August 2026

Per-month text deep dive.

State of the Upwork · Redwater Revenue edition. Continued from work originally published by Aaron Melton at Ascend Automation Agency. Same taxonomy, same voice, same emoji legend — now with AI-tier tracking added on top.

Jump to

← Main Report (Upwork Automation Market Analysis)App-Comparison ↓High-Paying Guide ↓

Upwork Automation Market: Platform & Application Comparison

August 2026 Analysis

Report Date: September 2026 Data Period: June 2025 baseline; detailed series September 2025 – August 2026 (12 months shown) Total Jobs Analyzed: 2,391 (platform set, August 2026)

Executive Summary

August 2026's headline is quiet: the median hourly rate printed $32.50 and the platform mix barely moved. On the volume line, August is the first month since the SOTU handoff where a clean month-over-month comparison is genuinely hard to make — the platform scrape failed on two August days, those days were backfilled at month-end, and the backfill also re-tagged jobs across the whole month. So the raw stored total (2,391) is not directly comparable to July's 2,549. On the honest comparison — platform jobs per weekday over cleanly-captured days, measured before the backfill — August ran about 88 jobs/weekday against July's 97, a high-single-digit softening. That is a normal late-summer easing, not a contraction.

Zapier held at 1,070 jobs and n8n at 960, with Make.com at 421 — the three platforms sitting in their usual order, Zapier the volume leader, n8n the technical-depth option, Make the weakest. None of these are a monthly "leadership" story; measured from the June 2025 baseline all three are down roughly the same ~30%. GoHighLevel stayed near its highs at 600 jobs, the single most durable application in the dataset. The median rate at $32.50 is the second time in the series it has printed below $35 — the other being December 2025's $30 — and the mean fell to $36.91 from July's $41.41, its lowest read in the series.

The AI tier led on counts and fell on share: OpenAI/GPT (534 mentions) and Claude/Anthropic (507) lead the platform-set terms, with AI Agents at 363 and Voice AI at 184, but these are overlapping mention counts. The clean figure, the share of postings naming any AI term, fell a fourth straight month to 40.82% from 48.75% in May, and all ten categories lost share. Fixed-price share fell to 37.1%, a third straight monthly decline from 42.9% in May.

Key August Insight: This is a calm month wearing a noisy hat. Two things moved and both are measurement, not market — the lost-and-backfilled August days, and the re-tagging that inflated the stored August total. Strip those out and you get a market that softened a few points into late summer (matching Aaron's -2.2% July→August 2025) and sat on the fence on pay: the hourly postings split 462 at $30 or below against 462 at $35 or above, with four in the gap, so the $32.50 median is decided by a handful of listings. Fixed-price share and AI-term share both fell. Don't read a story into the volume line this month.

Platform Comparison: Twelve-Month Evolution

Major Automation Platforms

Platform	Sep	Oct	Nov	Dec	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Jul MoM	Since Jun '25
Zapier	2,233	2,132	1,662	1,753	1,767	1,721	1,827	1,785	1,021	972	1,223	1,070	📉 -12.5%	💥 -30.5%
n8n	2,099	2,024	1,729	1,663	1,535	1,403	1,465	1,252	805	815	1,055	960	📉 -9.0%	💥 -27.9%
Make.com	967	1,027	811	751	725	743	806	781	638	593	631	421	📉 -33.3%	💥 -30.7%
Power Automate	33	42	20	28	37	22	29	30	20	18	30	29	🟡 -3.3%	🟢 flat

Status Indicators:

🚀 Strong Growth (>10%)
🟢 Stable/Growth (0-10%)
🟡 Slight Decline (0-5%)
📉 Declining (>5%)
💥 Collapse (sustained >20% since June 2025 baseline)

A note on the MoM column: raw August counts are not directly comparable to July. July was measured live on the full hybrid basis; August lost two capture days (backfilled at month-end) and the backfill re-tagged ~127 jobs into the platform set across the month, inflating the stored August total ~6%. The MoM figures above use the stored counts and should be read as directional only. The capture-independent signals — median rate and platform share — are the real read.

Platform Analysis

Zapier (Volume Leader, Standing Structure)

Current: 1,070 jobs (43.7% of the three-platform total)
Trend: down since July, but the stored counts straddle the August backfill
Total: -30.5% since June 2025, in line with the other two platforms
Assessment: Zapier remains the volume leader as it has for the entire series. There is no leadership drama here — its ~30% drawdown from June matches n8n and Make almost exactly.
Recommendation: Primary platform for volume; the case rests on breadth of client demand, not on any monthly growth arrow.

n8n (Technical-Depth Option)

Current: 960 jobs (39.2% of the three-platform total)
Trend: n8n has oscillated in a 805–1,055 band across the SOTU era; August sits mid-band
Total: -27.9% since June 2025 — the same neighborhood as Zapier and Make
Assessment: The old "uniquely resilient n8n" thesis was a baseline artifact. On a June baseline, n8n declined about as much as everything else. What is real is that n8n remains the technical-depth / self-hosted option and shows up heavily in the AI-agent job descriptions.
Recommendation: Strong secondary or primary for technically-deep, AI-adjacent work; not because it's "winning," but because of the kind of client it attracts.

Make.com (Weakest of the Three)

Current: 421 jobs (17.2% of the three-platform total)
Trend: August's 421 is the lowest platform read in the series, but it sits behind the August capture noise — treat the single print cautiously
Total: -30.7% since June 2025, the deepest of the three by a hair
Assessment: Make has been the smallest and weakest of the three platforms all year. That is standing structure, not this month's news. The low August print is partly a light-capture artifact.
Recommendation: Secondary skill, not a primary specialization. Same advice as it's been all year.

Power Automate (Niche Enterprise Tool)

Current: 29 jobs
Trend: -3.3% MoM, effectively flat
Total: flat since June 2025
Assessment: Too small for standalone specialization; sits inside the Microsoft enterprise ecosystem.
Recommendation: Microsoft-shop clients only.

