Highly Scriptable: What the Pest Control Labour Market Says About the Work

1. Introduction: a claim made to buyers

This journal's article on service contract economics quoted an acquisition advisory firm explaining the sector to people buying pest control businesses. One clause in that explanation deserved its own article.

1.1 Why it matters commercially

A scriptable job can be done by a cheaper worker after a shorter training period, and the consequences of turnover are small. That is a statement about labour cost, made to somebody deciding what to pay for a business.

1.2 What this article does

Tests the claim against the labour data the same industry publishes, and then asks why, if the work is not scriptable, the market does not pay for the difference. Sections 14 to 16 are the answer.

2. What was claimed

The context, restated briefly.

The acquisition literature this journal examined listed four features buyers underwrite: subscription-like recurring revenue, retention above ninety per cent annually, demand inelasticity, and a regulatory moat from licensing and applicator certification. Scriptability appeared in the summary of why the sector is attractive.

2.1 Why we are testing it rather than dismissing it

Because it is partly true. A great deal of routine pest control work is genuinely procedural: a perimeter application, a bait station service, a monitor change. A competent script covers those and a new technician can follow one on their second day.

The question is whether that describes the whole job, and §19 is our answer.

2.2 The implicit argument

If the work is scriptable, then the technician is a labour input rather than a professional, the business does not depend on any individual, and a buyer is purchasing a route rather than a team. That is a coherent investment thesis and §§3 to 9 test whether it holds.

3. The turnover figures

What the trade reports about keeping people.

A recruiting analytics source states that turnover rates in field services are hovering near 40%.³

A training platform reports that an industry publication found that 18% of pest control companies had a retention rate below 79% and reported struggling with staff turnover.⁴

3.1 Reading those together

Forty per cent annual turnover across field services, with roughly one pest control company in five losing more than a fifth of its staff each year.³⁴

Both figures come from vendors selling retention products and are flagged accordingly. They are the only figures we found.

3.2 What forty per cent means in practice

A firm with ten technicians replaces four of them a year, which means that at any moment a substantial share of its field staff is inside the ninety-day window described in §4.³

The stable, experienced team that the retention material describes as the goal is, on those figures, not what most operations actually have. That arithmetic is ours.

3.3 The demand side

The same training source notes that demand for pest control services is growing, with 5.2% annual growth expected over the next five years, while finding and keeping skilled technicians is becoming harder and harder.⁴

4. The first ninety days

Where the departures concentrate.

The same source reports that 60% of turnover happens in the first 90 days, describing that period as a retention danger zone.³

4.1 What that pattern means

Most of the people who leave do so before they have become useful. The firm has paid for recruitment, onboarding and initial training and has recovered none of it.

4.2 And it implies something about the job

A new technician's first ninety days are the period of lowest competence and highest supervision requirement. If most departures happen then, the firm is repeatedly paying for the hardest part of the employment relationship and rarely reaching the productive part.

That reading is ours.

5. What replacement is said to cost

The estimates, which do not agree.

A trade publication reporting an industry cost study quotes the position that some estimates put the cost of replacing a technician at 50-150% of their annual salary.¹

A training platform states that it can cost up to six to nine month's salary to replace an employee.⁴

A retention software vendor asks whether readers know that replacing a single service worker can cost up to two times their annual salary.⁵

A hiring platform states that according to industry research, replacing a single front-line worker can cost you nearly $6,000.⁶

6. The spread is the finding

Because those four numbers cannot all be describing the same quantity.

Six thousand dollars against two hundred per cent of a salary reported elsewhere at forty-four thousand seven hundred and thirty is a difference of roughly fifteen-fold.⁶⁵²

6.1 What is probably going on

The low figure is likely direct hiring cost: advertising, screening and administration. The high figure includes lost productivity, supervision, training and the revenue consequences in §20. Nobody states which they are measuring.

That inference is ours and it is the charitable reading. The uncharitable one is that each vendor quotes the figure that best supports what it is selling.

6.2 Either way the direction is consistent

Every source says replacement is expensive, and none says it is cheap.

7. What the cost consists of

The components, as the industry describes them.

When a technician leaves, you're not just replacing a salary: the firm is losing productivity during the vacancy, spending time and money on recruitment, investing in training for the replacement and often seeing service quality dip during the transition.¹

7.1 The vacancy itself

Lost productivity during the vacancy is the item most likely to be underestimated, because it does not appear as a payment. A route still has to be served, so either another technician absorbs it, working longer or faster, or visits are deferred.

