Customer Analytics Service Providers

The latest edition of this category is published under a shorter name: The Forrester Wave: Customer Analytics Services, Q2 2025, which is what Customer Analytics Service Providers became. Published in May 2025 and authored by Zeid Khater, it scored ten providers against twenty six criteria. The dropped word is the first finding. This market no longer sells providers. It sells services.

The second finding is Tredence. Forrester introduced the firm in the 2023 edition as "the new kid on the block." In the 2025 edition it is the number one Leader, with the highest possible scores in fifteen criteria, and the criteria it won on describe a company automating itself. The new kid became the category's reference point in two years, which is either a tribute to the firm or a warning about the category's turnover rate. Both readings are supported.

The newest scorecard: The Forrester Wave: Customer Analytics Services, Q2 2025

The evaluation's public placements run two Leaders and a recognized specialist; the rest of the ten sit behind the paywall.

Tredence took the top spot, its third consecutive Leader recognition across the 2021, 2023, and 2025 editions. The fifteen maximum scores span data sources and types, customer data models, personalization, customer retention, ModelOps, insights delivery and activation, measurement and metrics, vision, partner ecosystem, and pricing flexibility. Forrester cited an "agents-as-a-service" automated stack, a "Foundry & Factory" delivery model, and end-to-end customer experience transformation. The reference detail matters: every reference customer praised the firm's data science and thought leadership, and in a services Wave, the references are the score.

ZS joined as a Leader with ten maximum scores, including partner ecosystem, pricing flexibility, and global delivery strategy. Forrester's line is the one to quote in a buying meeting: "ZS's personalization capabilities are exceptional." The fit verdict runs to large enterprises in healthcare, pharma, medtech, quick-service restaurants, airlines, and retail. Two product names carry the citation: Personalize.AI and a Max.AI agentic platform, wrapped in what the firm calls an "Amazon Prime of analytics" approach, speed and quality of service as the product.

Evalueserve earned the highest possible score in functional competencies and experience, with a maximum in measurement and metrics. Forrester's phrasing, "Evalueserve shows its power through functional competencies," is the quietest way to say what the 2025 scorecard values: depth of doing, not breadth of claiming.

What the 2025 criteria are actually scoring

The announcement blog names three capabilities the edition was built around, and the third one is the story of the whole category.

Depth of insights. Churn and propensity models, lifetime value, segmentation: commoditized. Differentiation now runs through richer contextual understanding, graph databases, vector embeddings. The math became table stakes, so the scorecard moved to what the math sits on.

Impact and actionability. Advanced decision optimization, real-time feedback, explainable AI, dashboards and data hubs that let clients watch value accrue and ROI calculate itself. The scorecard now measures whether the work changed a decision, not whether the deliverable was delivered.

Preparation for agentic AI. The blog's own words: leading providers are building automation pipelines and reducing reliance on traditional consulting. Read that sentence again. This is a Wave for consulting firms, and the criteria reward firms for making consulting less necessary.

The scorecard has started scoring the firms' own obsolescence. That is not a cost-cutting line from the analyst. It is a product line for the winners. Tredence's factory model and agents-as-a-service stack, ZS's Amazon Prime of analytics: these are firms industrializing the labor they used to sell by the hour.

The new kid's scorecard, three editions later

The lineage runs Q2 2019, twelve providers; Q3 2021, eleven providers against thirty criteria; Q2 2023, eleven providers against twenty nine criteria; and the renamed Q2 2025 edition, ten against twenty six. The field has been shrinking while the criteria have been climbing toward outcomes, which is what maturation looks like from inside.

Brandon Purcell authored the 2019, 2021, and 2023 editions, and his 2021 verdict is the category's founding sentence: vendors can no longer differentiate on advanced analytics capabilities alone, but on how well they help clients use model outputs to create business impact. The model was the invoice. The moment it changed a decision was the product. Every edition since has scored that moment with more precision, and the 2025 edition changed hands to Khater, a new author for a renamed category.

The 2023 edition is where the current leader was minted. Purcell called Tredence the new kid on the block and gave it the highest current offering score. Fractal led alongside it, credited for bridging the insights-to-action gap with design and engineering. The Strong Performer band held the names worth remembering: Evalueserve, with maximums in insights delivery and data literacy training; TTEC, told it "thrusts the tip of the spear"; and ZS, in its first evaluation, flagged for Personalize.AI and generative emails. The 2023 field was drawn from a Landscape of thirty eight candidates, which gives you the depth of the bench behind a ten-provider Wave.

What the report cannot know yet

Here is the honest limitation, and it is built into the method.

A services Wave is scored on references, and references are memories of the past. All of Tredence's references praised its data science. That is meaningful and also inherently backward-looking. The 2025 edition is simultaneously scoring firms' current consulting work and rewarding their pivot away from that work. The automation pipelines, the agentic stacks, the factory models: these are scored mostly on promise, because the delivery model they replace is the one the references describe. A buyer in 2026 is signing a contract with a firm whose scorecard describes, in part, a way of working that is being dismantled.

The second limitation is the author's. A renamed category under a new analyst is a definition in transition. Khater's criteria moved the weight from Purcell's insights-to-action framing toward depth, impact, and agentic preparation. Both framings are legitimate, and they are not the same Wave. When the author changes, the definition changes with them, and scores from different editions compare less cleanly than the vendors would like you to believe.

The third is the oldest one in services. The Wave scores the firm's best practice. Your engagement gets the team the deal affords. No scorecard on earth can tell you which team that will be.

Three questions before you sign the SOW

Which delivery model is priced into my statement of work: a factory or a team? If the firm's winning score is built on a Foundry & Factory model or an agents-as-a-service stack, your contract should specify which of those assets your engagement actually runs on. The score was earned on the factory. The invoice should not quietly revert to hourly labor.

What happens to my engagement when the firm automates itself? Ask for the agentic roadmap applied to your program specifically: which workflows get automated in which quarter, who owns the outputs, and whether the price comes down when the labor does. Forrester is scoring firms on reducing reliance on consulting. Make sure your deal is on the right side of that reduction.

Who are the references for my industry and my scale, and what did they keep after the engagement ended? The scorecard's references are aggregates. Yours should be specific. For European buyers, add the data question on the record: where the data lives, who processes it, and what happens to it when the engagement closes.

The category has changed what it sells twice: first from models to impact, now from labor to automation. The 2025 scorecard is the first to be written after that second change began. Read it as the market's answer to its own founding question, and sign nothing until a firm shows you which half of its scorecard your deal is built on.

The same labor-to-automation shift is playing out on the platform side of this market: see Customer Analytics Technologies, where the 2026 Wave scores decision engines rather than the analytics tools it scored just two years earlier.

Forrester does not endorse any vendor named here, and tier placement should not be read as a recommendation to buy.