Data Staffing in Salt Lake City. Both Sides Are Building the Same Middle.
Most Wasatch Front data teams aren’t short a seat. They’re grading toward each other from opposite ends and paying for the middle twice. KORE1 places data engineers, analysts, scientists, architects, analytics engineers and governance specialists across the Salt Lake Valley and Silicon Slopes.
KORE1 is a data staffing agency serving Salt Lake City and the Wasatch Front, placing data engineers, analysts, scientists, architects, analytics engineers and governance specialists on contract, contract-to-hire and direct hire. Our data desk averages 17 days to first qualified submit, and 92% of placements are still in seat at twelve months.
Last updated: August 25, 2026
In 1864 Congress told the Union Pacific and the Central Pacific they could grade up to 300 miles ahead of their own end of track. It paid both of them by the mile. It did not say where they were supposed to meet.
You can guess what happened.
The crews reached northern Utah. And kept going. They graded roadbeds past one another, in sight of each other for miles, both drawing the subsidy, neither one laying a rail on ground the other had already prepared. The National Park Service puts the end of it at April 10, 1869, when Congress finally named Promontory Summit and made the two of them stop.
Nobody set out to build the same fifty miles twice. It’s just what happens when the junction isn’t named.
Our IT staffing practice has been running since 2005, and the Salt Lake data desk runs into the 1869 problem constantly. Not a missing seat. Two halves of a team, both funded, both busy, both building toward a middle nobody owns. Nobody planned it.
The platform side builds a revenue model. Finance keeps its own workbook, because the model came six months late and the board meeting didn’t move. Both are maintained. Both are wrong. Just in different months. The req that lands on our desk asks for a second data engineer, and that’s usually the one hire that makes the overlap wider rather than shorter, because a second builder on a route that already has two crews working it does not shorten anything, it just adds a third grade for somebody to maintain.
Two things are deliberately not on this page. Anything that builds models belongs on AI and ML engineer staffing. Software, infrastructure and the wider technical bench belong on IT staffing in Salt Lake City. Everything below is data.
Six Places Both Sides Built the Same Middle
Six things every data organization has to have. The left bar is what the platform side built. The right bar is what the analytics and business side built. Where they cover the same ground, somebody paid twice. One track came out clean. Just one, and the reason is the whole argument.
Revenue and bookings
W A revenue model in the warehouse, tested, versioned, six months late.
E The workbook finance built in month two and never stopped using.
Who counts as one customer
W Match and merge rules in the pipeline, owned by data engineering.
E A second set of merge rules inside the BI tool, owned by whoever needed the report.
Whether the numbers are current
W Freshness monitors and failure alerts on every job.
E An analyst who checks a timestamp by hand every morning before the stand-up.
Who can see which rows
W Grants and row-level policies in the warehouse.
E Sharing rules and hidden tabs in the reporting layer, which is where the real answer lives.
What the metric actually means
W Definitions in the repo, accurate, read by four people.
E A wiki page, out of date, read by everybody.
Regulatory and audit reporting
W One definition, fixed early, signed by a named person.
E The same definition, because there was never a second one to choose from.
Look at what the clean track has in common with the other five. Same people. Same tools. Same budget. Same quarter. The only difference is that somebody was legally required to name an owner before anyone started building, so the junction got fixed in week one instead of in year three. That’s the seat we’re usually being asked to fill, and it’s almost never the seat on the req.
Six Seats, and Which Two Keep Colliding
Salt Lake postings mix these up as much as any market we work. The two in the middle are the ones that end up building the same thing.
Data engineer
Everything upstream of the first question anyone asks. Extracts off core systems, claims feeds, point-of-sale, telemetry, the jobs that have to finish before the 6am refresh means anything. Nothing downstream works without it. At real volume it splits in two, and the heavier half is a big data engineer search.
Analytics engineer
This is the junction seat. One definition of a customer, an order, a claim, plus a test that fails loudly when somebody quietly changes it. Post it as an analyst and you get an analyst, and the middle stays double-graded.
