Website Design in Snohomish County

Fifteen cities, and the widest growth spread in the region

The Washington Office of Financial Management publishes official April 1 population estimates for every incorporated place in the state under RCW 43.62.030. The April 2026 release records fifteen Snohomish County cities at or above 5,000 residents, holding 488,195 people between them.

Everett is the largest at 114,900, followed by Marysville at 76,820, Edmonds at 43,770, Lynnwood at 43,070 and Lake Stevens at 42,590. The middle of the distribution is unusually dense: Mountlake Terrace holds 24,750, Arlington 23,390, Mill Creek 21,630, Mukilteo 21,610, Monroe 20,930 and the Snohomish County portion of Bothell 20,530. Below those, the city of Snohomish holds 10,770, Stanwood 9,325, Sultan 7,480 and Brier 6,630.

What distinguishes this county is the growth spread. Between the 2020 census and the 2026 state estimate, Sultan grew 45.4 percent, Stanwood 21.0 percent, Lake Stevens 19.5 percent, Arlington 17.7 percent, Mountlake Terrace 16.3 percent and Lynnwood 11.7 percent. Over the same six years Mukilteo grew 0.3 percent, Brier 1.1 percent, Edmonds 2.1 percent and Mill Creek 3.4 percent. Everett itself grew 3.9 percent.

That is a range from under half a percent to over forty-five percent inside a single county, and it is the widest spread of any large county in the Puget Sound region. Five cities here grew faster than any city in Pierce County. Four grew more slowly than all but one. A business in Sultan is selling into a town that has added nearly half again its 2020 population, where a large share of households arrived recently and have chosen few suppliers. A business in Mukilteo is selling into a population that has effectively not changed in six years. Those are different problems, and the same marketing plan does not address both.

• Everett 114,900 · Marysville 76,820 · Edmonds 43,770 · Lynnwood 43,070 · Lake Stevens 42,590
• Fastest growth 2020 to 2026: Sultan 45.4%, Stanwood 21.0%, Lake Stevens 19.5%, Arlington 17.7%
• Slowest growth: Mukilteo 0.3%, Brier 1.1%, Edmonds 2.1%, Mill Creek 3.4%
• Combined population of cities at or above 5,000: 488,195 across 15 cities

Manufacturing is the largest employer and one of the best paid

The Employment Security Department publishes a Snohomish County labour market profile with a sector breakdown that looks nothing like the profiles for King or Pierce. Its June 2026 payroll estimates put total nonfarm employment at 307,200 jobs, comprising 264,100 private and 43,100 government. Of that total, 83,500 are goods-producing and 223,700 service-providing.

Manufacturing is the largest single sector at 57,400 payroll jobs. Government follows at 43,100, private education and health services at 42,400, retail trade at 34,200, professional and business services at 33,400, leisure and hospitality at 30,500 and mining, logging and construction at 26,100. The 2024 covered employment census gives the same top sector from a different methodology: manufacturing at 55,884, then government at 41,055, health care and social assistance at 36,531, retail trade at 34,523 and construction at 24,847.

The wage data is where this becomes commercially interesting. The 2024 average annual wage across all covered employees in the county was 79,964 dollars. Manufacturing averaged 117,339 dollars, roughly 47 percent above that county average. Only professional, scientific and technical services at 136,655, information at 125,996 and utilities at 122,132 paid more, and all three employ far fewer people here.

That combination is genuinely unusual. In most local economies the largest employer is a comparatively low-paying sector such as retail or food service, and the highest-paying sectors are small. In Snohomish County the biggest employer is also near the top of the wage table. For a business selling to local households, that means the single largest identifiable customer group in the county is both numerous and comparatively well paid, which is a materially different proposition from the retail and hospitality-weighted workforce next door in Pierce.

• June 2026 payroll jobs: 307,200 total, 264,100 private, 43,100 government
• Largest sector: manufacturing at 57,400 payroll jobs and 55,884 covered employees
• County average annual wage 2024: 79,964 dollars
• Manufacturing average wage: 117,339 dollars, about 47 percent above the county average

The county lost jobs, and manufacturing was not the reason

Between June 2025 and June 2026 total nonfarm employment in Snohomish County fell by 2,700 jobs, a decline of 0.9 percent. Private employment fell by 3,000 or 1.1 percent while government added 300 or 0.7 percent. That makes this the only one of the three central Puget Sound counties to record a net employment decline over that period, against Pierce at plus 1,300 and King at minus 800.