Application Ecosystem: August 2026

Top 15 Applications by Job Volume

Application	Sep	Oct	Nov	Dec	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Jul MoM	Since Jun '25	Status
GoHighLevel	662	659	579	562	645	585	693	687	476	483	585	600	🟢 +2.6%	🚀 strong	Most durable app
Google Sheets	719	726	597	552	537	495	526	532	313	261	268	241	📉 -10.1%	📉 down	Long decline
HubSpot	450	447	346	333	350	349	375	378	205	243	322	239	📉 -25.8%	🟢 steady	Behind capture noise
Airtable	604	611	505	492	467	424	435	447	308	266	255	208	📉 -18.4%	📉 down	Structural decline
Slack	460	461	354	321	307	316	336	323	218	157	237	193	📉 -18.6%	📉 down	Slack softening
Notion	325	414	321	341	273	243	229	230	157	111	132	127	🟡 -3.8%	📉 down	Low and flat
ClickUp	166	182	135	169	146	134	112	103	58	81	94	88	📉 -6.4%	📉 down	Continued erosion
Zoho	124	141	128	122	82	120	101	119	41	66	82	72	📉 -12.2%	🟡 volatile	Small-number swings
Salesforce	149	106	81	96	88	110	88	102	57	64	104	68	📉 -34.6%	📉 down	Behind capture noise
Excel	549	568	403	382	316	280	251	235	108	171	353	67	💥 -81.0%	💥 collapse	See note
Monday.com	98	97	101	90	102	101	104	111	60	49	61	52	📉 -14.8%	🟡 stable-ish	Most stable alt
QuickBooks	93	95	79	64	77	73	94	95	47	70	70	46	📉 -34.3%	📉 down	Behind capture noise
Asana	93	97	67	71	64	65	49	72	27	24	49	37	📉 -24.5%	📉 down	Small numbers
Xero	43	51	40	34	30	37	43	42	17	20	20	26	🚀 +30.0%	🟢 small growth	Small numbers
Trello	75	55	48	47	26	24	31	41	11	17	26	14	📉 -46.2%	💥 near-extinct	Single digits territory

Note on Excel: the July print (353) was itself anomalously high relative to the surrounding months, and August's 67 sits below June's 171 and May's 108. Both the July spike and the August drop straddle the August capture change and a substring-matching quirk ("excel" catching "excellent"). Do not read the July→August Excel move as a real market event; the multi-month direction is a slow decline.

CRM Platform Deep Dive

Dedicated CRM Systems

CRM	Jun	Jul	Aug	Jul→Aug	Multi-Month Read	Market Assessment
GoHighLevel	483	585	600	+2.6%	Held near highs	Most durable application in the dataset
HubSpot	243	322	239	-25.8%	Steady over the year	Behind August capture noise; treat single print cautiously
Zoho	66	82	72	-12.2%	Small-number swings	Volatile, thin data
Salesforce	64	104	68	-34.6%	Range-bound low	Behind August capture noise

Key Insight: GoHighLevel at 600 jobs is the standout — up +2.6% MoM and holding near its SOTU-era highs while nearly everything else softened into August. This matches the full-year pattern: GoHighLevel has been the single most durable application across every contraction month. The sharp Jul→Aug drops in HubSpot, Salesforce and QuickBooks all sit behind the August capture change and should not be read as demand events on a single print. The honest CRM story is unchanged from prior months: pick one CRM and go deep — GoHighLevel for agencies is the best-supported bet in this dataset. The month-to-month "CRM rotation" narrative was noise; ignore it.

Generic "CRM" Mentions

Month	Mentions	Change
May 2026	817	—
June 2026	808	-1.1% MoM
July 2026	1,084	+34.2% MoM (spans the June-24 hybrid cutover)
August 2026	1,037	-4.3% MoM

Generic CRM mentions held at 1,037 in August, close to July's 1,084. The July jump spans the June-24 scrape-coverage cutover, so the Jun→Jul step is partly measurement. On the post-cutover basis, CRM demand as a category is holding roughly flat and remains the most resilient sub-segment of the market.

Alternative CRM Solutions

Application	Jun	Aug	Jun→Aug	Assessment
Airtable	266	208	-21.8%	Structural decline intact
Notion	111	127	+14.4%	Low and roughly flat
ClickUp	81	88	+8.6%	Continued long-run erosion
Monday.com	49	52	+6.1%	Most stable alternative

Combined Alternative CRMs: 475 jobs (August) vs 507 jobs (June).

Insight: The alternative-CRM category remains structurally weaker than the dedicated CRMs, and that gap has held all year. Airtable continues its long slide; Notion, ClickUp and Monday.com are bouncing around small numbers. The standing advice holds: dedicated CRM specialization — GoHighLevel in particular — is the more durable niche than the Notion/Airtable-as-CRM play.

AI Tier

The AI tier is contracting as a share of postings, which is the opposite of how raw counts read. Two populations, kept separate:

The platform set (ai_keywords) — AI terms within the 2,391 Zapier/n8n/Make jobs.
The broader automation market (broad_ai_keywords over 3,636 broad-automation jobs).

The per-term figures below are mention counts and they overlap heavily — one job naming OpenAI, RAG, and an agent is counted three times, so they do not sum and must not be turned into a percentage of the market.

The share of postings naming any AI term is computable and is the one clean AI figure (July's report published it). It fell for a fourth straight month: 48.75% in May, 46.41% in June, 42.89% in July, 40.82% in August — a cumulative 7.9 points. The count fell too, 1,147 to 976, on a capture that is if anything more complete. Every one of the ten AI categories lost share in August, which has not happened before in the months this report has tracked directly.

