Why we left Crossover for Devlyn after 6 months

This is a real story from a CIO at a $300M enterprise software company. Names anonymised; calendar and numbers exact, as described in a CIO peer call last quarter. The pattern is not specific to Crossover — it shows up across every productivity-tracking-led model when the work needs senior judgement rather than tracked output — but Crossover was his vendor.

The opening: Crossover’s low-rate model fit the budget envelope

The CIO had a $300M company with a $4M annual engineering budget that the CFO wanted held flat year-over-year. His in-house team was forty-two engineers; he needed to add eight more contributors for a legacy-modernisation program without expanding the budget. Crossover’s positioning — global low-rate hiring with strict productivity tracking via WorkSmart and similar tooling — fit the budget math. Hourly rates in the $25–$40/hour range against $90+/hour for US contractors meant 8 more engineers were affordable inside the existing budget.

He briefed Crossover. Within five weeks he had eight engineers placed across legacy modernisation lanes. Combined burn was around $46,000/month. The math looked clean on paper.

Months one through three: the productivity-tracking culture cost

The Crossover engineers were technically capable. Crossover’s hiring funnel is rigorous and the engineers placed at the CIO’s company were not low-skill — the rate was lower because Crossover operates in markets with lower local cost-of-living, not because the engineers were less qualified.

The structural problem was the engagement culture. Crossover’s model includes mandatory productivity tracking — keystroke and screenshot monitoring through WorkSmart — and metrics-driven performance evaluation. The CIO’s in-house engineering culture was the opposite: senior-judgement-led, autonomy-driven, outcome-measured rather than activity-tracked. Putting eight tracked-productivity engineers into a forty-two-engineer judgement-led culture created friction the CIO had not modelled.

The friction showed up in three ways:

1. Hesitation to ask senior in-house engineers questions — the Crossover engineers were tracked on activity, so “five minutes asking a senior” cost their measured productivity. They opted to grind through unfamiliar code instead.
2. Risk-averse PR patterns — tracked engineers tended to ship smaller, lower-risk changes that hit activity quotas rather than the larger refactors the legacy-modernisation work needed.
3. Engagement asymmetry — the in-house team felt awkward managing colleagues whose every keystroke was watched.

Months four and five: cost saved, calendar lost

By month four the CIO ran the math the second way. The legacy-modernisation program was 35% complete against a target of 50%. Velocity per Crossover engineer was roughly 0.6× the in-house team’s pace on similar refactors — which is what the productivity-tracking culture cost in practice. He was paying $46,000/month for eight engineers shipping at roughly the equivalent of five in-house engineers’ velocity. The “budget savings” was real on paper and largely fictional on calendar.

In month five the board flagged the legacy-modernisation milestone as the next-quarter risk. The CIO’s CFO was happy with the budget line. The COO was unhappy with the milestone slippage. Trade-offs that look balanced at month one rarely stay balanced at month six.

By month six:

• $276,000 cumulative Crossover spend over six months.
• Eight engineers placed at lower rates than market.
• Velocity per Crossover engineer: 0.6× in-house pace.
• Effective output: ~5 engineers’ worth from 8 paid seats.
• Legacy modernisation: 50% complete against a target of 70%.
• Engineering culture friction reported in two consecutive employee pulse surveys.

The Devlyn discovery call

The CIO booked a 30-minute Devlyn discovery call. Brought the legacy-modernisation scope, the Crossover burn pattern, and the cultural-friction data from his employee pulse surveys. The discovery call ended with a recommended pod composition: a 4-engineer pod plus shared DevOps and QA leads, dedicated PM line, AI-augmented engineering as the workflow standard, retainer of $19,400/month.

Against the Crossover burn at $46,000, the math was: lower burn for fewer-but-faster engineers, AI-augmented workflow producing 4× compounding velocity vs Crossover’s 0.6× tracked-productivity pace, and an autonomy-driven culture that matched the in-house engineering team’s norms.

The total-output comparison: 8 Crossover engineers at 0.6× = 4.8 effective engineer-equivalents. 4 Devlyn engineers at 4× compounding = 16 effective engineer-equivalents. Same calendar, less than half the burn, three times the effective output.

Devlyn proposed a 3-day free trial against a real legacy-modernisation refactor. The trial ran Friday through Monday. The pod returned a working refactor that the Crossover team had been working through for eleven days. The 3-day output was AI-augmented workflow operating as advertised, with senior validation choosing the larger-scope refactor that tracked-productivity engineers had been avoiding.

