The reason most pitches to IEEE Spectrum fail is simple enough to state in a sentence: they describe a product or a company when the magazine publishes engineering.
That sounds obvious written down and yet it accounts for the overwhelming majority of rejections. Founders and communications teams send material that would work at a business publication, framed around market opportunity, funding and customer traction, and it arrives at a magazine whose readers are electrical engineers, computer scientists and researchers who want to know how the thing works. The pitch is not too promotional. It is about the wrong subject.
Getting it right requires understanding what the publication is for, and that history is useful rather than decorative.
What IEEE Spectrum Actually Is
IEEE Spectrum is the flagship magazine of the IEEE, the professional body for electrical, electronics and computing engineers, and it has been published since 1964.
That lineage explains the editorial standard. The magazine exists to serve a membership of working engineers and researchers, which means its readers are not a general technology audience who happen to be interested in gadgets. They are people who will recognise when a claimed efficiency gain is within measurement error, who have opinions about which benchmark you used, and who will write in when a schematic is wrong. Editors know this about their readership and commission accordingly.
The coverage spans the full IEEE field: semiconductors, robotics, power and energy systems, telecommunications, computing, aerospace, biomedical engineering and the policy questions those fields generate. It runs news, long features, explanatory pieces, first-person accounts from engineers about projects they worked on, and opinion. The range of formats matters when you pitch, because the same subject can be a 600-word news item or a 4,000-word feature and those go to different editors with different appetites.
It is worth being direct about one thing. Our own placement catalog runs to 1,016 publications, from $150 regional outlets to $200,000 for Bloomberg, and IEEE Spectrum appears nowhere in it at any price. There is no paid route. Everything below is about earning a commission, which is slower and harder and produces something a paid placement cannot: a byline in a publication engineers actually trust.
What that trust is worth has changed recently too. When an engineer asks an AI assistant about a technology, the models are drawing on sources with reputations, and a Spectrum piece sits near the top of that pile for its subjects. A single well-reported article there outperforms a great deal of volume elsewhere in how your work gets described.
The other reason to aim here is longevity. Engineering magazine archives get cited for decades, and a piece explaining how a technology works does not go stale the way a funding announcement does. People find these articles years later while researching a problem, which makes the byline an asset that keeps appreciating rather than a traffic event that decays in a week.
Before pitching anything, spend an evening reading. Not skimming headlines, reading four complete pieces in your subject area end to end. You are looking for three things: how deep the technical detail goes before the writer stops explaining, how companies are referred to when they appear, and how often a piece ends without a conclusion because the engineering question is genuinely open. Each of those will adjust your draft more than any guidance you could be given secondhand.
Rule One and Two: Find the Engineering, Then Find the Tension
The first rule is that the subject of the piece must be a technical problem, not an organization.
Run this test on whatever you were planning to pitch. Strip out every proper noun: your company, your product name, your customers, your investors. What remains? If the answer is nothing, there is no article. If the answer is a description of a hard problem and an approach to solving it, you have something, and the company can go back in as context rather than as subject.
A worked contrast. “Our startup raised $12M to transform grid storage with proprietary battery chemistry” has nothing left after the proper nouns come out. “Sodium-ion cells lose capacity at low temperature in a way lithium does not, and here is the electrode design change that recovers most of it” survives the test completely, and the company’s name belongs in the third paragraph where it says who built it.
The second rule is that the piece needs a tension, which is the thing that makes an editor assign rather than file. Engineering writing without tension is a datasheet. The tension is usually a tradeoff, a surprise, a thing that should have worked and did not, or a disagreement among competent people.
Tradeoffs are the most reliable source. Every real engineering decision costs something, and the sentence naming what you gave up is almost always the most interesting sentence in a pitch. You got the latency down and the power draw went up. You made it manufacturable and lost twelve percent of the theoretical efficiency. Say so. A pitch that claims improvement in every dimension reads as marketing to an audience that knows better, and a pitch that names its own cost reads as credible.
Surprises are the second source. Something behaved unexpectedly, you investigated, and the explanation turned out to be interesting. This is the shape of a large fraction of the magazine’s best first-person pieces, and companies sit on these stories constantly because they look like failures internally.