AI Mentions — August 2026

AI Category	Platform Set (of 2,391)	Broader Market (of 3,636)
OpenAI/GPT	534	619
Claude/Anthropic	507	648
AI Agents	363	501
Voice AI	184	227
RAG	104	116
MCP	73	118
LangChain	63	69
Vector DBs	60	64
AI Builders	36	39
Embeddings	25	29

Assessment: OpenAI/GPT and Claude/Anthropic run neck-and-neck at the top in both populations, with AI Agents a clear third. In the platform set, OpenAI/GPT appeared in 534 of 2,391 jobs and Claude/Anthropic in 507; these overlap, so they are mention counts, not a share of the market. The RAG figure (104) is notably lower than July's 375 — that print straddles the August capture change and the swing is large enough to treat as noise until a second month confirms direction. Qualitatively, AI fluency continues to trend toward table stakes in automation work: agent-building, voice agents (Retell/Vapi show up in the high-paying examples), and LLM pipelines dominate the premium tier. The premium in this data is in building production AI systems, not in mentioning AI.

August Market Dynamics

Volume: Read the Per-Weekday Figure, Not the Total

Do not read a month-over-month volume story off the stored totals this month. August's platform scrape failed on 2026-08-12 and 2026-08-18, wiping the preceding posting days (which originally captured just 13 and 8 platform jobs against a weekday norm near 95). Those days were backfilled at month-end (now 60 and 55 — partly recovered, not whole), and the same full-month re-query moved ~127 jobs from the automation tag into the platform tag across days that were never broken, inflating the stored August platform total ~6% relative to how July was measured.

Stored August platform total: 2,391
July-comparable August figure (backing out the re-tag): ~2,239
July platform total: 2,549

The honest comparison is per-weekday over cleanly-captured days, measured before the backfill: about 88 platform jobs/weekday in August against 97 in July — a high-single-digit softening. Weekend days run far lower in both months (~46 vs ~56) and are not averaged into that figure.

Assessment: August softened a few points into late summer. That is normal and small — Aaron's own series ran -2.2% July→August 2025 and -6.1% into September. This is not a contraction, a collapse, a pullback, or a rebound; it's a seasonal easing sitting on top of the year-long slow drift. Since the June 2025 baseline, the platform market is down materially on a like-basis (the SOTU-era counts are not basis-comparable to Aaron's earlier broad counts — see Methodology).

Rates: Lead With the Median

The median hourly rate printed $32.50 in August — the second sub-$35 read in the series, the other being December 2025's $30. Reading the median series straight off the data:

Sep $35 · Oct $35 · Nov $35 · Dec $30 · Jan $35 · Feb $35 · Mar $35 · Apr $35 · May $38 · Jun $35 · Jul $35 · Aug $32.50

The median has held at or very near $35 for nearly the whole series, with two downside prints (Dec, Aug) and one upside (May $38). August's $32.50 is a single-month move below the long-standing $35 line; treat it as a wobble until a second month confirms. The typical automation job on Upwork has paid around $35/hr all year, and August doesn't change that read.

The mean rate fell to $36.91, down from July's $41.41 — the lowest mean in the series. The mean has wobbled in a ~$38–43 band all year and is easily levered by a handful of high postings (the max this month was $250/hr, well below prior months' $500–999 maxes). A single-month mean move is not a re-basing. With both the median softening and the top-end max lower this month, August reads as a slightly softer rate month — but one print, not a trend.

Month	Median	Mean	High-Paying ($75+)
May 2026	$38.00	$42.20	67
Jun 2026	$35.00	$41.83	82
Jul 2026	$35.00	$41.41	105
Aug 2026	$32.50	$36.91	78

High-paying jobs ($75+) numbered 78 in August. Note the count moves with total capture, so read it alongside the volume caveat above rather than as a clean premium-demand signal.

Contract Type: The Slow Shift to Fixed-Price

fixed_share_pct: May 42.9% → Jun 40.1% → Jul 38.6% → Aug 37.1%.

Assessment: The fixed-price share has fallen every month since May, three consecutive declines totalling 5.8 points. Aaron's year-in-review flagged the opposite direction over 2025, but that series and this one are not basis-comparable, so treat them as two separate observations rather than a reversal of one trend. Within the SOTU era the direction is unambiguous and it points toward hourly work, not toward packages. At 37.1%, better than a third of postings are still fixed-price, so being able to quote a defined outcome remains worth doing — but do it because it prices the outcome, not because the data says clients are asking for it. They are asking slightly less each month.

Strategic Recommendations

Platform Strategy for Q3/Q4 2026

1. Zapier: Volume Leader

1,070 jobs, the largest of the three
-30.5% since June 2025 — the same drawdown as the others
Recommendation: Primary platform for breadth of demand. Choose it for client volume, not for a growth arrow.

2. n8n: Technical-Depth Option

960 jobs, mid its 805–1,055 SOTU-era band
-27.9% since June 2025
Recommendation: Primary or strong secondary for technically-deep, AI-adjacent work. It attracts the agent-building and self-hosted clients.

3. Make.com: Weakest of the Three

421 jobs (low August print sits behind capture noise)
-30.7% since June 2025
Recommendation: Secondary skill only. Same as all year.

4. CRM: GoHighLevel Is the Bet

GoHighLevel 600 jobs, near SOTU-era highs, most durable app in the data
HubSpot the steady enterprise option (single August print behind capture noise)
Recommendation: Pick one CRM and go deep. GoHighLevel-for-agencies is the best-supported recommendation in this dataset. Ignore month-to-month CRM "rotation" — it's noise.

