Natives and pollinators
Choosing and placing plants

Native plants and pollinators: what the research supports, and what a garden can do

Each finding stated at the size of its evidence, keystone genera doing most of the work, nesting designed for, bloom overlapped, and the planting laid out so it reads as deliberate.

Start here Essentials

Four decisions, and the one that costs nothing is the one most often skipped. These work for a single bed as well as for a whole yard. Everything marked Go deeper is optional on a first pass.

  1. Look up the keystone genera for your ecoregion and note which ones your space could hold (section 3).
  2. Pick one keystone woody plant your space can hold at mature size, if there is room for one (section 3).
  3. Add three perennials whose bloom months do not overlap, so there is flower across the season (section 5).
  4. Leave one square foot of bare, undisturbed soil and mark it on the plan so it does not get mulched (section 4).

Minimum time: about 45 minutes, mostly looking up lists. Done when: the bloom chart shows flower in at least three separate months, and one patch of bare ground is marked on the plan.

1. What "native" means, and why it is regional Understand

A native plant is one that occurs naturally in a particular region, having evolved there alongside the animals, fungi and microbes that live on it. The word is only meaningful with a place attached. "Native to North America" covers a continent with every climate on it and tells you almost nothing useful about whether a plant belongs in your yard.

The unit that is useful is the ecoregion, an area where the ecosystems and the environmental resources are broadly similar. The National Wildlife Federation organizes its plant lists this way, by Level I ecoregion, and notes that keystone plant genera are unique to local food webs within ecoregions.

There is a second layer, which is where the individual plant came from. The USDA Forest Service's native plant materials program is built around the idea of local ecotypes: populations of a species that have adapted to local conditions, so that seed of the same species collected two states away may behave differently in your garden and may be less use to the insects that depend on it. Convention It matters most for seed and for restoration work and least for a single shrub in a suburban bed, but it is the reason that "where was this grown" is a reasonable question to ask a nursery.

Two things native does not mean, and both are worth saying before anything else in this chapter:

  • Native does not mean no maintenance. It describes where a plant evolved, not how much attention it wants in a yard that has been graded, compacted and built on.
  • Native does not mean no watering. Establishment watering applies to native plants exactly as it does to everything else, and Planting and the first three years has the schedule.

2. What the research actually says

This is the section the chapter is built around, because the gap between what the research shows and what gets said about it is wide, and the research is more useful than the slogan.

The chickadee study. You will meet a figure, usually stated as "yards need to be 70 percent native," and it comes from one study. Desirée Narango, Doug Tallamy and Peter Marra published work on Carolina chickadees in residential yards in the Washington, D.C. area, reported by the University of Delaware. What the study found, as the university reports it: chickadees needed yards above roughly 70 percent native plant biomass to sustain population growth, and below that threshold the birds switched prey, laid fewer eggs or skipped breeding, so that such properties acted as population sinks.

Every part of that sentence is load-bearing. It is one bird species, the Carolina chickadee. It is one metro area. The threshold is about woody plant biomass, which is the mass of trees and shrubs, not a count of perennials or a percentage of the plants in your border. Convention This manual cites the University of Delaware's report of the study rather than the paper itself, because the publisher's site refused the request when this manual was researched and nothing here is quoted from a paper that was not read. If you want the paper, it is Narango, Tallamy and Marra, "Nonnative plants reduce population growth of an insectivorous bird," in the Proceedings of the National Academy of Sciences, 2018.

What that study supports: planting native woody plants, in quantity, increases the food available to insect-eating birds in a yard, and the effect is large enough to matter to breeding success for at least one species in at least one region. What it does not support: a claim that your garden will bring birds back, a 70 percent target for perennials, or a population-level effect from one lot.

The keystone figures. The National Wildlife Federation's keystone plants page carries three figures, each attributed on the page to named researchers:

  • 96 percent of terrestrial birds rely on insects for food sources.
  • Research by Doug Tallamy and his team at the University of Delaware has identified that 14 percent of native plants, the keystones, support 90 percent of butterfly and moth species.
  • Research by horticulturist Jarrod Fowler has shown that 15 to 60 percent of North American native bee species are pollen specialists, eating pollen from only about 40 percent of native plants.