Both outcomes degrade something. Our article on service economics reported a revenue per technician target of 150,000 to 200,000 dollars annually, which constrains how much slack exists to absorb a vacancy at all.

7.2 Four items, of which one is not a cost to the firm

Lost productivity, recruitment and training are borne by the employer. A dip in service quality is borne by the customer.

Which is the observation this article turns on, and it arrives in §16.

7.3 And the ramp-up

An advisory source notes that new team members, no matter how qualified, take time to get up to speed with your operations and customer base.⁷

No matter how qualified. A statement that formal qualification does not substitute for experience.⁷

8. Service quality dips during the transition

Isolating the sentence.

The industry's own cost study commentary states it as a matter of course, alongside the other consequences of losing a technician.¹

8.1 What it presupposes

That service quality is a property of the individual performing it rather than of the procedure being followed.

If the work were fully specified by a script, a replacement following the same script would produce the same result, and there would be no dip to report.

9. Which contradicts the claim

Putting it together.

Three statements from the trade's own labour literature are each incompatible with scriptability: that quality falls during transition, that new hires take time regardless of qualification, and that replacement costs a substantial fraction of a salary rather than the cost of printing a procedure.¹⁷

9.1 The technical term

What cannot be written down and must be acquired by doing is tacit knowledge, and its presence is exactly what makes replacement expensive.

9.2 Which of the two audiences is being told the truth

The scriptability claim is made to buyers, who want labour costs to be low and staff to be interchangeable. The ramp-up and quality-dip statements are made to operators, who want to be sold retention software.

Both audiences are being told what they want to hear, and the operator-facing claims are the ones consistent with everything else in this journal. That assessment is ours.

10. The wage comparison

What the work pays against comparable licensed trades.

A marketing source citing a national labour statistics agency reports that the median annual wage for pest control workers was $44,730 in May 2024, and that this puts pest control at a significant disadvantage compared to competing trades like electrical work ($62,350 median) and plumbing ($62,970 median).²

10.1 Roughly eighteen thousand dollars

Against two trades with comparable licensing requirements, comparable vehicle and tool costs, and comparable physical demands.²

10.2 What the gap does to recruitment

A candidate choosing between licensed trades is choosing between comparable entry requirements and a wage difference of roughly forty per cent. The shortage in §11 is not mysterious.²

10.3 The obvious question

Why the gap exists. We did not find an explanation in any source and will not invent one. We note that a trade whose quality is unverifiable, per §16, has less scope to compete on quality and therefore less scope to pay for it.

That is a hypothesis, flagged as ours, and we have not tested it.

11. The shortage

The demand side of the same market.

The same source reports that the technician shortage consistently ranks as the top challenge and that the statistics agency projects 13,400 annual openings for pest control workers.²

11.1 Openings against a wage gap

Thirteen thousand four hundred positions a year in a trade paying eighteen thousand dollars less than the alternatives a candidate with similar aptitudes could enter.²

11.2 The recruiting friction

A recruiting source reports that 61% of candidates ghost after an interview, that 76% of recruiters report being ghosted by candidates at some stage in the funnel, and that candidates are applying to 10 jobs via Easy Apply buttons in 60 seconds.³

That source sells recruiting software and every figure in it serves that sale. We report the figures as claims.

12. And the response

What firms say they are doing.

A 2025 analysis is reported to have found that 89% of pest control companies planned to raise technician wages to stay competitive.³

12.1 Eighty-nine per cent

Which is close to everybody, and it tells you the shortage is being felt rather than anticipated.³

12.2 Which is the correct response to a shortage

And it is a cost increase that has to be recovered somewhere, which our article on service economics describes as landing on a subscription price the customer cannot evaluate against quality.

13. Direct labour as a share of revenue

The number that sizes the problem.

An industry cost study is reported to show that direct labor typically represents 26% of revenue.¹

13.1 A quarter of the top line

Which is large enough that a labour cost increase cannot be absorbed and small enough that it is not the dominant expense.¹

13.2 What it does not include

Direct labour is the technician's wage against revenue. The turnover costs in §5, the supervision in §4 and the quality consequences in §8 sit elsewhere in the accounts or nowhere at all.¹

Which is why a cost study can report a labour line that looks manageable while the same publication describes turnover as a hidden cost. Both statements are true of different numbers, and only one of them appears on a profit and loss account.

