Fusepoint Insights review

Strong — 4 of 5 on this rubric’s scale, from Poor (1) to Excellent (5). This dimension carries 25% of the total score.

Benchmark — across the 16 other agencies evaluated in this discipline, this dimension runs Excellent 1 · Strong 3 · Adequate 12. The typical agency here scores Adequate, and 1 of them score higher than this one.

What this dimension measures: The heaviest weight, because an unvalidated model is an opinion with decimal places. Look for a described methodology, stated assumptions, and above all how the model is VALIDATED: holdout periods, backtesting, or reconciliation against a real experiment. A described method with no validation story scores 3.

Scores high — A firm that publishes its validation approach and names its limits scores 4-5.

Scores low — 'Proprietary algorithm' with no methodology at all scores 1-2 — proprietary is not a method, and in this category it is the single least checkable claim a vendor can make.

What we found — The methodology page describes a five-stage loop (Model, Media Planning, Incrementality Testing, Refine, Optimize) and states the MMM is built on '24+ months of geo-level data across platforms', producing saturation curves and diminishing-returns estimates. Validation is not left abstract: they say they 'validate and calibrate our models using real-world geo-experiments', name synthetic controls and 'the tidysynth R package' for test/control pairing, and state every test carries confidence intervals and causal impact models. The MMM page positions this explicitly against backtesting alone: models 'proven in the real world, not just backtested against history'. The measurement page adds the named BEATS triangulation framework (Business results, Experiments, Analyses, Technology, Surveys) and says divergence between the five is the signal to dig. Held below Excellent because everything is self-published and nothing quantifies model quality: no stated adstock or decay assumptions, no seasonality handling, no holdout-period or backtest accuracy figure, no sample deliverable, and no named limits or failure cases. That is the high band above, which is why it scored Strong.

On the record — “MMM is calibrated against live geo-experiments rather than backtesting alone; the MMM page markets a model 'proven in the real world, not just backtested against history'.” fusepointinsights.com ↗

On the record — “Named validation machinery: 'We use synthetic controls and the tidysynth R package to ensure our test/control pairs are statistically reliable. Every test is validated with confidence intervals and causal impact models.'” fusepointinsights.com ↗

deliverability — deliverability: whether cold email actually lands in the inbox rather than spam. The tell is infrastructure talk — warmed sending domains kept separate from your main domain — because a burned domain outlasts the engagement.

Evidence: vendor stated — the agency’s own claim, recorded as theirs rather than ours.

Excellent — 5 of 5 on this rubric’s scale, from Poor (1) to Excellent (5). This dimension carries 20% of the total score.

Benchmark — across the 16 other agencies evaluated in this discipline, this dimension runs Strong 7 · Adequate 5 · Weak 4. The typical agency here scores Adequate, and none scores higher than this one.

What this dimension measures: Whether the firm can establish causality rather than only correlation: geo holdout tests, matched-market design, PSA/ghost-ad tests, switchback designs, or reconciliation of modelled results against live experiments.

Scores high — Named experiment designs with described execution score 4-5.

Scores low — Correlation-only modelling with no experimental capability scores 2-3 — legitimate and common, but the buyer should know they are buying a correlational estimate. A firm that presents modelled attribution as proven causality scores 1-2, and the overclaim should be named in the verdict.

What we found — Multiple named experiment designs with described execution, not a single label. The incrementality page lists geo tests ('We match predictive geographic markets and isolate the variable (revenue, CAC, leads) to measure true incremental lift against a control'), channel-level tests on Meta, TikTok, Amazon DSP and CTV, tactic-level holdouts on audiences/creatives/placements, and growth-versus-holdout designs. The methodology page adds Holdout Tests, Growth Tests, Media Mix Tests, audience holdouts, sequential testing, time-based blackout tests, creative lift analysis and cross-channel halo measurement, plus synthetic controls via tidysynth. Critically for this category, they describe experiments as the calibration source for the model rather than a separate product line: test results feed back 'as priors for continuous improvement'. The Petal and Pup case study describes a matched-market test on Pinterest at a named client. No overclaim of modelled attribution as proven causality was found on any page read; the site consistently separates the model (hypothesis) from the test (proof). That is the high band above, which is why it scored Excellent.

On the record — “Named experiment designs with described execution: geo/matched-market tests, growth tests, media mix tests, audience holdouts, sequential testing, time-based blackout tests, creative lift analysis, cross-channel halo measurement; channel-level tests on Meta, TikTok, Amazon DSP and CTV.” fusepointinsights.com ↗

On the record — “MMM is calibrated against live geo-experiments rather than backtesting alone; the MMM page markets a model 'proven in the real world, not just backtested against history'.” fusepointinsights.com ↗

attribution — working out which marketing touch actually caused a sale. Good practice names its model and its blind spots; bad practice quotes each ad platform’s self-graded numbers, which overlap and overclaim.

CAC — customer acquisition cost: total sales and marketing spend divided by customers won — the number that says whether growth is affordable.