Application Strategy

High-Value Specializations:

AI Systems / Agent Building ($100–250/hr)
- August's top postings are AI/LLM data engineering ($250/hr), GTM/AI-SDR engineering ($200/hr), AI ops agents on Retell + n8n ($150/hr), and senior AI systems architecture ($150/hr)
- The premium is in building production AI, not mentioning it
CRM + Automation Specialist ($75–150/hr)
- GoHighLevel holding near highs; HubSpot RevOps work at $120/hr in the examples
- Generic CRM mentions holding ~1,037, the most resilient sub-category
Voice AI Integration ($100–150/hr)
- Voice AI at 184 platform mentions; Retell/Vapi-style agent work recurring in premium postings
n8n Technical Depth ($80–150/hr)
- n8n threads through the AI-agent and reactivation-campaign premium jobs

Avoid/Deprioritize:

Trello: single digits territory (14 jobs), near-extinct
Excel-only: long structural decline (the July spike/August drop is capture noise; direction is down)
Notion/ClickUp-only as a CRM substitute: structurally weaker than dedicated CRMs
Generalist platform-only positioning: the market pays for a CRM or AI specialization on top

Rate Strategy for Q3/Q4 2026

Current Market:

Median: $32.50/hr (single-month dip below the year-long ~$35 line)
Mean: $36.91/hr (low end of the $38–43 band; one print)
High-paying ($75+): 78 jobs (read alongside the capture caveat)
Ultra-premium ($150+): 9 jobs; premium tier: 33 jobs
Max rate: $250/hr

Positioning:

Entry: $18–35/hr (the persistent low end; the median lives near here)
Experienced: $35–50/hr (requires a real specialization to escape the median gravity)
Specialist: $50–100/hr (CRM + platform + industry or AI)
Premium: $100–250/hr (production AI systems, agent architecture, voice AI)

Key: The typical job has paid ~$35/hr all year, and August's $32.50 doesn't change that read on one print. Escaping the median requires a genuine specialization — AI building or a deep CRM niche — not a repositioning statement. Package your work: fixed-price is better than a third of the market.

Q3/Q4 2026 Outlook

Expected September–November Trends

Volume: Expect a market in the low-2,000s per month on the current hybrid basis, drifting gently — Aaron's 2025 series showed a normal late-summer easing (-2.2% Jul→Aug, -6.1% into Sep). Plan for competition per posting, not a rebound.
Median rate: Watch whether September confirms August's $32.50 or reverts to $35. One sub-$35 print is a wobble, not a re-basing.
GoHighLevel: The most durable app in the data; watch it hold near 600 as confirmation of agency-CRM demand.
AI tier: Agent-building, voice AI and LLM pipelines remain the clearest premium driver, but AI vocabulary is losing share of postings four months running. Track whether that decline extends or the July retrieval-layer spike returns.
Platform mix: Zapier leads, but on share of platform mentions n8n has risen every month since May (34.33 → 36.52 → 39.17) and Make.com has fallen every month (24.68 → 21.50 → 17.18), each step larger than the last. Don't narrate a leadership change; do note the mix is consolidating.
Fixed-price: 37.1% and down every month since May. Watch whether the slide continues or stabilizes; the drift is toward hourly.

Market Structure Evolution

What's actually happening (SOTU era):

Volume drifting gently lower, with a normal late-summer softening into August (~88 vs ~97 platform jobs/weekday)
Median pay near $35/hr all year (Dec and Aug the two downside prints)
Mean range-bound $38–43, levered by a handful of high postings; August the low end at $36.91
AI vocabulary losing share of postings four months running (48.75% → 40.82%); naming it is table stakes, the premium is in building it
GoHighLevel the most durable specialization in the data

What's next:

A ~2,000–2,600 job/month platform market on the current basis
Median pay around $35/hr for generalists; specialization the only escape
AI/agent and CRM specialization the two most durable niches
Package/outcome pricing on its own merits, noting fixed-price share has fallen three straight months to 37.1%

Methodology

Data Source: Upwork job data scraped via Apify (platform set: jobs mentioning Zapier, Make.com/Integromat, n8n, Power Automate) Analysis Method: Keyword search in job titles and descriptions (substring matching) Rate Calculation: Median and mean of posted hourly maximums where provided (August hourly sample: 928 jobs)

Limitations & measurement notes:

June 2025 is the clean baseline. May 2025 is missing its first five days and is structurally invalid as a denominator; growth is anchored to June 2025 throughout.
Counts are keyword mentions and overcount — a job naming two platforms is counted under both. Platform mentions are never summed into a single "total platform jobs" headline; the headline uses unique job counts (2,391).
The SOTU-era platform counts are not basis-comparable to Aaron's earlier/broad counts. A May-2026 measurement change and a 2026-06-24 hybrid-query cutover both shifted capture; compare rates and share across the boundary, not raw counts.
August 2026 lost two capture days (2026-08-12, 2026-08-18) which were backfilled at month-end. The backfill partly recovered those days (60 and 55 vs a ~95 weekday norm) and re-tagged ~127 jobs into the platform set, inflating the stored August total ~6% relative to July's measurement. The honest volume comparison is per-weekday over cleanly-captured days measured before the backfill (~88 vs ~97).
Per-term AI keyword counts overlap heavily and are not a share of jobs. The any-AI-term share IS computable (40.82% in August) and is the only AI figure that should be read as a percentage.
Rates are posted maximums, not contracted rates. Substring matching may catch incidental mentions (e.g., "excel" in "excellent").

Report Generated: September 2026 Next Update: September 2026 data (expected October 2026) Questions/Feedback: Via GitHub Issues

This analysis is part of the ongoing Upwork Automation Market Analysis project tracking the evolution of the automation consulting marketplace.

High-Paying Automation Opportunities Guide

August 2026 Market Analysis

Report Date: September 2026 Data Period: June 2025 - August 2026 High-Paying Jobs Tracked: 78 jobs at $75+/hr in August

Executive Summary

August 2026 held the same calm story the data has been telling for a year: the median hourly rate sat at $32.50 and the typical premium job pays what it has always paid. High-paying opportunities registered 78 jobs at $75+/hr in the platform set. Before anyone reads that as a collapse against July's 105, note that August is not raw-comparable to July — the platform scrape query failed on two August posting days and was backfilled at month-end, which both under-captured two days and re-tagged ~127 jobs into the platform set. The stored August platform total (2,391) is roughly 6% inflated relative to how July was measured. The honest volume read is per-weekday, before the backfill: about 88 platform jobs/weekday in August against 97 in July — a high-single-digit softening, in line with the normal late-summer easing Aaron logged in 2025.

The average rate printed $36.91/hr, below the $41–42 band the last four months ran in. Treat that as a single-month move on a mean that is easily levered by a handful of high postings, not a re-basing — the median ($32.50) is the number that actually describes the typical job, and it has been within a couple dollars of $35 for the entire series. The premium tier remains dominated by AI-native engineering: the two top rates in August ($250/hr) are both LLM/data-pipeline builds, and the AI-agent architecture work that surfaced through the spring is still the ceiling.