13.3 And it explains the scriptability claim's appeal

If a quarter of revenue is labour, then a buyer wants to believe that labour is substitutable. That is not evidence that it is.

14. The training investment problem

The economics underneath all of this, which is standard and worth stating plainly.

Skills divide into those useful only at the firm that taught them and those useful anywhere. A firm can profitably invest in the first kind, because the worker cannot take the value elsewhere. It struggles to invest in the second, because a competitor can hire the trained worker and capture the return without paying the cost.

14.1 Pest control training is almost entirely the second kind

Pest identification, inspection technique, application method, label interpretation and diagnosis are portable to any competitor in the province. Nothing about them is specific to one employer.

That classification is ours, applied to this trade from standard labour economics.

14.2 The exception worth naming

Some knowledge genuinely is firm-specific: the account history, the building's layout, the client's tolerances, where the previous technician found activity. That is worth investing in and it cannot be taken to a competitor.

But it is also the knowledge lost fastest when somebody leaves, and no training programme creates it. It accumulates only through continuity, which is §21.1.

14.3 Which produces a predictable result

Every firm would prefer trained technicians and would prefer somebody else to have trained them. The industry association response described in one source is a workforce development programme helping members attract, train and retain quality technicians more effectively, with members building stronger recruitment pipelines through partnerships with technical schools and community programs.¹

Which is the textbook remedy: move the training cost to a collective body or to an institution outside the firm.

15. Why the usual solution is blocked

The other standard remedy, and why it does not work here.

Where a firm cannot capture the return on general training, the worker can pay for it instead, by accepting lower wages during the training period in exchange for a skill worth more afterwards.

15.1 That requires headroom

A wage that can fall during training and rise afterwards. Section 10 reports a median already eighteen thousand dollars below two comparable trades, and §11 reports a shortage that is pushing entry wages up rather than down.²³

15.2 So neither party can finance it

The firm cannot recover the investment because the worker leaves. The worker cannot fund it through reduced pay because the starting wage is already uncompetitive against the alternatives.

That is our analysis and it is the cleanest explanation we can construct for why a trade that everybody agrees needs skilled technicians has a forty per cent turnover rate.

16. Why the market does not correct it

And the mechanism that makes the whole thing stable.

This journal's article on pest control as a credence good established that the buyer cannot verify service quality before, during or frequently after the work. Our detection article established that finding nothing is not evidence that nothing is there.

16.1 Apply that to the quality dip in §8

A customer receiving degraded service from a new technician during a transition has no way to detect it. The visit happened, the invoice arrived, no insects were reported. Our service economics article described exactly that situation.

16.2 Which removes both halves of the market signal

A firm that skips training is not punished, because the customer cannot see the difference. A firm that invests cannot charge a premium, because it cannot demonstrate the difference.

Underinvestment in an unobservable quality is not a failure of diligence by anybody. It is what the market structure produces, and that framing is ours.

16.3 Which explains something this journal keeps finding

Our articles on remote monitoring, on detection dogs, on automated identification and on certification all found the same pattern: claims running ahead of validation, and nobody with an incentive to close the gap.

Section 16.2 is the general form of that finding. Where the buyer cannot check, nothing in the market requires the claim to be true, and the phenomenon is not specific to any one technology.

16.4 The one exception

Failures that become visible. A bed bug treatment that does not work, a rodent programme that fails an audit, a callback. Our credence goods article noted that the credence property weakens where the outcome eventually declares itself.

Which means the market does discipline the most visible failures and does not discipline anything else, including everything preventive.

17. The certification floor

The intended remedy.

Licensing and applicator certification exist partly to impose a minimum competence where the market cannot detect it. This journal's article on applicator certification examined the evidence behind that and found it thinner than the trade assumes.

17.1 Why a floor is the natural instrument

Because a regulator can verify that somebody passed an examination and cannot verify that a treatment was correctly diagnosed. Licensing measures what is measurable, which is the same constraint our food safety audit article found in a different system.

17.2 What a floor does and does not do

It excludes the worst. It does not create a gradient above itself, because the customer still cannot distinguish a technician at the floor from one well above it.

