Research brief: dashboards for SMBs — what's worth showing, and when an embedded one earns its keep (June 2026)

Status: Synthesised June 2026. Fourth brief in the SMB-advice quartet with Research brief: customer-facing calculators & tools for SMBs — the honest case (June 2026), Research brief: live data and data-driven tools for SMBs — when it's an edge, when it's overkill (June 2026), and Research brief: client portals for SMBs — the honest case (June 2026).

TL;DR

A dashboard is the surfacing layer that moves data from "queryable but ignored" to "informs daily decisions" — if it is actually used. The honest evidence says most of that promised value never lands:

• Independent analyst data puts BI/dashboard adoption at only ~25-30% of employees (BARC / Eckerson Group (March 2022, 214 D&A leaders surveyed Nov-Dec 2021) — BI/analytics adoption at 25% of employees on average; "stuck around 20% for many years" per Eckerson, Gartner — "pervasive business intelligence remains elusive, with BI and analytics adoption at about 30% of all employees" (doc 3753469, Howson & Sallam)) and has for over a decade.
• SMB-specific BI dashboard adoption is far lower — Software Advice put it at 5% of 243 SMBs (Software Advice (online survey of 243 US business owners/managers) — 5% of SMBs report using a BI dashboard to analyze data; Excel/Google Sheets dominate at 65%).
• The viral "60-70% of dashboards go unused (Gartner)" statistic is vendor folklore — could not be traced to any named Gartner report (CORRECTION: the viral "60-70% (or 60-80%) of dashboards go unused (Gartner)" stat cannot be traced to any named Gartner report — vendor folklore). Use BARC ~25% / Gartner ~30% adoption figures instead.
• The "72% of users regularly abandon dashboards for spreadsheets" stat is Luzmo's vendor-run survey of ~200 SaaS/product leaders ("72% of users regularly abandon dashboards for spreadsheets" — Luzmo vendor-run State of Dashboards 2025 survey of ~200 SaaS/product leaders; "Dashboards: Dead, Dying or Evolving?"), not independent research. Same with Logi Analytics 45% BI usage ("Only 45% of people with access to BI tools actually use them" — traces to Logi Analytics (BI vendor) report, not independent research).

The two decision tracks

INTERNAL dashboards are worth it when there is a recurring, owned decision tied to a metric with a threshold — cash flow, DSO, job-cost variance, utilization, at-risk customers. They are not worth it as a "single source of truth" wall of vanity metrics nobody opens after launch.

• Tied to a named owner + named decision + named threshold (R1 — Before building any dashboard, qualify every proposed metric: named owner + named decision + action threshold; if it fails the "what would I do if it moved?" test, drop it, The "what would I do if this number doubled or halved tomorrow?" test — if no action follows, it's vanity).
• Default to the cheapest BI tool that fits the stack: M365 → Power BI Pro (Power BI pricing (April 2025): Pro $14/user/mo, Premium Per User $24/user/mo (both raised from $10/$20); Fabric F2 ~$262/mo to F64 ~$5,000/mo (viewers no per-user); Embedded A1 ~$735/mo); Google/marketing data → free Looker Studio (Looker Studio: free tier widely used; Looker Studio Pro $9/user/Google-Cloud-PROJECT/month (per-project detail materially inflates agency costs); separate Looker (Google Cloud core) $35K-60K+/yr); SQL-comfortable / future-embedding → Metabase (Metabase pricing (2026): Open Source free (AGPL v3 self-host); Cloud Starter $100/mo (5 users included, +$6/user); Pro $575/mo (10 users, +$12/user; white-label + embedded + RLS + SSO); Enterprise from ~$20K/yr). (R2 — For internal dashboards, default to the cheapest tool that fits the existing stack: Power BI Pro for M365 shops, free Looker Studio for Google/marketing data, Metabase for SQL-comfortable / future-embedding).
• Budget ~20-30% of build effort annually for maintenance (R4 — Budget 20-30% of build effort annually for maintenance from day one; assign a metric-definitions owner; audit quarterly and archive any dashboard unopened in 30+ days).

CUSTOMER-FACING dashboards (especially embedded in the website) earn their keep in a narrow band: recurring clients + frequently-updating data-rich results + retention/differentiation lever + support deflection (Customer-facing embedded dashboard earns its keep in a NARROW BAND: recurring clients + frequently-updating data-rich results + retention/differentiation lever + support deflection). Most SMBs delivering periodic results to a handful of clients should send a scheduled PDF or shared reporting link — see AgencyAnalytics benchmark — a majority of agencies still send static reports to clients while relying on live dashboards internally; live unfinal numbers invite client misinterpretation and R3 — For customer-facing reporting, default to the lightest delivery first: scheduled PDF or shared reporting link; embed only when the four-condition narrow band holds.

• 20+ client tenants or analytics-as-product? Evaluate an embedded-analytics platform (Luzmo / Explo / Cube / Embeddable — Embedded-analytics platforms (2026): Luzmo $995-3,100/mo embedded plans; Explo $795/Growth, $2,195/Pro; Cube Core open-source + Cube Cloud consumption; Embeddable still in beta) before a custom build.
• Custom-embedded is justified only when analytics is core differentiation AND ≥1-2 engineers will own it permanently. 3-year TCO ~$300K-$630K with ~20-30% annual maintenance (Custom-embedded 3-year TCO converges on $300K-$630K (multiple, mostly vendor sources) with 20-30% annual maintenance; hidden costs are multi-tenancy/RLS, performance at scale, perpetual maintenance).

Source-incentive meta-finding

The dashboard literature is overwhelmingly vendor-published (Microsoft, Salesforce/Tableau, Google, Metabase, Luzmo, Explo, Cube, Logi Analytics, Sisense). Retention / churn / ROI claims (e.g., "31% retention increase," "27% churn drop") are vendor-sourced — treat as marketing, not evidence. See Caveats for the dashboards brief: pervasive BI/embedded-analytics vendor sourcing; the viral "60-70%" stat is folklore; SMB data thin; retention claims unproven.

The article

The publication-ready prose draft lives at [[article-when-is-a-dashboard-worth-it-for-your-business]] (Candid /writing/ candidate, SMB audience).