Data architect
Platform choice, storage tiers, retention, and whose cost center the compute bill lands in. Somebody has to decide. Also the person who can tell you whether the overlap is worth unwinding or worth living with. On a big enough estate one person stops covering it, and the modeling half belongs to a data warehouse engineer.
Data analyst
The other half. Denial rates, churn, throughput, shrink, cost per acquisition. Good ones build their own extracts when the platform is slow, which is rational and which is also how the second grade gets started. When the interface itself is the deliverable, that’s a visualization engineer instead.
Data scientist
Risk, forecast, fraud, propensity, anything where the number has to be produced rather than looked up. Sits downstream of everything above. Give one a double-graded middle and you get two confident answers. Both defensible.
Data governance analyst
Who is allowed to know what, and who signed for it. Boring until it isn’t. In regulated Utah shops this seat is the reason STA 06 came out clean. At enterprise scale the same problem arrives as a chief data officer search.
We Find Out What Already Got Built Twice
It’s one extra call. It happens before we source anybody.
We ask both sides the same six questions. Platform lead and reporting lead, separately, twenty minutes each. Where the two answers disagree about who owns a definition, that’s your overlap, and it’s usually wider than either of them thinks, because each side has spent years quietly absorbing the other’s gaps and neither one still remembers which parts were only ever supposed to be temporary. About a third of the Salt Lake reqs we take change shape after that pair of calls.
We put the overlap on paper before we write the search. Not a maturity assessment. A list of the things that exist twice, in the order they cost you money. Nothing fancy. One Draper client looked at theirs and cancelled the second analyst req they’d been carrying since January.
We say so when the req is a third grade. Hiring another builder into an unnamed junction doesn’t shorten the overlap, it lengthens it, and we’d rather lose the placement than take a fee for making that worse. It costs us searches. We keep doing it, because the ones we do run land and stay landed, which is what we get measured on.
No framework here. It’s forty extra minutes on the phone that most desks won’t spend, because a fast submit scores better internally than a right one.
The Tech Center of This Metro Is Thirty Miles South of It
Salt Lake City is the state capital, the health systems, the banks and the airport. The software payroll is mostly in Utah County. Two markets, one freeway.
That gap is the first thing to settle on a data search here, and employers underestimate it constantly. Adobe’s campus is in Lehi. Ancestry is in Lehi. Qualtrics is in Provo. Domo, which is a data platform company, is in American Fork. A downtown Salt Lake req and a Lehi req are competing for overlapping candidate pools separated by a stretch of I-15 that behaves very differently at 8am than it does on a map.
Downtown holds its own on the data side, and it holds a specific kind. Intermountain Health and University of Utah Health sit on most of the clinical and claims data in the Intermountain West, which is healthcare IT and revenue cycle work as much as it is data work. Zions Bancorporation is headquartered downtown and Goldman Sachs runs one of its largest US offices a few blocks away, which is why an analyst who can survive an audit prices differently in this city than the same resume does elsewhere. That is banking IT staffing territory as much as it is data work. Recursion is downtown. Genuinely heavy compute. Health Catalyst sits in South Jordan and has been selling healthcare data warehousing longer than most of the market has been buying it.
North of the city the picture changes again. Hill Air Force Base and the defense contractors around Ogden and Roy carry cleared work, and that population moves on its own timeline and its own pay scale. The Kem C. Gardner Policy Institute at the University of Utah publishes the honest version of the wider employment picture if you want the numbers rather than the pitch.
Practically, a Utah data resume is easy to misread. Someone out of a health system arrives careful, documented and slow to commit, because in that world a number that turns out to be wrong is a regulatory problem rather than an awkward meeting. Someone out of a Silicon Slopes SaaS company arrives fast and casual about definitions. Neither is a defect. They were graded for different work. Ask which you need.
One Valley, One Freeway, Five Hiring Markets
The Wasatch Front is about eighty miles of metro strung along a single interstate with mountains on one side and a lake on the other. Where your office sits on that line is a hiring input, not a detail.
Downtown Salt Lake City
Banking, health systems, state government, and the newest platform work in the city proper. Best transit access in the state, and the likeliest place to find a stack that was stood up inside the last two years. Transit actually matters here.