So certification addresses the bottom of the distribution and leaves the incentive problem in §16 intact for everybody above it. That reading is ours.

18. And the entry barrier

The same requirement from the buyer's side.

Our service economics article recorded acquisition advisers listing a regulatory moat, on the basis that licensing and applicator certification limit new entrants, among the features that make the sector attractive.

18.1 Both things are true at once

A competence floor that protects the public, and an entry restriction that protects incumbents. The same rule does both and the acquisition literature names only the second because that is what its readers are buying.

18.2 Which is not an argument against licensing

The products this trade applies are capable of harm, and our articles on applicator exposure and on the legal force of the label describe why competence in handling them is not optional. A floor is justified on that ground alone.

The observation is only that a rule justified by public protection also produces a private benefit to incumbents, and that the people who benefit are unlikely to be the best judges of whether the rule is set correctly.

18.3 The uncomfortable combination

Restricted entry, a documented shortage, wages eighteen thousand dollars below comparable trades, and a quality the market cannot price. We are not claiming those facts have a single cause. We are noting that they coexist and that nothing in the structure obviously resolves them.

19. What this journal has spent eighty-nine articles on

The direct test of the scriptability claim.

Is a single ant indoors a forager from an outdoor colony or evidence of a satellite nest in the wall? Is the small fly problem the drain, the mop bucket or something outside? Does a clean inspection mean an empty building or an inadequate inspection? Is the client's skin complaint a bed bug, a bird mite after host removal, or neither?

19.1 And each has a documented failure rate

Our article on detection probability put numbers on how often an inspection misses a present population. Our article on canine detection found handler effects. Our article on spider bite misdiagnosis found a whole category of complaint routinely attributed to the wrong cause.

A trade with documented diagnostic failure rates is not a trade whose central task is scripted.

19.2 Each of those is a diagnosis

Each has an article in this journal because each is genuinely difficult, each has a documented failure rate, and each determines whether the subsequent treatment has any chance of working.

19.3 A script can specify a treatment

It cannot specify which treatment, because that depends on a judgement about what is actually happening in a building, made by somebody standing in it.

Our view is that the scriptable claim is true of application and false of diagnosis, and that selling the second as though it were the first is the specific error. That is our position and it is what the rest of this journal is evidence for.

20. The customer attrition point

One consequence the trade sources identify that we had not considered.

An advisory source describes that perhaps the most significant risk associated with high employee turnover is customer attrition.⁷

A retention source adds that customers value relationships with familiar faces and that when technicians stay, they build trust with clients, which translates into repeat business and positive reviews.⁵

20.1 Which is a credence good result

If quality cannot be verified, the customer substitutes something they can observe, and familiarity is the obvious candidate. Our service economics article found acquisition advisers describing stickiness through habit, scheduling and relationship rather than through results.

20.2 So the trained technician is valuable for the wrong reason

Retention is rewarded because the customer recognises the person, not because the person is better. Both effects point the same way, which is fortunate, and only one of them is about competence.

That inference is ours.

21. What follows for a firm

The practical implications, stated without pretending they are easy.

The first ninety days are where the money goes. Sixty per cent of departures occur there, so supervision and structure in that window return more than anything later.³

Training is a cost that cannot be recovered through price. Section 16 is why, and a firm that invests should do so knowing the customer will not pay a premium for it.

Which means it has to be justified internally. Through fewer callbacks, fewer failed treatments and fewer escalations, all of which are measurable by the firm even when they are invisible to the client.

And the collective route is the rational one. Industry training programmes and partnerships with technical schools move the cost off the individual firm.¹

21.1 The one thing a firm can do that the market will reward

Keep the same technician on the same accounts. Section 20 establishes that continuity is the quality proxy customers actually respond to, so the return on retention is capturable in a way the return on training is not.

Which is a slightly dispiriting conclusion, because it means the commercially rational investment is in familiarity rather than in skill. Both follow from retaining people, which is why the advice converges even though the reasons do not.

22. What follows for a client

Three questions, since the market will not answer them for you.

How long has the technician been doing this? The trade's own sources say new hires take time to get up to speed regardless of qualification.⁷

Will it be the same person? Continuity is the one quality proxy a customer can actually observe, per §20.

And what did they find, not what did they do? A finding is evidence of diagnosis. An application is evidence of a visit.

