--- title: "Fidelity score bands by channel (dataset) | AdBuyMCP" description: "Every fidelity score AdBuyMCP's compilers can return, from 85 on CTV to 38 on paid social, with the grade, what your brief must supply and a CSV." url: "https://adbuymcp.com/data/fidelity-bands" --- [Home](https://adbuymcp.com/) / [Channels](https://adbuymcp.com/channels) / Fidelity score bands Dataset # How precisely each UK channel can reach your customer A fidelity score grades, out of 100, how much of your audience survives into a channel’s own targeting language. This table lists every score AdBuyMCP’s compilers can return on seven channels, what your brief has to supply to reach it, and which channels never move whatever you write. It downloads as a CSV. [Start now](https://adbuymcp.com/signup) [Book a working session](https://adbuymcp.com/contact#book) A short application: five questions, no card. Approved accounts get the whole loop in their own sandbox, up to a monthly cap. Live spend is open to paid design partners. [Download the CSV](https://adbuymcp.com/data/fidelity-bands.csv) - **Highest score**: 85 / 100 - **Lowest score**: 38 / 100 - **Bands published**: 13 - **Last checked**: 25 September 2026 One rule everywhere Naming a city the UK geography table has no mapping for costs 15 points on any channel, and the plan names the city. The table ## Seven channels return thirteen bands between them Three channels move with what your brief says, across three bands each. The other four return one score whatever you write, because that is the finest claim their targeting language can carry. | Channel | Score | Grade | What your brief supplies | Moves with the brief | | --- | --- | --- | --- | --- | | [CTV & streaming TV](https://adbuymcp.com/channels/ctv-streaming) Band 1 of 3 | 85 | person | At least one marketplace segment matches the persona strongly. | Yes | | [CTV & streaming TV](https://adbuymcp.com/channels/ctv-streaming) Band 2 of 3 | 65 | household | Segments matched only weakly, so the plan comes flagged for a person to review. | Yes | | [CTV & streaming TV](https://adbuymcp.com/channels/ctv-streaming) Band 3 of 3 | 60 | household | Nothing in the catalogue matched the persona's terms. | Yes | | [Podcast & digital audio](https://adbuymcp.com/channels/podcast-audio) | 65 | content | Every audio plan lands in this band, because the audio targeting vocabulary goes no further. | No, fixed | | [Addressable & linear TV](https://adbuymcp.com/channels/addressable-tv) | 55 | household | Every addressable TV plan lands in this band, because the channel targets households. | No, fixed | | [Cinema](https://adbuymcp.com/channels/cinema) | 40 | venue | Every cinema plan lands in this band, because the channel targets venues and films. | No, fixed | | [Digital out-of-home](https://adbuymcp.com/channels/dooh) | 50 | geo-cohort | Every DOOH plan lands in this band however well the persona matched, because 50 is the best claim the vocabulary carries. | No, fixed | | [Paid search](https://adbuymcp.com/channels/paid-search) Band 1 of 3 | 82 | content | The brief names the search topics your buyers actually type. | Yes | | [Paid search](https://adbuymcp.com/channels/paid-search) Band 2 of 3 | 68 | content | The brief names interests rather than intent. | Yes | | [Paid search](https://adbuymcp.com/channels/paid-search) Band 3 of 3 | 45 | content | The brief is a persona label and nothing else. | Yes | | [Paid social](https://adbuymcp.com/channels/paid-social) Band 1 of 3 | 62 | content | The brief supplies interest terms. | Yes | | [Paid social](https://adbuymcp.com/channels/paid-social) Band 2 of 3 | 48 | content | Demographics but no interests. | Yes | | [Paid social](https://adbuymcp.com/channels/paid-social) Band 3 of 3 | 38 | content | Neither interests nor demographics. | Yes | Scores and grades last checked against the product’s channel compilers on 25 September 2026. [Download the CSV](https://adbuymcp.com/data/fidelity-bands.csv) One row per band, with the geography penalty and the check date on every row. Licence: [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/). Credit AdBuyMCP and link to this page. The grades ## The grade says how close a channel gets to one person The grade says what the channel’s targeting language can point at. Each step down the list is one step further from the individual you described. 1. 01 ### Person The channel’s targeting language can carry a claim about the person you described. 2. 02 ### Household It reaches the home that person lives in. 3. 03 ### Content It reaches what they read, watch, listen to or search for. 4. 04 ### Geo-cohort It reaches screens in an area, chosen by venue type, place and time of day. 5. 05 ### Venue It reaches cinemas and the films shown in them. The full explanation, with worked examples, is the guide [What a fidelity score measures and why we publish it](https://adbuymcp.com/guides/what-a-fidelity-score-is). Each channel page prints its bands with the compiler’s own rationale for each. [See all seven channels](https://adbuymcp.com/channels) Next step ## See the score your own audience gets. Bring a real brief to a 45-minute session and we’ll plan it live. You describe one audience and watch it turn into native targeting on 7 channels, with the fidelity score and the lawful-basis manifest on screen. It runs in the deterministic sandbox, so it needs no vendor credentials and no card. [Start now](https://adbuymcp.com/signup) [Book a working session](https://adbuymcp.com/contact#book) A five-question application, no card. · [Live spend is open to paid design partners.](https://adbuymcp.com/pricing#design-partners) [See three worked plans](https://adbuymcp.com/planner#examples)