Those are reported by a conservation organization, with the researchers named, and they are the basis for the ecoregional lists discussed in the next section. The 15-to-60-percent range is worth noticing for its width: it is a range across regions and groups rather than a single figure, and anybody quoting it as "60 percent of native bees are specialists" has dropped the lower bound.

Limits

What a garden does is add forage and nesting in one place. The Xerces Society's framing of a pollinator garden is in exactly those terms: flowers through the growing season, shelter and nesting sites, protection from pesticides. No study read for this manual supports a claim that one garden changes a population, and this manual does not make one. That is not a reason to do less. It is a reason to say what you are doing accurately, which also happens to be the version that tells you what to plant.

3. Keystone genera: a few plants do most of the work

The practical consequence of the 14-percent figure is the most useful idea in the chapter: a small number of plant genera support a disproportionate share of the insects, so choosing from those genera buys more than choosing at random from a list of natives.

The National Wildlife Federation publishes keystone plant lists by ecoregion, summarized by Level I ecoregion along with the top plants used by pollen-specialist bees. Homegrown National Park, an advocacy organization founded by Doug Tallamy, publishes a more detailed keystone list by Level II ecoregion. Convention Both are advocacy sources rather than peer-reviewed publications, which is not a reason to ignore them; it is a reason to say what they are when you repeat them.

How to use a keystone list in a small garden:

  1. Find your ecoregion and open the list.
  2. Read the woody genera first. The biggest contributors are usually trees, and a tree is the single highest-value plant you can put in a yard that has room for one.
  3. Apply the filters from Choosing plants that fit the place. Zone, band, soil, drainage, and above all mature size. A keystone oak is a wonderful thing and does not belong in a 12-by-4-foot strip.
  4. If no keystone tree fits, take a keystone shrub. If nothing woody fits, take the perennial genera on the list, which usually include the ones that feed specialist bees.
  5. Put the genus on the plant list with what it feeds in its own column, so the reason survives.

The honest version of this for a small lot: one or two keystone woody plants, chosen for mature size, plus perennials that cover the season, is most of the available benefit. A yard cannot hold an ecosystem. It can hold the plants that the local insects evolved on, which is a different and achievable thing.

Go deeper

Level I against Level II lists, and why they differ

Ecoregions are nested. Level I divides North America into a small number of very large regions; Level II subdivides those. The National Wildlife Federation summarizes its data by Level I, which gives broad lists that apply across a wide area; Homegrown National Park's Level II lists are finer-grained and will sometimes name different genera for the part of a Level I region you are actually in. Neither is wrong. If the two disagree about your area, the finer-grained list is usually the more specific advice and your own extension service or native plant society is the tiebreaker, because they are looking at your state rather than at a continental dataset.

What the chickadee paper measured, and what it did not

Worth being clear about, because this figure gets stretched. The measure was woody plant biomass, which is the mass of trees and shrubs on a property, not the number of plants and not the share of the perennial border. The outcome measured was the population growth of one species in residential yards in one metro area. It did not measure whether adding natives to an individual yard changes that yard's bird population, which is a different question requiring a different design, and it did not establish a threshold for other species or other regions. A reader who wants the primary source should go to the PNAS paper; this manual cites the university's report of it, because that is what was read.

4. Nesting, which is the part gardens remove

Here is the fact that changes a garden design, and it is not about flowers.

The Xerces Society states that 70 percent of native bees are ground nesting, creating burrows in soil, and that they need access to bare soil and may be affected by tilling. The other 30 percent are cavity nesting, such as leafcutter and mason bees, needing cavities in dead wood, hollow stems or brush piles. Bumble bees nest in cavities underground or in trees, preferring abandoned rodent burrows or sheltered spots such as under brush piles.