He hired Tuesday. Pod was in his Slack and repos within 24 hours.

Want to see the model against your actual roadmap? Book a 30-minute Devlyn discovery call → — no contracts, no commitment.

What changed: months seven through nine

The CIO offered the Crossover engineers a transition window. Three of the eight transitioned to autonomy-driven engagements with other vendors at higher rates; the other five wound down on existing work. The bulk of legacy-modernisation work moved to the Devlyn pod.

By month eight the legacy-modernisation program was 80% complete; by month nine the target milestone was shipped. The employee pulse survey showed the culture-friction issue resolved by month eight. The CFO was unhappy about the higher per-engineer rate and happy about the lower total burn — net result was a net budget reduction of $26,600/month against the prior Crossover engagement.

The structural reason was that AI-augmented engineering compounds when the engineers can exercise senior judgement. Productivity-tracking cultures dampen senior judgement. The two are structurally incompatible at the workflow level.

The honest reckoning: when Crossover was still right

Crossover was not the wrong vendor for the wrong company. Crossover’s model fits organisations whose engineering culture is itself productivity-tracking-led — large enterprises running at scale with metric-driven performance evaluation built into the in-house norms. The cultural match would not have produced the friction the CIO experienced.

The vendor became wrong because the cultural mismatch was structural. The CIO’s $300M company ran a senior-judgement culture; placing eight tracked-productivity engineers into it produced friction that compounded against velocity. The lower hourly rate did not survive the velocity dampening.

The CIOs who get this right in 2026 match engagement culture to in-house culture. The CIOs who get it wrong assume the rate-card savings will dominate the velocity dynamics and end up at month six with budget line happy and calendar line slipping.

What the numbers looked like, side by side

Lever	Crossover months 1–6	Devlyn months 7–9
Engagement model	Eight contractors with productivity tracking	Four-engineer pod with senior judgement
Monthly burn	$46,000	$19,400
Velocity per engineer (vs in-house pace)	0.6× (tracked-productivity dampening)	4× (AI-augmented compounding)
Effective engineer-equivalents	4.8 from 8 paid seats	16 from 4 paid seats
Legacy modernisation progress	50% at month 6	Milestone shipped at month 9
Culture friction (pulse survey)	Two consecutive negative quarters	Resolved by month 8
Budget line	Held flat	Reduced $26,600/month

The line that mattered most was effective engineer-equivalents. Headcount and rate cards are headline numbers; effective output is the number that makes the milestone. Crossover’s 8-seat headcount delivered ~5 effective engineers; Devlyn’s 4-seat pod delivered 16 effective engineers. The math compounds across a quarter.

What he tells other enterprise CIOs now

I asked the CIO what he tells his peers. His answer:

“Crossover’s lower rate is real on paper. The cultural friction with senior-judgement in-house teams is also real and the velocity dampening at 0.6× erases the rate-card savings. The right mental model is effective output per dollar, not rate per hour. Pods at higher rate per hour producing 4× velocity beat tracked-productivity engineers at lower rate producing 0.6×, by three to four times on the math. The CFO needs to see effective output not rate cards.”

He still recommends Crossover for organisations with productivity-tracking-led in-house cultures. The framing is tracked-productivity-fit-mode versus senior-judgement-fit-mode.

What to do if you are at month three or four with Crossover

If you are reading this from inside a Crossover engagement that produces friction with the in-house team and slipping milestones — the pattern is structural. The diagnostic questions are:

1. Does your in-house engineering culture run on productivity tracking or senior judgement? Mismatch produces the velocity dampening.
2. Have you measured velocity per Crossover engineer against in-house pace? If under 0.7× you are paying for friction.
3. Are pulse surveys flagging culture issues? Cultural friction is a leading indicator.
4. What is the effective engineer-equivalent calculation? Headcount × velocity multiplier. Compare across vendors on this metric, not on rate cards.

Cheapest move from month four is parallel evaluation. Keep the Crossover engagement running. Open a 30-minute Devlyn discovery call. Run a 3-day free trial against legacy-modernisation work. Decide based on effective output per dollar, not rate per hour.

If you are running a $50M–$500M enterprise IT organisation and the budget envelope plus cultural fit plus velocity multiplier all matter, the productivity-tracking model compounds against you when the in-house culture runs on senior judgement. Book a 30-minute Devlyn discovery call → — no contracts, no commitment.