Evidence: partly checkable — corroborated in part against the sources below; the remainder rests on the agency’s own account.

Strong — 4 of 5 on this rubric’s scale, from Poor (1) to Excellent (5). This dimension carries 15% of the total score.

Benchmark — across the 16 other agencies evaluated in this discipline, this dimension runs Excellent 1 · Strong 8 · Adequate 6 · Weak 1. The typical agency here scores Strong, and 1 of them score higher than this one.

What this dimension measures: Attributable client work at a stated scale and category. Treat any 'we found X% waste' claim as a vendor-stated number unless a client is named and corroborates it.

Scores high — Named clients with described engagements score 4-5.

Scores low — Anonymised case studies score 2-3 — common here for genuine confidentiality reasons, so do not penalise beyond the band, but do not credit unverifiable lift figures either.

What we found — Four testimonials attributed to named individuals with titles at named companies: Caleb Barber (Global Media and Measurement, Sound United), Liz Bradfor (Head of Digital Marketing, Kurt Geiger), Sal Aziz (CEO, Zenwise) and Amanda Byers (Head of Integrated Marketing, eero at Amazon). The Barber attribution is consistent with a LinkedIn profile surfaced in search at HARMAN International, which acquired Sound United, though the profile was not read first-hand. One of nine case studies names its client (Petal and Pup, Pinterest matched-market test). The remaining eight are anonymised, and every headline number sits in an anonymised one: $1.3M profit growth (healthcare apparel), 2.18x ROI growth (luxury handbag), 85% efficiency lift (fashion), $400K incremental profit and 167% lift to incremental revenue (voluntary insurance). Per this rubric those lift figures are vendor-stated and are not credited here. That is the high band above, which is why it scored Strong.

On the record — “Four testimonials attributed to named individuals with titles: Caleb Barber (Sound United), Liz Bradfor (Kurt Geiger), Sal Aziz (Zenwise), Amanda Byers (eero at Amazon).” fusepointinsights.com ↗

On the record — “Of nine published case studies, only one names its client (Petal and Pup, a Pinterest matched-market test); all quantified outcomes appear in anonymised studies.” fusepointinsights.com ↗

attribution — working out which marketing touch actually caused a sale. Good practice names its model and its blind spots; bad practice quotes each ad platform’s self-graded numbers, which overlap and overclaim.

Evidence: partly checkable — corroborated in part against the sources below; the remainder rests on the agency’s own account.

Adequate — 3 of 5 on this rubric’s scale, from Poor (1) to Excellent (5). This dimension carries 15% of the total score.

Benchmark — across the 16 other agencies evaluated in this discipline, this dimension runs Strong 11 · Adequate 3 · Weak 2. The typical agency here scores Strong, and 11 of them score higher than this one.

What this dimension measures: What the engagement needs from the buyer (spend, conversion, and channel data at what granularity) and — critically — whether the firm also buys the media it is measuring. A firm that measures media it does not sell is structurally more credible; where the same firm buys and grades its own work, that conflict must be disclosed and should be named in the verdict whether or not the firm names it.

Scores high — Clear data requirements plus independence from media buying scores 4-5.

Scores low — Undisclosed conflict scores 1-2.

What we found — Data requirements are unusually explicit for this category. The methodology page lists what the engagement needs: first-party conversion and revenue data, marketing activity logs (spend, impressions, creative metadata), product/region/pricing/promo data, retailer and wholesale performance, subscription and retention data, and web analytics (GA, Adobe, Shopify), with the MMM built on 18-24+ months of geo-level history. Independence is where this is capped. Every page footer reads 'fusepoint is a Data Intelligence consultancy by Power Digital', and the About page says fusepoint was 'Born from Power Digital's data-driven DNA'. Power Digital's own site sells Paid Social, Paid Search, Amazon, TikTok and Programmatic - it buys and manages the media. The homepage asserts 'We don't run your media; we make sure the strategy and measurement behind it can survive the boardroom', and the media planning page says its output is 'everything you need to hand off to any competent media team and execute'. So the entity-level separation is stated and the parentage is disclosed on every page - this is not an undisclosed conflict. But nowhere on the site is the actual question addressed: what happens when the media being graded was bought by the parent, and whether any wall exists between the two. The Clutch review tags its one engagement 'Media Planning and Buying', which is Clutch's combined category and does not settle it either way. That is between the two bands, which is why it scored Adequate.

On the record — “The homepage states 'We don't run your media; we make sure the strategy and measurement behind it can survive the boardroom.'” fusepointinsights.com ↗

On the record — “Power Digital, the parent, sells and manages paid media: paid social, paid search, Amazon, TikTok and programmatic advertising.” powerdigitalmarketing.com ↗

programmatic — automated auction-buying of ad space across the web, rather than direct placements.

Evidence: verified — checked against a named source you can open; the links under Sources below are where to check it yourself.

Strong — 4 of 5 on this rubric’s scale, from Poor (1) to Excellent (5). This dimension carries 10% of the total score.