Now consider what a well-kept pollinator bed usually looks like: mulched edge to edge, every stem cut down in fall, every bit of dead wood tidied away. Physical principle That bed has flowers for the adults and nowhere for the next generation. Mulch wall to wall removes the ground nesting; cutting every stem in fall removes the cavity nesting.

What to do instead, in order of how little it costs you:

ProvisionWhat it isWhat it costs
Bare groundA patch of soil, sunny, undisturbed, not mulched and not dugNothing. It is a decision not to do something
Standing stemsHollow or pithy stems left up over winter rather than cut in fallNothing, and it looks untidy to some eyes, which section 9 addresses
Dead woodA log, a stump, a few branches in a cornerNothing if you already have prunings
A brush pileA loose heap in an out-of-the-way placeSpace, which a small yard may not have
A bee hotelA manufactured or built nesting blockMoney, and it needs maintenance to avoid disease and parasites; the free options above come first

The bare-ground patch is the one to design in, because it has to survive your own tidying. Mark it on the plan, exclude it from the mulch order in the mulch calculator, and tell whoever else works in the garden that it is deliberate. A square foot or two in a sunny spot is a real provision; the plant count calculator has a field for the percentage of a bed you are keeping clear, which is this manual's own addition to the arithmetic rather than a published rule.

On cutting back, the honest position is that leaving stems standing over winter provides nesting and leaving them in spring provides it for longer, and that at some point you will want to cut them. Where exactly to draw that line is a judgment rather than a published rule, and it varies with your climate and your tolerance. What is clear from the Xerces figures is that cutting everything to the ground in fall removes the resource entirely.

5. Bloom across the whole season

The Xerces Society's list of what a pollinator planting needs puts food first and defines it specifically: an abundance of flowering plants with pollen and nectar throughout the growing season. Its habitat planning guidance says to use a wide variety of plants with overlapping flowering periods, so that food and nesting resources are available through the growing season, and notes that willows in particular nourish bumble bee queens in spring. That guidance is written for farms and hedgerows rather than for a 12-foot strip, which is worth saying when you borrow the rule.

The design consequence is a chart rather than a list. The bloom chart is a row per plant and a column per month, and the point of it is the empty columns.

Three gaps are common, and each has a different answer:

  • Early spring. The commonest gap and the most consequential, because queens emerging from hibernation need forage before most garden perennials start. Early-flowering woody plants and bulbs fill it.
  • Midsummer. A gap that appears when a planting is built from spring and fall perennials bought at those seasons, which is when nurseries sell them.
  • Late fall. The last forage before winter, and the one most often cut down just as it becomes useful.

Fill the chart from what the plants do where you live, not from the national range on a catalog page. A plant described as flowering June to September may do July to August in a short northern season. Test it The reliable way to fill the chart is to tick a month after you have seen the plant flower in that month in your own area, and to mark the rest as expected rather than observed.

6. Nativars: some are and some are not

A nativar is a cultivated variety of a native species: a selection or a hybrid bred for a flower color, a shorter habit, a longer bloom or a different leaf. The question is whether it is as good for pollinators as the straight species, and it is asked in a way that invites a yes or a no.

The most useful evidence available is Annie White's doctoral research at the University of Vermont, published on her own site. The trials were replicated experimental pollinator gardens in northern Vermont, zones 4a and 4b, running from 2011, monitoring pollinator visits to 12 native species and 14 native cultivars, with nectar production work on a further 2 species and 4 cultivars.

What she found, in her own framing:

  • The answer "is complicated; some cultivars are and some cultivars aren't."
  • Open-pollinated seed cultivars, such as Asclepias tuberosa 'Hello Yellow' and Monarda fistulosa 'Claire Grace', attracted as many pollinators as the species. She adds a caveat of her own about 'Claire Grace': it was not as cold hardy in zone 4, being a selection from a southern ecotype.
  • One cultivar, Veronicastrum virginicum 'Lavendelturm', attracted significantly more total pollinators than the species and bloomed for longer, which she reads as showing there is potential for pollinator-friendly cultivars to be selected for and marketed.
  • The clear trend: the more manipulated the cultivar, the less attractive it was, with repeated-selection and hybrid cultivars attracting significantly fewer pollinators in nearly all groups.
  • On nectar, Lobelia × speciosa 'Fan Scarlet' offered under 20 percent of the nectar energy of Lobelia cardinalis.