Disagreement is the third and the most underused. If two groups of competent engineers have reached opposite conclusions about an approach, that is a story, and it is a story an editor can commission without having to vouch for either side. Pitching a disagreement you are a party to requires being honest that you are a party to it, and then being accurate about the other position. A pitch that describes the opposing argument fairly is unusually persuasive, because it demonstrates you could write the piece without turning it into advocacy.
What all three have in common is that they centre a question rather than an achievement. The pitch that gets commissioned does not say look at what we did. It says here is something that was not obvious, and here is what it turned out to be.
Rule Three, Four and Five: Evidence, Format and Timing
The third rule is that every claim needs a method behind it.
An editor at an engineering publication will ask how you measured it, and the pitch should answer before they ask. Give the test conditions, the instrumentation, the sample size and the comparison baseline. If the numbers come from simulation rather than hardware, say that plainly in the pitch, because it will come out in edit and discovering it late damages the relationship more than disclosing it early costs you.
Published work helps enormously here. A conference paper, a preprint or a peer-reviewed article gives an editor something to check and signals that the work survived review by people in your field. It also removes the commonest objection, which is not disbelief but unwillingness to be the first party to vouch for an unverified claim.
The fourth rule is to pitch a format, not just a topic. Read four recent pieces in your subject area and note their length, structure and voice. Then say which you are proposing: a news item on a specific result, an explanatory feature on how a class of technology works, a first-person account of a project, or an opinion piece arguing a position. Editors commission into slots, and a pitch that names its slot is far easier to say yes to than one that offers an interesting subject and leaves the shaping to them.
Length matters more than people expect. A pitch proposing 3,500 words on a narrow result is asking for a commitment the subject cannot carry. A pitch proposing 800 words on a subject that deserves a feature undersells itself. Match the ambition to the material.
Visuals belong in the pitch too, and this is a real advantage engineers have over most pitchers. Technical magazines run diagrams, charts and photographs of actual hardware, and they are chronically short of good ones. Saying you can supply a clean schematic, a measured performance curve and high-resolution photographs of the test rig makes a commission easier to approve, because it removes work from the art side. Offer them. Do not attach them to the first email.
One more thing to get right in the pitch: say who else is working on this. An editor’s first question after deciding a subject is interesting is whether it is new, and the fastest way to answer is to name the other groups in the space and say what is different about your approach. Pretending to be alone in a field invites a quick search that contradicts you.
The fifth rule is timing, and engineering publications have a specific rhythm. Research results cluster around conference calendars in each subfield, so a piece tied to a result presented at a major conference is most pitchable in the weeks around it. Policy and standards stories follow regulatory and standards-body timelines, which are slow and published in advance, which makes them unusually easy to plan for.
One practical note on who to email. Spectrum’s masthead and author pages identify editors by area, and the right target is the editor whose name sits on recent pieces in your subfield. One named editor, one specific proposal, a subject line that states the technical claim. Follow up once after two weeks. An editor who declines a well-made pitch will often say why, and that reply is worth more than most acceptances elsewhere, because it tells you exactly what the magazine wants from someone in your field.
When You Are Not the Right Author
The hardest version of this advice is also the most useful: sometimes the correct move is to not write it.
An editor may be interested in the subject and unconvinced that you should be the one writing about it, usually because you have a commercial interest in the conclusion. That is a reasonable position and it has a productive answer. Offer yourself as a source instead. Give the editor access to the engineers who did the work, the data, the test setup and the people at other companies who disagree with you. A piece reported by a journalist, quoting your engineers, carries more weight than a byline you wrote yourself, and it costs you nothing but control.
This is where companies with strong technical work routinely lose. They want the byline, decline the source role, and end up with neither. Across more than 2,000 published articles and a journalist network running past 1,200 contacts, the pattern we see is consistent: becoming a reliable source for a beat reporter generates far more coverage over two years than any single placement, because the reporter comes back.
Build the technical story until it survives the proper-noun test, name the tradeoff honestly, attach a method to every number, and send it to one editor who covers your subfield. If the answer is that they want to report it themselves, say yes.