The received explanation for a manufacturing county losing jobs would be that manufacturing shrank. The state data says the opposite. Manufacturing added 1,700 jobs, a rise of 3.1 percent, and was the largest single sector gain in the county. Goods-producing employment as a whole rose 2,100 or 2.6 percent, while service-providing employment fell 4,800 or 2.1 percent.

The losses were concentrated in services, and one of them is dramatic. Other services fell by 3,100 jobs, a decline of 28.7 percent, which is by far the largest proportional sector movement in any county profile we have read this year. Financial activities fell 1,100 or 8.5 percent, retail trade 700 or 2.0 percent, information 500 or 13.2 percent, wholesale trade 300 or 3.3 percent and transportation, warehousing and utilities 200 or 2.2 percent. Against those, leisure and hospitality added 500, professional and business services 400, construction 400 and private education and health services 200.

A 28.7 percent fall in a sector is large enough that we would want to see the next release before treating it as settled, and a category as broad as other services can move sharply for definitional or reclassification reasons rather than because that many jobs disappeared. We are stating what the published profile records rather than explaining it, because we do not know the cause and the profile does not give one. What is safe to say is narrower: the decline in this county was a services story, not a factory story, and anyone describing it as industrial decline has the direction backwards.

• Total nonfarm employment: down 2,700 jobs or 0.9 percent year over year
• Manufacturing: up 1,700 jobs or 3.1 percent, the largest sector gain
• Goods-producing up 2,100 (+2.6%) while services fell 4,800 (-2.1%)
• Largest loss: other services down 3,100 jobs or 28.7 percent

Why the unemployment rate here moves for reasons other counties do not share

Snohomish County unemployment was 5.0 percent in June 2026, with 22,915 people seeking work out of a labour force of 455,204 and 432,289 employed. That was up from 4.8 percent in May 2026 and 4.7 percent in June 2025, and above both the state figure of 4.7 percent and the national figure of 4.4 percent. The rate peaked at 6.0 percent in January 2026 before easing back.

The profile makes a point about this county that it does not make about its neighbours. It states that the Snohomish County unemployment rate is highly dependent on changes in aerospace parts manufacturing. That is a concentration risk stated plainly by the state agency: a single industry can move the headline number for the whole county, which means the rate here is a less reliable proxy for general local business conditions than the same figure would be elsewhere.

There is a second and less obvious effect the profile identifies. Residents of Snohomish County who commute to work in King County are counted in the King County employment figures, not the Snohomish ones. So the published Snohomish rate describes the county as a place of work more than as a place of residence, and a meaningful share of the people living in Lynnwood, Mountlake Terrace, Bothell and Mill Creek do not appear in it at all.

The practical consequence for marketing is worth stating. If you are selling to Snohomish County residents rather than to Snohomish County employers, the local unemployment rate is a weaker signal than it looks, because it partly measures a different population. The city-level population and growth figures from the Office of Financial Management describe your actual addressable market more accurately, and they are not affected by where anybody works.

• June 2026 unemployment 5.0 percent, up from 4.8 in May and 4.7 a year earlier
• Labour force 455,204 with 22,915 people seeking work
• Peaked at 6.0 percent in January 2026
• ESD states the rate is highly dependent on aerospace parts manufacturing

Snohomish is not a Seattle bedroom community, statistically speaking

The common description of this county is as a residential overflow for Seattle, a place people live because King County is expensive. There is real commuting, and the profile confirms it matters enough to distort the unemployment statistics. But the description fails as an account of the local economy, and using it as a marketing premise leads to the wrong plan.

The county supports 307,200 payroll jobs of its own, which is a larger employment base than Pierce County had covered employees in 2024. Its largest sector, manufacturing, is not a commuter sector, and it grew while the county overall shrank. Its average wage of 79,964 dollars sits well above the Pierce figure of 70,396. A county with its own dominant, well-paid, growing primary industry is not functioning as a dormitory.

The difference matters for what a website needs to say. If you accept the bedroom-community framing you will write copy aimed at commuters, emphasise evening and weekend availability, and pitch against Seattle providers. If you look at the employment data you will notice that the largest single group of local earners works in manufacturing inside the county, on shift patterns, at wages nearly half again the county average. Those two audiences want different things and search for them differently.