The AI signal continues to climb. In the platform set, OpenAI/GPT appeared in 534 jobs, Claude/Anthropic in 507, and AI Agents in 363 (these overlap — a single job naming two counts under both, so these are mention counts, not a share of the market). Across the broader automation market (3,636 jobs), OpenAI/GPT hit 619 and AI Agents 501. The direction is unambiguous: AI is trending toward table stakes at the premium tier.

Bottom Line: Nothing broke in August. The median held near where it's held all year, the premium tier is still real but scarce (78 jobs at $75+), and AI-native engineering is still the ceiling. The August average dip and the volume print are both measurement noise — don't reposition off a single month. CRM specialization plus AI capability remains the durable play.

Rate Tiers & Market Reality

August 2026 Rate Breakdown

Tier	Rate Range	Jobs Available	Requirements	Market Trend
Ultra-Premium	$150-250/hr	9 jobs	AI/LLM + autonomous systems + industry	Concentrated in AI engineering
Premium	$100-150/hr	33 jobs	Multi-platform + CRM + AI integration	Stable band
Expert	$75-100/hr	36 jobs	Platform certified + CRM + proven results	Steady
Experienced	$50-75/hr	162 jobs	Multi-platform competency	Steady
Mid-Tier	$40-50/hr	117 jobs	Single platform proficiency	Steady
Entry	$25-40/hr	244 jobs	Basic automation tasks	Largest bucket

Key Insight: The shape hasn't changed. Entry-level work (244 jobs) is still the biggest single bucket of the hourly sample, and the premium tiers together (78 jobs at $75+) remain a thin slice at the top. The ultra-premium count (9) and premium count (33) are small enough that month-to-month wobble is noise — read them as "the premium tier exists and is scarce," which has been true all year, not as a directional move. The max rate observed was $250/hr.

Average Rate Evolution

Month	Market Average	Median	High-Paying Count ($75+)	Trend
Sep 2025	$38.74/hr	$35	—	Baseline range
Oct 2025	$38.17/hr	$35	—	Range-bound
Nov 2025	$38.79/hr	$35	—	Range-bound
Dec 2025	$37.58/hr	$30	144	Range-bound
Jan 2026	$40.20/hr	$35	165	Q1 uptick
Feb 2026	$43.27/hr	$35	170	Single-month outlier
Mar 2026	$40.66/hr	$35	167	Reverted
Apr 2026	$39.46/hr	$35	139	Baseline range
May 2026	$42.20/hr	$38	67	Measurement change*
Jun 2026	$41.83/hr	$35	82	Range-bound
Jul 2026	$41.41/hr	$35	105	First full hybrid month
Aug 2026	$36.91/hr	$32.50	78	Mean dip, median near $35

*Absolute counts from May 2026 forward reflect a scrape-coverage change and are not directly comparable to Aaron's earlier months. The median and rate distribution are the capture-independent signals.

Analysis: The average printed $36.91 — its lowest read in a year — but this is a mean, and the mean has wobbled in a narrow $37–43 band the entire series while being easily moved by a few high or absent high-rate postings. The median tells the more honest story: $32.50 this month, within a couple dollars of the $35 it's held in nearly every month of the series. Treat August's average dip as a single-month move until September confirms it. The premium-count drop (105 → 78) sits inside the August measurement change and should not be read against July as clean month-over-month.

Ultra-Premium Opportunities ($150-250/hr)

Profile Requirements

Technical Stack:

LLM/AI engineering (OpenAI, Claude/Anthropic APIs, agentic systems)
AI agent architecture (multi-step reasoning, tool use, memory)
Data pipelines and ETL (production-grade, scalable)
n8n or Zapier for workflow orchestration
Voice AI (Retell, Vapi — still emerging)
Custom development (Python, TypeScript, Node.js)

Industry Knowledge:

Home services / trades (roofing, restoration, construction)
Healthcare / men's health (growth and data leadership)
Financial services (B2B lead gen, fund administration)
AI-native businesses (agent-operated systems, growth engineering)
Event / fleet logistics (AI ops layers)

Proven Track Record:

Production AI/LLM deployments, not prototypes
$50K+ project portfolio
Quantified business outcomes
Active technical presence (GitHub, content)

August Market Reality

The ultra-premium tier held 9 jobs at $150+/hr — a small count, as it has been most of the year. The top of the rate stack is still AI-native engineering: the two highest rates in August ($250/hr each) are both LLM-driven data-pipeline and automation builds, and the $150+ roles skew toward AI systems architecture and fractional executive/advisory work in specific verticals. This is the same standing structure the spring reports described — the premium ceiling rewards building AI systems, not mentioning AI.

Entry Barrier: Very high — requires production LLM/data-engineering experience and, increasingly, domain depth in a specific vertical.

Actual August Examples

Example 1: AI & Data Engineer — LLM-Driven Workflows & Data Pipelines

Rate: $250/hr
Requirements: AI/data engineering, APIs, ETL, LLM integrations, scalable pipelines
Project: Build reliable data pipelines, LLM-powered workflows, and automation systems, turning raw data and business processes into scalable systems
Context: Highest rate observed in August; LLM + data engineering as the standalone premium tier

Example 2: GTM Engineer — Clay, Customer.io, APIs & AI Automation

Rate: $200/hr
Requirements: Programmatic outbound, AI SDR, product-signal routing, lifecycle automation, custom AI platform
Project: Revenue-as-an-engineering-problem build — systems doing the work of a twenty-person team
Context: AI-native growth engineering commands the top of the non-data-engineering stack

Example 3: Senior AI Systems Architect — Business AI Platform

Rate: $150/hr
Requirements: Senior AI systems architect / AI agent engineer, sophisticated agentic system design (explicitly "not a basic chatbot or simple automation")
Project: Design and build an AI-powered business operating system
Context: Industry-vertical AI agent architecture continues to command premium