Lehi, American Fork & the Point of the Mountain
Silicon Slopes proper. Deepest modern data stack bench in the state and the most competitive offers. Also the submarket where a candidate is most likely to already have three conversations running. Assume competition.
Draper, Sandy & the south valley
The compromise address. A genuinely good one. Reachable from both ends of the metro, which widens your pool more than any single perk you could add to the offer.
Ogden, Roy, Clearfield & Davis County
Defense, aerospace and the cleared population around Hill Air Force Base. Longest timelines on this list and the least elastic pool. Plan for it. If this is your pool, say so in week one, because nothing about it compresses later.
Provo, Orem & Utah Valley
BYU and UVU feed it, so the junior end is deeper and cheaper than anywhere else in the region. Senior scarcity is real here. A first data hire out of this pool needs somebody above them who has already done it once, and hiring two juniors into an empty room because the rate looked good is the most expensive saving available anywhere in this market.
Settle the commute before you settle the comp band. Do it first. A Lehi candidate weighing a downtown seat is weighing an hour each way in winter, and they will not raise it until the offer is in front of them. Park City is close enough that plenty of senior people live up there and will not drive down past the mouth of the canyon in February for anything less than a genuinely serious number, which is worth knowing before you build a band around a downtown address. We staff 30+ U.S. metros. If the right person for a Salt Lake seat turns out to be in Boise or Denver, you’ll hear that from us in week one with the relocation figure attached.
Nobody Here Is Trying to Leave Utah
People stay. That’s the good news, and it’s also the thing that quietly breaks searches.
Retention in this market is unusually strong, and not because employers are unusually good. People move here on purpose. Canyons forty minutes from a downtown desk, a housing math that still works next to the coasts, and in a lot of cases family within an hour. A senior data engineer in Sugar House is not idly browsing. The ordinary levers that pry someone loose in Dallas or Atlanta do very little here.
So the competition isn’t the company across the street. It’s a fully remote seat at a coastal employer paying a coastal band, taken without changing an address, a school or a ski pass. That offer never appears in a local comp survey. It never appears in a local comp survey, it never comes up in a screening call, and it is the single most common reason a Salt Lake search that looked healthy in week three goes quiet in week six.
Two things worth deciding before you build the offer. Three days onsite is what this market has settled on, and the big health systems and banks are the least willing to move off it. And contract-to-hire converts well here, better than it does in the Northeast markets on our desk, partly because the contractor population is large and nobody in Utah reads a contract start as a demotion. If you want to calibrate before we talk, the salary benchmark tool is open.
Three Ways to Buy the Seat
Pick on how settled the work is, not on which budget line has room this quarter.
Contract & Contract-to-Hire
You direct the work, we carry the employment, normally across a three to nine month window. Right call when the consolidation has to happen before anyone can honestly size a permanent seat. It usually converts.
Contract Staffing →Direct Hire
Junction seats belong here. Whoever gets to decide what a customer means should still be around in three years to defend it, and that is not a job you want re-explaining itself to a new contractor every nine months.
Direct Hire details →Project & Statement of Work
A migration, a lineage cleanup, or unwinding an overlap that’s been running for four years. Fixed end date. We scope it, staff it, and hand it back on a date you can commit to in a plan.
Project Staffing →Common Questions
What does it cost to hire a data engineer in Salt Lake City?
Salt Lake runs below the coastal metros and above the interior mountain West, and the spread inside the metro is wider than the gap between metros. Two things move it.
The first is which end of I-15 you’re hiring against, because a Lehi offer and a downtown offer are not priced the same even when the job is identical. The second is ownership. Whether the person has ever carried a platform outright, budget and roadmap included, rather than working inside one somebody else paid for. We put a current band in front of you on the first call instead of publishing one that ages badly in six weeks. The BLS Occupational Outlook Handbook is the cleanest public baseline for the occupation itself if you want a national floor to argue from.
Should I be recruiting in Salt Lake City or in Silicon Slopes?