Both audiences exist, and in different cities. The southern cities closest to the King County line, including Mountlake Terrace, Brier, Bothell and Mill Creek, are where the commuter pattern is strongest and where growth ranges from 1.1 to 16.3 percent. The northern and eastern cities including Arlington, Stanwood, Monroe and Sultan are further from that boundary and have grown faster, from 6.2 to 45.4 percent. Deciding which of those two counties you are actually marketing into is a more useful first question than anything about the region as a whole.

• The county supports 307,200 payroll jobs of its own
• Average wage 79,964 dollars, above Pierce at 70,396 and below King at 132,276
• Manufacturing, the largest sector, is a local employer rather than a commuter sector
• Southern cities show the commuter pattern; northern and eastern cities grew faster

What the two official employment figures mean here

As in every Washington county, there are two authoritative employment numbers published by the same agency and they do not match. Snohomish County recorded 307,200 nonfarm payroll jobs in June 2026 and 297,617 covered employees as a 2024 annual average. Neither is wrong and neither supersedes the other.

The payroll series is a monthly sample-based estimate of jobs on employer payrolls, so someone holding two jobs is counted twice, and the figures are revised as more information arrives. It excludes agriculture, freelancers, contractors and the self-employed. The covered employment series is an administrative count of individual employees at firms covered by the Washington State Employment Security Act, built from actual unemployment insurance filings, and it includes agricultural employment. It is far more accurate and published annually, well after the period it describes.

In this county the two series happen to agree on the most important structural fact, which is convenient and not always the case. Both rank manufacturing first, at 57,400 payroll jobs and 55,884 covered employees. In Pierce County the two series disagree on the top sector entirely, so the agreement here is a genuine signal rather than an artefact.

Our rule is unchanged. Use covered employment when the claim concerns structure, because it is a census rather than a sample. Use the monthly payroll estimates when the claim concerns recent direction, because they are current. Never combine figures from the two series inside a single comparison, because the resulting number describes nothing that exists.

• Payroll jobs June 2026: 307,200 — monthly sample, counts jobs, revised
• Covered employees 2024: 297,617 — administrative census, counts people, annual
• Both series rank manufacturing first, unlike in Pierce County
• Use covered data for structure, payroll data for direction, never mixed

What the monthly detail shows, and where the data has gaps

Annual comparisons are the right place to start but they hide movement, and in this county the monthly detail points the other way from the yearly one. Between May and June 2026 total employment rose by 1,700 jobs. Leisure and hospitality added 700, private education and health services 500, manufacturing 400, mining, logging and construction 400 and government 300, partly offset by professional and business services losing 700 and retail trade 100.

So a county down 2,700 jobs over twelve months was up 1,700 over the most recent single month. Both figures are accurate and neither settles anything on its own. Our practice is to quote the annual figure when describing conditions, because a single month is noise, and to check the latest month before making a recommendation, because a plan written against a twelve-month trend that has since reversed is a plan written against the past.

There is a subtler point in the labour force numbers that is worth surfacing because almost nobody reads this far into a profile. The county labour force was 455,204 in June 2026, down from 461,367 in May and 460,758 in June 2025. Employed persons fell from 439,124 to 432,289 over the same month. So the labour force itself contracted by more than six thousand people while unemployment rose only from 4.8 to 5.0 percent. A rate can stay almost flat while the number of people participating in the workforce falls, and when that happens the rate understates what is actually going on. Anyone using a single headline percentage as a proxy for local demand should know that.

Two data caveats belong on the record. The federal Bureau of Labor Statistics cancelled its October 2025 unemployment publications, so there is a genuine hole in the monthly series for every county in the country including this one. And the county rate ran between roughly 3.9 and 4.8 percent through 2024 before peaking at 6.0 percent in January 2026, which means the current 5.0 percent is elevated against its own recent history rather than merely against the state average. We would rather state both of those than present a clean series that does not exist.

• May to June 2026: employment up 1,700 jobs, against down 2,700 year over year
• Labour force fell from 461,367 in May to 455,204 in June 2026
• BLS cancelled October 2025 unemployment publications, leaving a gap in the series
• The rate ran 3.9 to 4.8 percent through 2024 and peaked at 6.0 percent in January 2026

What we actually do first for a Snohomish County business

Our order of work is the same in every market because the arithmetic underneath it is the same. What differs is which step is the binding constraint, and in this county it is usually clarity about service area rather than mobile performance or search visibility.

