Our Predictions, Dated & Scored

Innovation Vista Predictions

Dated, Sourced, Scored

The list below includes every prediction Innovation Vista has published which did not qualify as “majority opinion” and/or which predicted a major impact on dynamics which had been stable for over a decade, as of the date of the prediction.

Each entry in the ledger lists the date the prediction was made, the date it resolves, the source that decides it, and its current status. The thesis from which these predictions flow is summarized on its own page. Entries are never deleted; misses count the same as hits in the denominator.

How the ledger works

Every entry below is dated and linked to its first publication. Each states the claim, the date it resolves, the source that decides it, and a confidence level: High, Moderate, or Contrarian, the last meaning that fewer than half of the experts we could find would have agreed with us at the time. Where a consensus existed when we wrote, we snapshot it and link the snapshot. Entries are never deleted or reworded after posting; a correction is a new entry that cites the old one. Conditional entries are labeled as conditional and are void, not scored, if their condition fails. A miss on a Contrarian call is calibration. A miss on a High-confidence call is a lesson, and we will write it up as one.

Our Prediction 2 · TRUE · Cyber underwriters ask about post-quantum readiness by late 2026

“By late 2026, we expect underwriters to begin asking whether critical infrastructure has been assessed for PQC migration.” Migration to PQC will not be required for a few more years, but we believe Cyber carriers will want to gather this data and will look to incentivize companies to prioritize this upgrade.

Active Predictions:

Prediction 3 · Vendors eventually give away what companies are building

Most companies are building AI in the wrong places, and “their software vendors are going to hand them for free within the next 18 to 24 months.” Custom AI in those zones is “rented advantage, not owned advantage.”

  • Predicted December 2025 in “Brighter Headlights: Seeing Where SaaS AI is Headed So You Can Build Elsewhere”.
  • Majority view when predicted: Vendors would build AI into the software companies already owned, but would charge for it. Gartner said the generative AI features spreading through existing software cost more, and forecast 15.2% growth in software spending for 2026.
  • Resolves December 31, 2027; decided by vendor release notes and price lists showing the post’s named capability categories bundled at no extra charge. Confidence: Moderate. Status: Open.

Prediction 4 · The Great Repricing shrinks the Labor market

“US white-collar and professional-services payrolls will remain below their pre-contraction peak through the end of 2027 due to AI disruption, and the college wage premium will narrow further.”

  • Predicted March 2026 in “The Great Repricing“.
  • Majority view when predicted: The white-collar slowdown was not mainly about AI. The Yale Budget Lab found no discernible broad disruption from AI, Fed Chair Powell called AI a factor but not the main one, and the New York Fed put the college wage premium near its all-time high. Professional and business services payrolls were already below their May 2023 peak.
  • Resolves December 31, 2027; decided by the final BLS payroll series for 2027 and the Federal Reserve Bank of New York’s college wage-premium series. Confidence: High. Status: Open

Prediction 10 · Pricing units switch from per-seat and T&M

“By the end of 2027, at least three major software or IT-services providers will have publicly moved offerings from time-and-materials or per-seat pricing to fixed-bid or outcome pricing anchored at or above pre-AI price levels, while disclosing material AI-driven reductions in their own delivery cost.”

  • Predicted in May 2026 in “Vendor Partnerships in the Era of AI Arbitrage“.
  • Majority view when predicted: Outcome pricing was coming, but at lower prices. HFS Research found only 16% of enterprise leaders expected to still use headcount-based contracts within two years, and analysts assumed AI savings would pass to clients: Kotak projected 16% gross revenue deflation on existing IT-services spending over three years, and PwC said it had already cut some prices.
  • Resolves December 31, 2027; decided by earnings calls, published price lists, and trade press. Confidence: High. Status: Open.

Prediction 13 · Digital workers get a price tag

“By the end of 2027, at least five of the twenty largest enterprise software vendors will sell per-agent or outcome-priced digital-worker offerings as standard products, and supervisory capacity will surface as the reported bottleneck in major CIO surveys.”

  • Predicted June 2026 in “Last Year Your Competitors Subscribed to AI Tools · Now They’re Hiring Them“.
  • Majority view when predicted: Analysts already expected pricing to move off the seat; IDC forecast that pure per-seat pricing would be obsolete by 2028, and Salesforce had priced Agentforce per conversation and per action in 2025. But the bottleneck was expected elsewhere: IT leaders surveyed by Gartner pointed to governance and security gaps and did not cite supervisory capacity.
  • Resolves December 31, 2027; via published price lists and CIO surveys. Confidence: High on the first clause, Moderate on the second. Status: Open.

Prediction 16 · Buyers get agents before most sellers build docks

“Through at least the end of 2027, deployed buyer-side procurement agents will outnumber seller-side agent interfaces in the mid-market by an order of magnitude.”

  • Made August 2026 in “Your Next Customer Won’t Be Human · MCP and the Rise of Agent Docks“.
  • Majority view when predicted: Buyers were already expected to move first; Deloitte found nearly 40% of B2B buyers using agentic AI in purchasing, against 24% of suppliers using agents in sales. We found no forecast of a gap anywhere near ten to one.
  • Resolves December 31, 2027; decided by published procurement-AI adoption data set against a dock census, which we may have to run ourselves and will publish if we do. Confidence: High. Status: Open.