Convention That is a trial of a particular set of plants in one region, not a law about the word "nativar." What it supports is a working rule rather than a ban: prefer the straight species where you can get it; where you cannot, or where a cultivar is what fits the space, prefer an open-pollinated seed selection over a heavily bred hybrid; and treat a double flower, a radically changed color or a dwarf hybrid as the cases most likely to have lost something.

To check a particular plant rather than reasoning from the trend, Mt. Cuba Center's trial garden research publishes genus-by-genus reports comparing species and cultivars with pollinator observation. That is the route when the question is "is this specific cultivar worth buying," which is a question no general rule answers.

7. The design is the pest plan

The Xerces Society lists protection from pesticides as one of the requirements of pollinator habitat, alongside food and nesting. This manual's position follows from that and from the scope it set in What the garden has to do: it recommends no pesticide, because the design is the plan.

What that means in practice:

  • Choose plants suited to the site, which is the whole of Choosing plants that fit the place. A plant in the wrong light and the wrong soil is a stressed plant, and stressed plants attract more trouble.
  • Mix the planting. A monoculture of one shrub is one pest away from a bare bed.
  • Accept some damage. A planting intended to feed caterpillars will have chewed leaves on it, and that is the planting working rather than failing.
  • Ask what a plant was treated with before it reached you. It is a reasonable question for a nursery and a revealing one.
Caution

If you do use any pesticide, herbicide or fertilizer, the label is the legal document and it governs: what it may be used on, at what rate, and what precautions apply. Nothing in this manual overrides a label, and this manual recommends no product. For a diagnosis of what is actually wrong with a plant, your county extension office is the place to go, and Learn Vegetable Gardening's chapter on diagnosing before you treat covers the method of identifying a problem before acting on it.

8. Where the plants came from

Three questions to ask when buying for a pollinator planting, each of which a good nursery can answer:

  1. Is this the species or a cultivar? The tag answers this if you read the single quotes. It matters for the reasons in section 6.
  2. Where was it grown? Provenance matters most for seed and for anything intended to persist and self-sow, and least for a single specimen shrub. The USDA Forest Service's native plant materials program is the background to why.
  3. Was it treated with anything systemic? A systemic insecticide applied in production can persist in the plant, and a nursery that cannot answer is giving you an answer of a sort.

Beyond the nursery, the sources worth knowing are regional: your state's native plant society, your extension service, and Pollinator Partnership's free ecoregional planting guides, which are organized by zip code. When you use a guide, note its own date on the plant list, because these documents are revised and the one you downloaded is a snapshot.

9. Making a native planting read as deliberate

This is a design problem rather than an ecological one, and ignoring it is how a well-intentioned planting becomes a letter from an HOA or a coolness between neighbors.

The underlying issue is that the visual cues people read as "cared for" are mostly about edges and repetition, and a planting of mixed native perennials left standing over winter has neither unless you put them there. The planting is doing its job; it is being read as neglect.

What supplies the cues, in rough order of how much they buy you:

CueWhat it doesCost
A clean edgeThe single strongest signal. A cut edge, a mown strip, an edging board or a path says somebody decided where this stopsAn hour, repeated
A mown frameA strip of mown grass or low groundcover around or in front of a taller plantingSpace, and the mowing
RepetitionThe same plant repeated through a bed reads as a pattern; one of everything reads as a collectionNothing; it is a choice when you count plants in Drawing the planting plan
A fence, wall or low hedgeFrames the planting as a thing inside something, rather than as the lawn going wrongMoney, where one is not already there
A signStates the intent, which does a surprising amount of work with neighborsLittle, and some people find it twee
Height gradingLower at the front, taller behind, so the planting has a legible frontNothing; it is a plan decision

Alabama Extension's point from Beds, lawn and use areas applies with full force here: landscape plants should be arranged in defined, mulched beds rather than dotted individually into lawn. Convention A native planting inside a defined shape reads as a planting. The same plants scattered through grass read as a lawn nobody is mowing.