Both, and they behave like two markets that happen to share one freeway. Downtown skews health, finance and government data. Lehi through Provo skews product and SaaS data. Candidates cross the gap, but not casually and not for a lateral move.
The practical rule is to name your actual office address early rather than writing “Salt Lake City area” and hoping. Someone in Orem reads a downtown address as a real life change. Someone in Bountiful reads a Lehi address the same way. Name the office. Getting that on the table in the first conversation removes the single most common week-six surprise in this market.
How long does KORE1 take to fill a data role in Salt Lake City?
17 days to first qualified submit is our desk average, and Salt Lake analytics and engineering searches usually land near it. Architect searches take four to six weeks. Cleared roles around Hill Air Force Base are their own category and nothing about them is quick.
We would rather be judged on retention, and ours is 92% at twelve months. A quick submit is worth nothing if the person is gone by spring. It’s rarely sourcing. It’s a hybrid policy or a commute that surfaces in week five instead of week one, and each of those resets the clock, because everyone already in flight was measured against a different job and has to be looked at again.
My team keeps arguing about which revenue number is right. Is that a hiring problem?
Usually yes, but not the hire you think. Two numbers means two definitions, which means two teams built the same middle. Adding a builder to either side widens the overlap. The seat that closes it is normally an analytics engineer or a governance owner.
We’d rather find that out on the first call than at month four. It takes about twenty minutes with each side and the answers rarely agree, which is the useful part, because the gap between what the platform lead believes they own and what the reporting lead believes they own is exactly the ground that got graded twice. If the overlap turns out to be narrow and the real gap is capacity after all, we’ll say that too and run the search you originally asked for.
Are Salt Lake data roles remote, hybrid, or onsite?
Three days in the office is the default here, and the health systems and the banks are the slowest to bend on it. Fully remote is now reserved for platform and architecture seats nobody can fill locally. Cleared defense work is onsite. No exceptions.
State the policy on the first call, and state the office with it. The office part carries more weight here than almost anywhere, because this metro is eighty miles end to end. A Provo candidate hears “hybrid in Salt Lake” very differently than a Bountiful candidate does, and neither will tell you so until the offer stage.
Does the cleared work around Hill Air Force Base change a data search?
Completely, if you actually need it, because an active clearance narrows the Utah data pool by an order of magnitude and carries a premium that has nothing to do with the technical work. Most reqs don’t need one. A surprising share ask for a clearance the job never uses.
Write the real requirement instead of “clearance preferred,” which filters out strong uncleared people while attracting no cleared ones. Write the real requirement. And separate active from previously held, because those are two different candidate pools with two different timelines and two different rates. Blurring them is how a Davis County search quietly loses a month.
Do people relocate to Utah for data roles?
More than they used to, and the pitch mostly writes itself. The harder problem runs the other way, which is keeping the senior people already here from taking a fully remote coastal seat without moving at all.
Inbound candidates are usually leaving California or the Northeast with a partner, a mortgage calculation and an outdoor hobby, and they close on arithmetic plus lifestyle rather than on the role alone. Ask in week one. A Salt Lake search with relocation attached is a different search, and it’s much better to know that early than to find out after a finalist asks.
Which data platforms come up most in Utah?
Snowflake leads, Databricks is second, and AWS sits underneath both more often than Azure does, although Azure runs deep inside the health systems. dbt has become the default modeling layer, Power BI and Tableau split reporting, and Domo shows up here far more than its national share would suggest, which is what happens when a business intelligence company is headquartered forty minutes down the freeway and half the analysts in the valley have either worked there or interviewed there.
Two local pockets are worth naming in a req if you need them. Health systems in the Intermountain West still run deep Epic and Cerner reporting estates, and older manufacturing and distribution shops along the Wasatch Front carry real legacy ETL work, which is usually an ETL developer search rather than a modern data engineering one. Platform-specific searches get pulled from our national Snowflake recruiters and Databricks recruiters desks, which we lean on for Utah reqs constantly.
Send us the req and we’ll tell you which side already built it.
One call with your platform lead, one with your reporting lead, and a short list of everything that exists twice. You get that whether or not you hire us.
Talk to a Data Recruiter →