Example 4: AI Operations Agent for Event Fleet Logistics

Rate: $150/hr
Requirements: Retell + n8n voice agent stack, Google Sheets ops integration, full AI ops layer
Project: Build a complete AI ops layer on top of an existing live voice-agent driver check-in system
Context: Voice AI + n8n in a specific vertical; the emerging-premium voice niche persists

Example 5: GoHighLevel Consultant — Home Services / Roofing (Advisory)

Rate: $150/hr
Requirements: GoHighLevel depth + roofing/home-services domain, 5+ years, advisory not execution
Project: Advisory support for a large home-services GHL build where the agency lacks contractor-space depth
Context: CRM depth + industry domain commands premium even in an advisory-only role

Premium Opportunities ($100-150/hr)

Profile Requirements

Core Competencies:

Deep expertise in one platform + working knowledge of a second
CRM specialization (GoHighLevel, HubSpot)
AI integration (production API work, not just chat)
Industry-specific positioning
Strategic consulting beyond implementation

Differentiators:

Published case studies with quantified ROI
Platform certifications (n8n, HubSpot, Zapier)
Visible thought leadership
Strong client references and repeat business

August High-Demand Segments

1. GoHighLevel Agency Automation ($100-150/hr)

Opportunity: 600 jobs — the most resilient application in the dataset

Skills Needed:

GoHighLevel platform mastery (workflows, CRM, funnels, snapshots)
n8n or Zapier for external integrations
Marketing automation and funnel optimization
Agency and industry-vertical advisory

Why Premium: GoHighLevel is the standing bright spot across the whole series — it has held near its highs even in softer months. August's $150/hr roofing-advisory role shows the premium comes from pairing GHL depth with a vertical. Agencies keep investing in it.

August Insight: GHL at 600 jobs remains the strongest CRM signal in the data. Pick a vertical and go deep.

2. AI / LLM Engineering ($120-250/hr)

Opportunity: The premium ceiling, dominated by AI-native builds

Skills Needed:

OpenAI and Claude/Anthropic API mastery (tool use, agents, memory)
Data pipelines, ETL, RAG architecture
Vector databases (Pinecone, Weaviate, Chroma, Qdrant)
Production deployment (latency, reliability, cost)

Why Premium: August's two $250/hr roles were both LLM/data-engineering builds; the $150–200 band was AI systems architecture and AI-native growth engineering. Generic AI-enhanced workflows sit at $75–100/hr — standalone AI systems engineering commands $150–250/hr. This has been the structure since the spring.

3. HubSpot RevOps & Multi-CRM Integration ($100-150/hr)

Opportunity: HubSpot at 239 jobs; generic CRM mentions at 1,037

Skills Needed:

HubSpot RevOps (custom integrations, coded workflows)
Migration and data-quality expertise
Multi-system orchestration (Xero, Chargebee, billing)
Industry-specific CRM configuration

Why Premium: August's $120/hr HubSpot RevOps role and $115/hr fractional B2B marketing role (financial services + HubSpot) show CRM depth plus a vertical continues to command premium. CRM demand is the most durable sub-segment in the data.

4. n8n Technical Specialist ($100-150/hr)

Opportunity: 960 jobs — the technical-depth platform

Skills Needed:

n8n self-hosting (Docker, custom nodes, scaling)
Claude/GPT API integration with n8n workflows
Complex workflow and AI-agent orchestration
Voice AI + n8n stacks (Retell)

Why Premium: n8n's technical premium holds at the top of the stack — August's $150/hr AI-dev/n8n reactivation-campaign role and the $150/hr Retell + n8n fleet-ops build both anchor to n8n depth. As standing structure, n8n is the technical-depth option, not a horse-race winner — pair it with CRM or a vertical to insulate from platform-level wobble.

Expert Tier Opportunities ($75-100/hr)

Profile Requirements

Core Skills:

Deep expertise in one platform OR competency in two
50+ completed automations
Industry-specific knowledge
Project management capabilities

Minimum Credentials:

10+ client references
Platform certification
Public portfolio with case studies

Resilient Segments

1. CRM Implementation ($75-100/hr)

Opportunity: GoHighLevel (600), HubSpot (239), generic CRM mentions (1,037)

Focus:

GoHighLevel agency setup and configuration
HubSpot onboarding and RevOps
CRM data migration
Multi-system orchestration

Why Stable: CRM is the most resilient category in the data. Even in softer months, CRM demand holds while other categories thin.

2. Marketing & Growth Operations ($75-100/hr)

Opportunity: B2B lead-gen and growth-ops work surfaced repeatedly in August's premium examples

Focus:

Outbound automation (LinkedIn, cold email, SMS)
Funnel and lifecycle systems
Marketing-ops infrastructure (technical marketing manager roles)
AI-based campaign optimization

Why Stable: Multiple $100–130/hr growth-ops and campaign-manager roles in August. Growth is increasingly scoped as an engineering problem.

3. Accounting & Finance Automation ($75-100/hr)

Opportunity: QuickBooks (46), Xero (26), plus fractional finance advisory

Focus:

QuickBooks / Xero integration
Financial reporting automation
FP&A and controller-support workflows

Why Stable: Accounting automation delivers measurable ROI. August's $150/hr fractional controller role shows the finance-advisory tier reaches into premium.

4. Zapier Advanced Implementation ($75-100/hr)

Opportunity: 1,070 jobs — the volume leader

Focus:

Zapier Tables and Interfaces
Complex multi-step workflows
Error handling and monitoring
Migration from legacy automation

Why Relevant: Zapier remains the volume leader, which means a steady flow of advanced implementation work. It's the breadth-and-consistency platform.