Where an HOA is involved, the sequence matters more than the cues. Read the covenant, find out what needs approval and how long it takes, submit the base plan from Measuring and drawing your site with the planting on it, and get the answer in writing before you plant. Ask the authority Some states have passed laws limiting what an association may prohibit in this area and others have not; what applies to you is a question for your state and your association, not for this manual.

10. See it: two pollinator plantings See it

Fictional example

Rosalind's strip, Burlington, Vermont. Known: 12 by 4 feet, with a marked gas line and its no-dig zone down the middle, so the plantable ground is narrow strips either side; part sun; sandy, free-draining; zone 5a; she wants pollinators and does not want the neighbors to think she has stopped mowing. Keystone step: her ecoregion's list names several woody genera, and none of them fits. The largest has a mature spread over 15 feet; the strip is 4 feet wide with a no-dig zone in it. Assumed: that she cannot plant anything woody on the no-dig zone, which she confirmed rather than assumed, by asking at the time of the locate. What she chose: no tree, which is the right answer and worth saying because the keystone advice is written for yards with room. Instead: three perennials from the keystone list's herbaceous genera, chosen for mature spreads of 2 to 3 feet, with bloom in May, July and September so that the chart has three separate months ticked. A 1-square-foot patch at the north end, sunny and away from the walking route, is marked on the plan as bare ground and excluded from her mulch order. Her design cue: a 6-inch mown strip along the driveway edge and the same plant repeated three times down the strip rather than three different plants, which is what makes it read as a planting from the sidewalk. Open: whether 1 square foot of bare ground does anything measurable, which no source read for this manual answers. She is doing it because it costs nothing and because wall-to-wall mulch demonstrably removes the resource.

Fictional example

Harlan's front yard, Austin, Texas. Known: 30 by 22 feet, full afternoon sun, zone 8b, a two-day watering restriction, an HOA that approves plans and requires a "maintained" front appearance, and a contractor doing the work. Keystone step: his ecoregion's list does include woody genera that fit a 660-square-foot front yard, and one small tree is in the plan. Assumed: that the HOA's "maintained" requirement is about appearance rather than about turf specifically; this is his assumption and it is the one thing that could stop the whole project, which is why it went to the board in writing before anything was ordered. What he chose: the design cues as requirements rather than preferences. A mown strip of low groundcover along the sidewalk, a defined steel edge on every bed, the same three plants repeated in drifts rather than a collection of twenty, height graded so the tallest material sits against the house, and a small sign. The bare-ground provision is in the back, not the front, because an unmulched patch in an HOA front yard is a conversation he does not need. Open: the board's answer, which nobody else can give, and whether the planting survives a two-day watering regime through its first summer, which is Planting and the first three years and depends on whether new plantings are exempt from the restriction.

11. Practice: a keystone, three perennials and a bare patch Practice

Minimum: one keystone plant that fits, three perennials with non-overlapping bloom, and a marked square foot of bare ground. Time: about 45 minutes. Materials: the plant list and its bloom chart, an ecoregion keystone list, and your own survey.

  1. Find your ecoregion's keystone list and read the woody genera first.
  2. Run each candidate through the filters from Choosing plants that fit the place, mature size last and hardest. Write down the one that fits, or write down that none does, which is a real answer.
  3. Add three perennials and tick their bloom months on the chart as you go, aiming for flower in as many separate months as your season allows.
  4. Look at the empty columns and decide whether to fill the earliest one, which is usually the most useful.
  5. Mark the bare patch on the plan, somewhere sunny and undisturbed, and note on the plant list that it is excluded from mulch.
  6. Write one line about edges: what will make this read as deliberate from wherever it is seen.
Go deeper

Buying from a native plant nursery

Worth the trip if there is one within reach. The questions that distinguish a good one: can they tell you the provenance of their stock; do they sell straight species as well as cultivars; can they tell you what a plant feeds; and will they tell you when something is a poor choice for your site. A nursery that answers all four is worth more than a plant list.