The reason is the geography. A county containing both Brier at 6,630 residents and Everett at 114,900, spread from the King County line to Stanwood, means a plausible customer may be forty miles from the business and not know whether they are served. So the first thing we confirm is that the site states which cities the business covers, in text, in the first screen, without a scroll. The second is that it loads and converts on a mid-range phone on a mobile connection. The third is that enquiries are measurable, so calls and forms can be attributed to a source. Only then would we consider paid traffic.

The sequencing is not a preference. Advertising multiplies whatever conversion rate already exists, so buying traffic before fixing conversion multiplies the wrong number at a cost per click that is indifferent to how good the business is. Fixing the conversion path first is nearly always cheaper than the advertising spend it saves.

The failure we see most often in the fast-growing cities here, places like Sultan, Stanwood, Lake Stevens and Arlington, is a website that has not caught up with the town. It was built when the population was materially smaller, it names one city when the business now serves four, and it ranks for nothing in the newer areas. In a city that has grown 20 or 45 percent since 2020 that is not a cosmetic problem, it is a large share of the addressable market unable to establish that the business serves them.

• State the specific cities served in the first screen, in text
• Confirm mobile speed and conversion before buying traffic
• Make enquiries measurable so channels can be compared
• In fast-growing cities, check the site has caught up with the service area

What Nine99 is and is not in Washington

Nine99 Inc. is a Wyoming corporation. It has no office, no staff and no telephone number in Washington state, and none in Snohomish County. We build and manage websites for businesses here remotely, and we would rather say so on a page competing for attention in this market than let a customer assume otherwise.

Two consequences follow and both are worth stating. Nine99 does not appear in Google map results or the local pack for any Washington city, because those surfaces require a verified business location or service area that we do not have. And we do not publish local business structured data implying a Snohomish County presence, because that would misrepresent us to customers and to search engines at the same time.

None of this limits what we can do for a business that genuinely operates here. Your business has a real address, real service areas and a real Google Business Profile, and those are the assets that earn map visibility. Our work is to make the site behind them worth arriving at, to earn organic rankings for the terms your customers actually use, and to make sure you can tell which channel produced each enquiry.

The same standard applies to every figure above. This page contains no invented statistics about local website prices, click costs or conversion rates, because we hold no verified dataset for any of them and an invented average is the most quotable false claim a page like this could publish. Every population, employment and wage number is attributed to a named state agency publication with a retrieval date, and anything that is our judgement rather than a published fact is labelled as ours. An agency willing to show you a local address it does not occupy is showing you how it will treat your own listings.

• Nine99 Inc. is a Wyoming corporation with no Washington office, staff or phone
• We do not appear in Washington map results and publish no local business markup
• Organic ranking needs no local address; map and local pack results do
• Every figure here comes from a named state agency with a retrieval date

What most sources get wrong

Each claim below is one we found published and being repeated. The correction is drawn from primary documentation.

The Nine99 view

Our own judgement, not drawn from the sources above. Treat it as opinion formed in practice rather than documented fact.

The most useful thing in this profile is a division most marketing misses entirely: this county behaves as two markets separated by distance from the King County line. The southern cities, Mountlake Terrace, Brier, Bothell and Mill Creek, carry the commuter pattern and grew between 1.1 and 16.3 percent. The northern and eastern cities, Arlington, Stanwood, Monroe and Sultan, grew between 6.2 and 45.4 percent and are anchored to local employment rather than to Seattle. Deciding which of those two counties a business is actually selling into changes the brief more than any regional generalisation, and both facts are free to look up.

We treat the reported 28.7 percent fall in other services as unexplained rather than as a finding. A category that broad can move sharply because of reclassification as easily as because jobs vanished, and the profile offers no cause. Our practice in that situation is to report the published figure, say plainly that we do not know why it moved, and decline to build a recommendation on it. Presenting an unexplained statistic as evidence for a marketing conclusion is how a plausible-sounding plan ends up resting on nothing.

Our judgement on sequencing here differs from the neighbouring counties. In Pierce we would attack mobile conversion first; in this county the binding constraint is more often service-area clarity, because the geography is long and a prospective customer forty miles away genuinely cannot tell whether they are covered. In the fastest-growing towns the specific version of that problem is a site built for a smaller version of the town, naming one city when the business now serves four. That is a copy problem, it is cheap to fix, and it is usually the highest-return change available.

Common questions