None of these questions is hostile and a competent provider will welcome them, because they are the only questions available to somebody who cannot inspect the work itself.

23. Our own position

The disclosure, which in this case is simple.

We are a sole proprietorship. There is no technician turnover to disclose because there are no employed technicians, and the training investment problem in §14 does not apply to us in the form described.

23.1 Why we wrote it anyway

Because the structural argument in §16 applies to every firm in this trade including ours. A customer cannot verify our competence either, and nothing in the market rewards us for having it.

23.2 Which is the reason this journal exists

Publishing the reasoning, with sources attached, is one of the few mechanisms available for making competence visible in a market that cannot otherwise price it. Ninety articles is a slower route than advertising and it is checkable, which advertising is not.

That is a commercial motive and we state it as one.

24. The Manitoba position

What we could and could not establish locally.

Every wage, turnover and cost figure in this article is American. We found no Canadian pest control labour data, no provincial wage comparison against the licensed trades here, and no figures on turnover in this province.

24.1 One thing that probably transfers

The credence good argument in §16 is structural rather than jurisdictional. A Manitoba client can no more verify a treatment than an American one, so the incentive problem does not depend on any of the figures being locally accurate.

That is the part of this article we would defend without the data.

24.2 What is likely to differ

Manitoba has a shorter active season than most of the markets these figures describe, which changes the seasonal employment structure and probably the turnover pattern with it. Our article on overwintering invaders and our degree-day article both describe how compressed the season is here.

That is a reason to doubt the transfer, not a measurement of it.

25. Limitations and open questions

Every source is a vendor. Recruiting software, retention software, a training platform, a hiring platform, a marketing agency, an accounting practice and a trade magazine reporting an association's own cost study. Each sells something to the firms it is describing and each is flagged in the reference list.³⁴⁵⁶

No peer-reviewed literature was located. As with our article on service contract economics, the only published account of this trade's economics is written by people selling into it.

The figures are unverifiable and mutually inconsistent. Section 6 documents a fifteen-fold spread in one quantity, and no source states its methodology.¹⁶

The wage figures are reported at second hand. We have not consulted the labour statistics agency directly and the comparison trades may not be matched for experience or region.²

The human capital argument is textbook, not measured. Sections 14 and 15 apply standard labour economics to this trade. We found no study testing whether pest control firms underinvest in training relative to an optimum, and there may not be one.

And we have no data on whether training improves outcomes. Our certification article found that evidence thin, which means §21's recommendation rests on a plausible mechanism rather than a demonstrated effect.

Sections 1.2, 4.2, 6.1, 7.1, 9, 10.2, 13.2, 14.1, 15, 16, 17.1, 18, 19, 20.2, 21, 22 and 23 are our reasoning. The test of the scriptability claim, the reading of the cost spread, the human capital analysis, the credence good explanation for underinvestment, the certification floor argument, the diagnosis argument and the practical recommendations are ours rather than sourced positions.

26. Conclusion

An acquisition advisory firm told buyers that the underlying technical work in this sector is highly scriptable. The same industry's labour literature reports that service quality dips during a transition, that new team members take time to get up to speed no matter how qualified, that turnover in field services runs near forty per cent with sixty per cent of it in the first ninety days, and that replacing one technician costs somewhere between six thousand dollars and twice an annual salary depending on which vendor you ask.¹⁷³⁵ Those are descriptions of work that lives in a person rather than in a procedure.

The wage sits at forty-four thousand seven hundred and thirty against roughly sixty-two thousand for electrical and plumbing work, with thirteen thousand four hundred annual openings and eighty-nine per cent of firms planning increases.²³ The skills are general rather than firm-specific, so the employer who trains cannot capture the return; and the wage has no headroom below it, so the worker cannot finance the training either. Both of the standard solutions to the training problem are closed.

What keeps it stable is the property this journal identified thirty articles ago. Pest control quality is unverifiable by the person buying it, so a firm that skips training is not punished and a firm that invests cannot charge for it. Underinvestment in an unobservable quality is not carelessness; it is what this market structure produces. Which is why the scriptability claim is believed: it is false about diagnosis, true about application, and nobody in the transaction is in a position to tell the difference.