Leaving stems standing, and the spring cleanup question

The Xerces figures make the case for leaving hollow and pithy stems over winter: 30 percent of native bees nest in cavities, and standing stems are one of the places they do it. The question that follows, when exactly to cut in spring, does not have a published answer this manual found, and the advice that circulates varies with climate and with who is giving it. What is defensible: cutting everything down in fall removes the resource; leaving stems up through winter provides it; and cutting later in spring provides it for longer. Where you draw the line between that and a tidy bed is yours, and it is worth making the decision rather than defaulting to the fall cleanup.

If you have no room for anything woody

Then you have no room for the highest-value part of the keystone advice, and that is worth being clear about rather than pretending a container of perennials is equivalent. What a small space can still do: perennials from the keystone genera, flower across the season, bare ground, standing stems, and no pesticides. That is a real contribution and it is smaller than a tree. Both things are true.

12. Your garden Your garden

13. Check your work Check your work

Criteria

  • Your ecoregion is named, and the keystone list you used is identified with its publisher.
  • Any keystone plant on the list passed the mature-size filter rather than being chosen for its name.
  • The bloom chart has flower in at least three separate months, and the empty columns have been looked at.
  • Bloom months are ticked from what plants do where you live, with expected months marked as expected.
  • A patch of bare, undisturbed ground is marked on the plan and excluded from the mulch order.
  • The plant list says, per row, whether each plant is the species, a cultivar of a native, or not native here.
  • There is a written decision about edges and repetition.
  • Nothing in your notes claims an effect larger than the evidence behind it.

Worked example

Rosalind's planting (fictional) is the better example and the reason is that it says no. The keystone advice says plant a keystone woody genus; her strip is 4 feet wide with a no-dig zone down it; the smallest keystone woody plant on her ecoregion's list would reach 15 feet across. Writing "no tree, no room" on the plant list, rather than squeezing one in, is the correct application of the advice, and it is the application most readers resist. What she did instead is the achievable part: three perennials from the keystone genera, three separate bloom months, a square foot of bare ground marked and excluded from the mulch, and a mown strip and repetition so it reads from the sidewalk. Her open question is also honest: she does not know whether a single square foot of bare ground does anything measurable, and no source read for this manual tells her. She is doing it because it costs nothing and because the alternative, mulching wall to wall, demonstrably removes the resource. That is the right shape of reasoning when the evidence runs out.

Common mistakes

  • Stating the 70 percent figure as a target for your border. It is one study, one bird, one metro area, measured as woody biomass.
  • Treating "native" as a maintenance claim. It describes where a plant evolved.
  • Skipping establishment watering because the plants are native.
  • Mulching wall to wall in the bed you planted for bees.
  • Cutting every stem down in fall, which removes the cavity nesting.
  • Buying a heavily bred cultivar and assuming it does the same job as the species.
  • Rejecting every cultivar, which the Vermont trials do not support either.
  • Filling the bloom chart from a catalog's national range rather than from your own season.
  • Planting first and asking the HOA afterward.
  • Leaving out the edge, and discovering that the planting is read as neglect.

What is still unresolved

  • How many of each plant the bed holds: Drawing the planting plan, where repetition stops being an idea and becomes a count.
  • Exactly when to cut stems in spring: no source read here settles it; your own climate and tolerance do.
  • Whether a particular cultivar is as good as the species: Mt. Cuba Center's genus-by-genus trials, not a general rule.
  • What your HOA will permit: the board, in writing, and your state's law if it has one.
  • Whether a plant was treated with a systemic in production: the nursery.
  • What is actually eating something: your county extension office.

