. For questions, contact hello@precise.ai. See also our privacy policy.