References

1. Profit by the numbers, trade magazine article reporting an industry association cost study. Trade publication reporting an association's own commissioned study, with the underlying dataset available to clients of the accounting practice that produced it; cited as attributed material. Used for the report that data show direct labour typically represents 26 per cent of revenue; for the quoted account that when a technician is lost the firm is not just replacing a salary but losing productivity during the vacancy, spending time and money on recruitment, investing in training for the replacement and often seeing service quality dip during the transition; for the quoted estimate that some estimates put the cost of replacing a technician at 50 to 150 per cent of their annual salary; and for the description of an association workforce development programme helping members attract, train and retain quality technicians more effectively, with members building stronger recruitment pipelines through partnerships with technical schools and community programmes. https://www.pctonline.com/article/profit-by-the-numbers-/
2. Seven critical threats to pest control company profitability, marketing agency article. Commercial source selling marketing services to the firms described, cited as attributed material. Used for the report, attributed to a national labour statistics agency, that the median annual wage for pest control workers was 44,730 dollars in May 2024, and that this places the trade at a significant disadvantage against competing trades with medians of 62,350 dollars for electrical work and 62,970 dollars for plumbing; for the statement that the technician shortage consistently ranks as the top challenge and that the agency projects 13,400 annual openings for pest control workers; and for the observation that companies developing competitive compensation packages and strong retention cultures will have an advantage. https://cubecreative.design/blog/pest-control-marketing/7-owner-challenges-solutions
3. Pest control hiring and retention, recruiting technology vendor article. Commercial source selling recruiting software, with every figure serving that sale; cited as attributed material. Used for the statements that turnover rates in field services are hovering near 40 per cent and that 60 per cent of turnover happens in the first 90 days, described as a retention danger period; for the reported figures that 61 per cent of candidates cease contact after an interview, a figure said to have risen 9 per cent since the previous year, that 76 per cent of recruiters report the same at some stage, and that candidates apply to ten jobs in sixty seconds through one-click applications; and for the report that a 2025 analysis found 89 per cent of pest control companies planned to raise technician wages to stay competitive. https://www.nurtureme.ai/blog/pest-control-hiring-retention
4. A guide to pest control technician training and retention, field service software vendor article. Commercial source selling operations software, cited as attributed material. Used for the statement that demand for pest control services is growing with 5.2 per cent annual growth expected over the next five years while finding and keeping skilled technicians is becoming harder; for the report, attributed to a trade publication, that 18 per cent of pest control companies had a retention rate below 79 per cent and reported struggling with staff turnover; for the estimate that it can cost up to six to nine months of salary to replace an employee; and for the argument that thorough technician training leads to fewer mistakes and higher customer satisfaction. https://gorilladesk.com/learn/pest-control-technician-training/
5. Technician retention articles from a workforce engagement software vendor. Commercial source selling retention software, cited as attributed material. Used for the claim that replacing a single service worker can cost up to two times their annual salary; for the description of turnover costs including hiring expenses, job advertisements, recruiter fees and interview time; and for the arguments that experienced technicians know systems and processes and ensure consistent quality, that high turnover disrupts teams and leads to burnout, and that customers value relationships with familiar faces so that retained technicians build trust which translates into repeat business and positive reviews. https://www.applausehq.com/blog/how-retaining-your-home-services-technicians-saves-thousands-vs-hiring
6. How to hire pest control technicians, hiring platform guide. Commercial source selling hiring software, cited as attributed material. Used for the statement that turnover in pest control is expensive and disruptive and that, according to industry research, replacing a single front-line worker can cost nearly 6,000 dollars; and for the recommendations to offer clear career paths and ongoing training on the basis that employees who see a future with the company are less likely to leave. https://www.workstream.us/hire/service-pest-control-technician.html
7. The true costs of employee turnover, accounting and advisory practice article aimed at pest control business owners. Commercial source selling advisory services to the firms described, cited as attributed material. Used for the statement that retaining a strong consistent team is critical and that for pest control companies talented technicians are valuable, while preventing excessive turnover remains a challenge; for the observation that one of the most visible costs is recruiting and training new staff but that such direct costs represent only a small part of the picture; for the statement that new team members, no matter how qualified, take time to get up to speed with operations and the customer base; and for the assessment that perhaps the most significant risk associated with high employee turnover is customer attrition, alongside the note that compensation is a key factor in retention and that staying in line with market rates is essential. https://kempanderson.com/2025/02/14/the-true-costs-of-employee-turnover/