Prediction 17 · Mid-market tech budgets stop rolling forward

“By the fall 2027 planning cycle, most mid-market technology budgets will carry governance and data as standing rows and software as a committed-plus-variable pair. The budgets that do not will belong to the companies whose AI spending rose fastest and whose EBITDA did not.”

  • Predicted September 2026 in “If Your Tech Budget Looks Like Last Year’s, You Missed the Turn”.
  • Majority view when predicted: Analysts agreed on the direction but found practice lagging. IDC called AI governance a strategic budget priority, while Gartner’s 2027 planning research found 73% of enterprises had no plans to set rules for who owns technology costs, with IT budgets projected to rise an average of just 3.7% in 2027.
  • Resolves December 31, 2027; decided by next year’s Innovation Vista Report, as the post says, cross-checked against published IT budget benchmarks such as Gartner IT Key Metrics. Confidence: Moderate. Status: Open.

Prediction 5 · The Great Repricing drives national policy

“By the end of 2028, a major federal income-support expansion explicitly responding to AI displacement will be enacted or formally adopted as a platform position by the leadership of a major party.”

  • Predicted March 2026 in “The Great Repricing“.
  • Majority view when predicted: Both parties favored retraining over income support. The White House AI Action Plan answered AI displacement with rapid retraining, Democrats’ emerging playbook leaned on retraining and apprenticeships with its details unsettled, and safety-net ideas such as Sen. Mark Kelly’s call for more generous unemployment insurance came from individual lawmakers, without party leadership support.
  • Resolves December 31, 2028; decided by the legislative record and published party platforms. Confidence: High. Status: Open.

Prediction 7 · Deflation follows, once a safety net exists

“The deflationary signal appears in the same series during 2028, as redesigned contracts catch up with delivery costs.” 

  • Predicted March 2026 in “The Great Repricing“.
  • Majority view when predicted: Prices would keep rising through 2028. CBO projected 2.3% consumer inflation that year, and Fed officials’ December 2025 projections had a median of 2.0% PCE inflation; we found no mainstream forecast of falling prices in any broad price index.
  • Resolves with the December 2028 PPI release; decided by the relevant BLS Producer Price Index series. Conditional on prediction #5; if #5 misses, this entry is void. Confidence: Moderate. Status: Open

Prediction 8 · The old consulting business model fails

“By the end of 2028, at least two of the Big Four will report advisory-arm revenue declines in consecutive fiscal years, or aggregate advisory headcount more than ten percent below peak, while expert-direct and fractional models gain share.”

  • Predicted March 2026 in “Why the Old Big 4 Consulting Model is Dying“.
  • Majority view when predicted: Consulting would keep growing. Source Global Research expected technology consulting growth to reach 7% in 2026, and Deloitte, PwC, EY, and KPMG all grew global consulting or advisory revenue in fiscal 2025. The clearest declines were in national arms, such as the advisory business of KPMG’s UK and Swiss group, down 3% in fiscal 2025.
  • Resolves with the firms’ fiscal 2028 global revenue reports; decided by those reports. Confidence: Moderate. Status: Open.

Prediction 11 · Waiting on vendors does not pay this time

The claim: “Companies in the top quartile of the 2026 Innovation Vista AI Maturity Index will outperform bottom-quartile peers on revenue growth AND margin by the 2028 edition, and the gap will widen rather than close.”

  • Predicted June 2026 in “The Last Tech Wave Rewarded Waiting for Vendors · This One Won’t“.
  • Majority view when predicted: Opinion was split. BCG already reported AI leaders beating laggards on revenue growth and margins, with the gap widening, while MIT’s widely cited study found most generative AI pilots delivering little or no measurable profit impact, and vendor-bought tools succeeding more often than internal builds.
  • Resolves with the 2028 Index. Confidence: Moderate. Status: Pending prerequisite.

Prediction 15 · Agent docks will become a standard commercial surface

“By mid-2029, an agent-facing, transactable interface will be as expected of a mid-market B2B seller as a website was by 2005. By the end of 2027, at least two major ERP or B2B marketplace ecosystems will formally support autonomous counterparty agents completing discovery-to-order without a human starting the session.”

  • Made August 2026 in “Your Next Customer Won’t Be Human · MCP and the Rise of Agent Docks“.
  • Majority view when predicted: Mixed opinions on magnitude, not direction; Gartner did forecast that AI agents would intermediate 90% of B2B buying by 2028, so we’re aligning with them. Others expected the plumbing to come later: Deloitte placed fully autonomous agent-to-agent commerce at the final stage of B2B commerce maturity, and platforms such as Alibaba’s still required a person to approve each purchase order.
  • Resolves December 31, 2027 for the ecosystem clause and June 30, 2029 for the expectation clause; decided by vendor documentation and public announcements. Confidence: High. Status: Open.