Navigating the Current Market

August Market Reality

Market Status:

Platform-set jobs: 2,391 (not raw-comparable to July — see below)
Average rate: $36.91/hr (single-month mean dip)
Median rate: $32.50/hr (near the year-long $35 level)
High-paying jobs: 78 at $75+/hr
Ultra-premium ($150+): 9 jobs
Expert tier ($75-100): 36 jobs
Max rate: $250/hr

A note on the numbers you can and can't trust this month. The platform scrape query failed on two August posting days and was backfilled at month-end. The backfill partly recovered those days and also re-tagged ~127 jobs into the platform set, inflating the stored August total roughly 6% relative to how July was measured. So do not read 2,391-vs-2,549 as a clean month-over-month figure. The honest volume comparison is per-weekday over cleanly-captured days, measured before the backfill: about 88 platform jobs/weekday in August against 97 in July — a high-single-digit softening. For calibration, Aaron's series ran -2.2% July→August 2025 and -6.1% into September; a small late-summer easing is normal. The capture-independent signals — median rate and platform share — are the ones to steer by.

The Signals That Actually Matter

Median rate held near $35 all year. August's $32.50 sits inside the same narrow band that's defined the series. The typical automation job pays what it has paid the whole time — that's your gravity, and escaping it requires a real specialization, not a repositioning statement.

AI keeps climbing. OpenAI/GPT (534), Claude/Anthropic (507), and AI Agents (363) in the platform set (overlapping mention counts, not a share). In the broader market these run higher still. AI fluency is no longer a premium niche — it's the baseline expectation, and the premium is in building, not mentioning.

Action Plan

If Revenue Growing: Build AI-Native

Immediate:

Audit positioning for AI-native opportunities (the $250/hr work is LLM/data engineering)
Build a production AI-agent or data-pipeline demo with metrics
Target ultra-premium AI work selectively — it's scarce (9 jobs), win it with proof
Deepen GoHighLevel if you have the agency relationship

Q4 Strategy:

Position as AI systems architect, not just automation builder
Develop one industry vertical (home services, healthcare, financial services all surfaced in August)
Lock in retainer relationships before year-end budget cycles

If Revenue Stable: Hold Rates, Specialize Deeper

Actions:

Median near $35 means discipline matters — don't discount to chase volume
Premium share is thin; specialization is the moat
Add CRM specialization if not already (GoHighLevel most resilient)
Pair platform expertise with one industry vertical

Goal: Maintain $75+/hr positioning even through late-summer softness

If Revenue Down: Volume Strategy with Specialization

Immediate:

Increase proposal volume — the market is softer and more competitive
Target the $50-75/hr range (162 jobs) with a specialization narrative
Add CRM certification (GoHighLevel or HubSpot)
Build AI-integration capability for premium positioning

30-Day Plan:

Complete one CRM certification
Build 2 AI-enhanced automation portfolio projects
Target 25+ proposals per week
Apply to the $50-75/hr range with strong specialization framing

Rate Optimization Strategies

1. Reposition Around AI Systems Engineering

Strategy: The premium ceiling rewards building AI systems, not integrating them

Before: "Automation specialist with AI integration experience" After: "AI systems engineer — I build LLM-driven pipelines and production agents"

Tactics:

Build a production data-pipeline + LLM demo with metrics (the $250/hr work)
Master one agent framework and one vector DB deeply
Position around outcomes (autonomous workflows, decision agents)

Rate Impact: +50-100% positioning premium for AI-native vs generic AI-enhanced

2. CRM + Industry Stacking

Strategy: CRM is the most durable specialization in the data — pair it with a vertical

Before: "GoHighLevel specialist" After: "GoHighLevel lead for home services / roofing"

Tactics:

GoHighLevel for agencies and trades (600 jobs, most resilient)
HubSpot for B2B / financial services (239 jobs)
Industry framing creates pricing power (see August's $150/hr advisory role)

Rate Impact: +30-50% for industry-specific CRM positioning

3. Value-Based Pricing for Stability

Strategy: Decouple from hourly compression. Note the data does not show clients moving toward packages: fixed-price share has fallen every month since May (42.9 → 40.1 → 38.6 → 37.1). Package because it prices the outcome and insulates you from a sliding low end, not because demand is shifting that way

Package Examples:

"AI Systems Build: $25,000-50,000" (LLM pipeline + agent)
"GoHighLevel Agency Setup: $12,000-18,000" (CRM + automation + funnels)
"CRM Migration & Consolidation: $20,000-40,000"
"Accounting Automation: $8,000-12,000" (QuickBooks + reporting)

Benefits:

Insulates from hourly-rate wobble
Premium via outcomes, not hours
Higher margins than time-and-materials

4. Retainer Conversion

Strategy: Convert project clients to recurring revenue before year-end

Retainer Tiers:

Maintenance: $3,000-5,000/month (monitoring, updates, support)
Optimization: $5,000-12,000/month (continuous improvement, new workflows)
Strategic: $12,000-30,000/month (consultation, training, roadmap)

August Context: With volume softening into late summer, recurring revenue is more valuable than chasing one-off bids.

Platform Strategy for Q4 2026

August Landscape

August Facts (platform set, note the August measurement caveat):

Zapier: 1,070 jobs — the volume leader
n8n: 960 jobs — the technical-depth option
Make.com: 421 jobs — the weakest of the three
Power Automate: 29 jobs — marginal

Strategic Implications:

Zapier Positioning:

Standing volume leader across the whole series
Most accessible mid-market platform
Strategy: Primary platform for breadth and consistency

n8n Positioning:

The technical-depth option; premium niche intact at the top of the rate stack
Pair with CRM or a vertical to insulate from platform wobble
Strategy: Niche premium positioning, not primary-only

Make.com Positioning:

The weakest of the three, as it's been all year
Strategy: Secondary skill, not a primary specialization

On the horse-race: Don't read month-to-month platform counts as lead changes. Over the full record all three declined roughly the same from peak; the standing structure — Zapier for volume, n8n for depth, Make weakest — is what to plan around, not this month's arrows. August's per-platform counts also sit inside the month's measurement caveat.