Benchmark — across the 16 other agencies evaluated in this discipline, this dimension runs Excellent 1 · Strong 5 · Adequate 10. The typical agency here scores Adequate, and 1 of them score higher than this one.

What this dimension measures: What arrives and how often: a one-off model, a refreshed quarterly model, a live dashboard, a decision workshop. A model delivered once and never refreshed is a snapshot of a market that has moved, and should be scored as the ceiling it is.

Scores high — Stated deliverables with a stated refresh cadence score 4-5.

Scores low — Undefined deliverables score 2.

What we found — The MMM page states 'Monthly Agile Model Refreshes' - 'We update your model monthly with the latest data' - with 'Ad hoc refreshes available when something big changes', which answers the snapshot-versus-living-model question directly. Deliverables are described by audience rather than as a single artefact: executive summaries, tactical breakdowns and financial translations for the CFO, 'each one structured around the decisions that person actually makes', plus weekly performance monitoring through statistical signals and saturation curves on the methodology page. Time-to-first-model is stated as 'weeks, not months'. Short of Excellent because no sample report or dashboard is published, engagement length is never stated, and it is not said whether the model itself is handed over or only its outputs. That is the high band above, which is why it scored Strong.

On the record — “Stated refresh cadence: 'Monthly Agile Model Refreshes ... We update your model monthly with the latest data', with ad hoc refreshes available on material change; deliverables split into executive, tactical and CFO-facing versions.” fusepointinsights.com ↗

deliverability — deliverability: whether cold email actually lands in the inbox rather than spam. The tell is infrastructure talk — warmed sending domains kept separate from your main domain — because a burned domain outlasts the engagement.

Evidence: vendor stated — the agency’s own claim, recorded as theirs rather than ours.

Weak — 2 of 5 on this rubric’s scale, from Poor (1) to Excellent (5). This dimension carries 10% of the total score.

Benchmark — across the 16 other agencies evaluated in this discipline, this dimension runs Strong 2 · Adequate 3 · Weak 11. The typical agency here scores Weak, and 5 of them score higher than this one.

What this dimension measures: Note any minimum spend and whether the model licence continues after the engagement ends.

Scores high — Published fees, ranges, or a stated engagement structure with numbers score 4-5.

Scores low — A described structure without numbers scores 2-3. Bespoke-only with no anchor scores 1-2.

What we found — Measured absence across the homepage, services index, MMM, incrementality, measurement, data infrastructure, media planning, methodology and contact pages: no fee, no range, no minimum engagement, no retainer structure, not even a stated engagement shape with numbers attached. Every commercial path is a 'free measurement assessment', 'free MMM consultation', 'free incrementality assessment' or a 30-minute growth call. The Clutch profile lists minimum project size as 'Confidential' and no hourly rate. A buyer cannot form any budget expectation before entering a sales conversation, and nothing states whether a model licence or the model itself survives the end of the engagement. That is what the low band describes, which is why it scored Weak.

On the record — “No pricing, fee range, minimum engagement or retainer structure appears on the homepage, services index, MMM, incrementality, measurement, data infrastructure, media planning, methodology or contact pages; Clutch lists minimum project size as 'Confidential'.” clutch.co ↗

On the record — “The homepage states 'We don't run your media; we make sure the strategy and measurement behind it can survive the boardroom.'” fusepointinsights.com ↗

retainer — a fixed monthly fee regardless of hours or output — predictable, but worth tying to a defined scope.

Evidence: inferred — our reading of indirect evidence, not a documented fact.

Stronger here: Nepa scores Strong on the same dimension.

Weak — 2 of 5 on this rubric’s scale, from Poor (1) to Excellent (5). This dimension carries 5% of the total score.

Benchmark — across the 16 other agencies evaluated in this discipline, this dimension runs Adequate 6 · Weak 10. The typical agency here scores Weak, and 6 of them score higher than this one.

What this dimension measures: Independent, verified reviews or checkable references. Low weight deliberately: measurement work sells through procurement and referral, so a thin public review footprint is normal and the method evidence above matters far more.

Scores low — A substantial verified base scores 4-5; a handful scores 2-3; none located scores 2.

What we found — One independent review located and read first-hand: a Clutch profile showing 5.0 overall from a single review, by the CEO of Bullibone, covering customer personas, surveys, media planning/buying and incrementality testing over January-April 2025, with 5.0 across quality, schedule, cost and willingness to refer and claimed improvements to Amazon and DTC ROAS. No G2, Trustpilot or Google review base was located. Note that the Clutch profile's firmographics belong to the parent, not the sub-brand: founded 2012 and 1,000-9,999 employees, where fusepoint itself launched publicly in January 2025. A single review is a real but very thin independent footprint.

On the record — “Fusepoint was launched publicly by Power Digital on 30 January 2025.” cmswire.com ↗

DTC — direct-to-consumer: brands selling on their own site rather than through retailers.

ROAS — return on ad spend: revenue per dollar of advertising. Platform-reported ROAS overstates; independently measured ROAS is the honest version.

Evidence: verified — checked against a named source you can open; the links under Sources below are where to check it yourself.