14. Knowledge check and scenario

Somebody tells you a yard needs to be 70 percent native. What is the accurate version of that claim?
It comes from one study of Carolina chickadees in residential yards in the Washington, D.C. area, by Narango, Tallamy and Marra, as reported by the University of Delaware. The finding was that chickadees needed yards above roughly 70 percent native plant biomass to sustain population growth, and that below it they switched prey, laid fewer eggs or skipped breeding. It is one bird species, one metro area, and the measure is woody plant biomass, not a count of perennials or a share of your border.
Why is mulching a pollinator bed edge to edge a problem?
Because the Xerces Society puts about 70 percent of native bees as ground nesting, creating burrows in soil and needing access to bare soil. Mulch over the whole bed removes that access. The provision costs nothing, a patch of sunny undisturbed soil left unmulched, and it has to be designed in and marked, because otherwise it gets tidied away by you.
Are nativars as good for pollinators as the straight species?
Some are and some are not, which is Annie White's own summary of her Vermont trials of 12 native species against 14 cultivars. Open-pollinated seed cultivars such as Asclepias tuberosa 'Hello Yellow' attracted as many pollinators as the species, and Veronicastrum virginicum 'Lavendelturm' attracted significantly more with a longer bloom. The clear trend was that the more manipulated the cultivar, the less attractive it became. Prefer the species where you can get it; prefer an open-pollinated selection over a heavily bred hybrid; check a specific plant against Mt. Cuba Center's trials.
What does the keystone idea buy a small garden?
Focus. The National Wildlife Federation reports Doug Tallamy and his team's finding that 14 percent of native plants, the keystones, support 90 percent of butterfly and moth species. So choosing from the keystone genera for your ecoregion buys more than choosing at random from a list of natives. In a small yard that usually means one keystone woody plant if there is room for one, plus perennials from the list, and saying so if there is not room.
Why does this chapter spend a whole section on edges and mown strips?
Because a planting that is doing its ecological job can be read as neglect, and that reading has consequences: a letter from an association, a complaint, pressure to mow it. The cues that say "deliberate" are mostly edges and repetition, and they cost almost nothing when designed in. Alabama Extension's point that plants belong in defined, mulched beds rather than dotted into lawn is the same idea from the design side.
(Go deeper) Why does this manual cite a university news report rather than the chickadee paper itself?
Because the paper could not be read. The publisher's site refused the request when this manual was researched and again when it was written, and nothing here is quoted from a source that was not read. The university's report of its own authors' study is what was read, and the chapter says so. The paper is Narango, Tallamy and Marra, "Nonnative plants reduce population growth of an insectivorous bird," PNAS, 2018, if you have access.
Scenario: you want a pollinator border along the front of a house in an HOA neighborhood. You have a list of twenty native perennials, a plan to leave everything standing over winter, and a free Saturday.
Two separate problems. The HOA one is procedural: a covenant that requires approval requires it before, and planting first puts you in a position where the cheapest resolution is removal. The design one is that twenty different perennials in one front border reads as a collection rather than a planting, with no repetition and no legible front, which is the reading that generates the complaint in the first place. Neither problem is about whether native planting is worthwhile.
This is the strongest path and it loses nothing ecologically. Approval first is the only sequence that works where a covenant requires it. Five or six plants repeated in drifts feed as much as twenty singletons and read as deliberate, which is the whole of section 9. A clean edge and a mown frame are the strongest cues available and cost an hour. Putting the bare ground and the standing stems where they are least conspicuous keeps the nesting provision without putting it in the front row. What it leaves open is the board's answer, which only they can give.
A reasonable fallback and worth holding in reserve rather than reaching for first. The front yard is often the sunniest ground on a small lot, and giving it up costs real habitat. Try option b, and if the board refuses, you will at least have the refusal in writing and a design that is ready to go in the back.

Checklist

Records your learning only; it is not a qualification.