Prediction 6 · The Great Repricing, if it comes, drives inflation

“If prediction #5 proves true, its financing will be inflationary, despite the deflationary pressure AI exerts generally: the federal deficit stays above 2.5% of GDP and CPI reaccelerates above 4% within eighteen months of enactment.”

  • Predicted March 2026 in “The Great Repricing“, in the lineage of Lyn Alden’s work on debasement, credited there.
  • Majority view when predicted: The deficit clause matched consensus; CBO’s baseline already showed deficits between 5.6% and 6.7% of GDP every year through 2036. The inflation clause did not: the same baseline, which assumes no new income-support program, had CPI inflation easing to 2.4% in 2027 and 2.3% after.
  • Resolves by June 30, 2030; decided by Congressional Budget Office deficit figures and the BLS Consumer Price Index. Conditional on prediction #5; if #5 misses, this entry is void. Confidence: Moderate, conditional. Status: Open.

Prediction 12 · Mid-level talent becomes scarce and expensive

“A finished mid-level professional hired laterally in 2028-2030, in a market where every competitor is bidding for the same shrinking pool, will plausibly cost $150,000 or more.”

  • Predicted June 2026 in “Talent Arbitrage · Leveraging the Professionals AI Just Repriced”.
  • Majority view when predicted: AI would cool hiring demand rather than bid up pay. About four in ten employers planned to cut or replace roles with AI (WEF, Korn Ferry), and 2026 raises were expected to hold near 3.5% (Conference Board). Some, Korn Ferry among them, warned that a drying entry-level pipeline could lead to a long-term leadership crisis; we found none forecasting a pay premium for mid-level hires.
  • Resolves December 31, 2030; decided by BLS Occupational Employment and Wage Statistics for the professional occupations the post names, and published lateral-hire compensation surveys. Confidence: Contrarian. Status: Open.

Prediction 14 · Early headcount cutters produce failure stories, then a promotion gap

“The [AI] faith-based job cutters are going to generate spectacular failure stories over the next two years,” and “the bench gap surfaces in three to five years when there is no one ready to promote.”

  • Predicted July 2026 in “The AI Rorschach Test & the High Cost of Presumption”.
  • Majority view when predicted: Executives were cutting on AI’s promise, but analysts were already skeptical. Forrester expected half of AI-attributed layoffs to be quietly reversed, Gartner expected half of the organizations planning AI-driven customer-service cuts to drop them by 2027, and Korn Ferry warned of a thinning leadership pipeline. Our call differs mainly in expecting the reversals to be public.
  • Resolves July 31, 2028 for the failure-stories clause and July 31, 2031 for the bench-gap clause; decided by reported reversals of AI-driven layoffs (rehiring announcements, written-down restructuring charges) in the business press and company filings. Confidence: Contrarian. Status: Open.

Prediction 1 · HR will eventually fold into IT; the CIO will reach C-suite parity

“Traditional Human Resources functions will be entirely subsumed within IT by 2035.” The same post has the CIO at equal stature with the CFO and COO by 2035.

  • Predicted August 2025 in “The IT Org Chart of the Future”.
  • Majority view when predicted: IT leaders expected HR and IT to merge, mostly not under the CIO. In a 2025 Nexthink poll, most expected a complete merger within five years, but the largest group, 40%, wanted neither the CIO nor the head of HR to lead it. CIO stature looked steady, with 58% of CIOs reporting to the CEO, flat on the year before.
  • Resolves December 31, 2035; decided by published org-chart and reporting-line data (Fortune 500 proxy disclosures and the annual CIO surveys from Deloitte, Gartner and Foundry). Confidence: Contrarian. Status: Open.

Prediction 9 · Laggards will be sold for cheap

“Within ten years of publication, AI-laggard companies in at least three mid-market sectors will trade at distress: median acquisition multiples for bottom-quartile AI-maturity firms below 20% of the sector-median revenue multiple.”

  • Made May 2026 in Within a Decade You’ll Be Able to Buy Competitors for Pennies.
  • Majority view when predicted: AI was becoming a diligence question, not a cause of fire sales. Acquirers were scrutinizing AI’s risks and opportunities in targets, and one in five strategic dealmakers had walked away from a deal over AI’s expected impact on the target’s business, but we found no forecast of distressed pricing for AI laggards outside software.
  • Resolves May 2036; decided by PitchBook or Capital IQ transaction data, with the AI-maturity quartile defined by the Index. Confidence: Moderate. Status: Open.

Candidate entries, not yet posted

Our AI Dividend Map is refreshed every year, which makes it a natural source of dated claims: which sectors will cross the third gate and move from Slow Melt or Rollup into Rising Bar or Falling Floor by the 2028 edition, and which will not.  Nothing enters the ledger until it has been published as a claim somewhere else first.

The scorecard

So far only our prediction #2 has been resolved (True). A ledger a single prediction resolved is nearly all promise, and we know it; the near-term entries were chosen so the scoreboard fills in within eighteen months, not ten years.

We’ve laid out our overarching thesis that these predictions test, and of course other articles & pages provide deeper context on specific aspects of the AI revolution through which we are guiding mid-market organizations.

Thought Leadership from our team
Latest !nsights