Application Strategy

Safe Bets

1. GoHighLevel (600 jobs)

The most resilient application in the dataset, all year
Agency and trades automation demand structurally durable
Strongest CRM signal in the data
Positioning: Agency CRM + automation specialist, one vertical deep

2. HubSpot (239 jobs)

Steady enterprise-marketing and RevOps demand
Certification commands premium
Positioning: B2B / financial-services marketing automation

3. Airtable (208 jobs)

Durable as a database/CRM-alternative layer
Pairs well with AI and platform work
Positioning: Airtable + CRM/AI systems

4. Google Sheets (241 jobs)

Persistent ops-layer demand across small businesses
Frequently the integration target in AI-ops builds
Positioning: Ops-layer integration specialist

Watch List

Salesforce (68 jobs)

Enterprise CRM, volatile month-to-month on small numbers
Watch: September for direction

Slack (193 jobs)

Steady workflow-integration demand
Watch: Continued stability

QuickBooks / Xero (46 / 26 jobs)

Small but consistent accounting-automation niche
Watch: ROI-driven demand holds through year-end

Avoid/Deprioritize

Long-running structural decline (project-management tools):

Trello: 14 jobs — smallest tracked, long downtrend
Jira: 9 jobs — marginal
Asana: 37 jobs — long-run decline
ClickUp: 88 jobs — off its highs

Note on Excel: August's Excel print (67) sits inside the month's measurement caveat and against a wobbly recent series (171 in June, 353 in July). Excel/Notion-only positioning has been in long-run structural decline across the full record — that trend is the signal, not any single month's count.

Strategy: Exit PM-tool-only and spreadsheet-only specializations. Focus on CRM + platform + AI/industry stacking.

Q4 2026 Outlook

Expected September-November Trends

September 2026:

Volume likely eases modestly further (Aaron's 2025 series ran -6.1% July→Sept)
Median expected to hold near $35 — watch whether August's $32.50 mean dip persists or reverts
GoHighLevel resilience continues
AI-agent and LLM-engineering work stays at the premium ceiling

October-November 2026:

Volume in the softer late-year range on the current basis
Median steady near $35; no return to the February $43 outlier expected
CRM demand remains the primary durable driver
AI penetration continues climbing toward table stakes

Q4 Bottom Line:

A softer, competitive market where the typical job pays near $35
Premium share thin and won with proof, not positioning
CRM specialization the safest bet
AI systems engineering the premium ceiling

Final Recommendations

The Premium Positioning Formula (Q4 2026)

1. CRM Foundation (Choose One):

GoHighLevel: Most resilient (600 jobs), agency/trades focus, $100-150/hr
HubSpot: Steady (239 jobs), B2B/financial services, $90-140/hr
Airtable: Database/CRM-alternative layer, pairs with AI, $75-200/hr

2. Platform Proficiency:

Option A: Zapier (volume leader, breadth and consistency)
Option B: n8n (technical depth, premium niche; not primary-only)
Option C: Both (maximum flexibility, rate premium for verified multi-platform)

3. AI Capabilities (Now the Baseline):

OpenAI / Claude API mastery (production agents, not chat)
Data pipelines, ETL, RAG, vector DBs
Voice AI (Retell/Vapi) — still emerging premium
Agent orchestration and tool use

4. Proof and Positioning:

3+ case studies with quantified ROI
One industry vertical, deep
Active thought leadership
Value-based / package pricing

Formula Result: $100-200/hr sustainable rates, insulated from month-to-month noise

The Bottom Line

August 2026 was a calm month wearing a scary number. The average dropped to $36.91 and the high-paying count read 78, but the average is a mean that wobbles on a handful of postings, and the count sits inside a month whose scrape failed twice and was backfilled unevenly. Neither is a clean signal.

The numbers that are clean:

Median rate $32.50 — within a couple dollars of the $35 it's held all year
Per-weekday volume ~88 vs July's ~97 — a high-single-digit late-summer softening, normal for the season
AI mention counts climbing (OpenAI/GPT 534, Claude 507, AI Agents 363 in the platform set — overlapping, not a share)
GoHighLevel at 600 — the most durable specialization in the data

Thriving Segments:

AI/LLM systems engineering ($150-250/hr) — the premium ceiling
GoHighLevel agency + trades automation (600 jobs, most resilient)
Growth/marketing operations ($100-130/hr)
Industry-vertical AI agent work ($150+/hr)

Struggling Segments:

Generic platform expertise without CRM or AI overlay
Single-platform positioning
PM-tool-only and spreadsheet-only specialists
Sub-$50/hr generalist positioning

The market didn't break in August, and it didn't collapse. It softened a little, as late summer tends to, while the typical job paid what it has paid all year. For specialists with CRM plus AI positioning, the opportunity is where it's been. For generalists competing at the median, the gravity is the same $35 it's always been.

Softer volume. Median holding near $35. AI systems engineering as the ceiling. CRM as the most resilient specialty. Don't reposition off one noisy month.

Resources

Platform Mastery:

n8n: Self-hosting guides, custom node development, Docker deployment
Zapier: Tables/Interfaces, advanced features, error handling
Make.com: Complex scenarios, error handling

AI/LLM Engineering:

OpenAI Cookbook (agents, function calling, structured outputs)
Anthropic Claude API documentation, tool use, agentic systems
LangChain / LangGraph for agent orchestration
Vector databases (Pinecone, Weaviate, Chroma, Qdrant) for RAG

CRM Specialization:

GoHighLevel: Certification program, community, snapshots (most resilient in data)
HubSpot: HubSpot Academy (free certification)
Airtable: Automations, scripting, interface design
QuickBooks: Integration certification

Business Skills:

Value-based / package pricing frameworks
Case study development
Retainer model design
AI systems thought leadership

Report Generated: September 2026 Based on: August 2026 platform-set data (jobs mentioning Zapier / n8n / Make.com) High-Paying Jobs Tracked: 78 at $75+/hr in August Note: August platform capture was affected by two failed scrape days that were backfilled at month-end; absolute counts are not directly comparable to July. Median rate and platform share are the capture-independent signals. Next Update: September 2026 data (expected October 2026)

Part of the Upwork Automation Market Analysis Project Tracking the evolution of automation consulting opportunities

Data: JSON · CSV