How'd you fix Leadership Connect's revenue issues in 2026?

The two options on the table: broaden the database or narrow to a signal

Every stalled data business eventually faces the same fork, and Leadership Connect's version of it was unusually clean. Option A was the incumbent instinct: keep investing in coverage. Add more people records, more committee staff, more state legislators, more agency org charts, and win on comprehensiveness. Option B was the harder pivot: accept that coverage is table stakes, freeze the database roadmap at "good enough," and pour engineering and go-to-market energy into turning that data into a timely signal that changes what a customer does this week.

Option A is seductive because it is measurable and it feels like progress. You can put a number on it — records added, coverage percentage, refresh cadence — and every number goes up and to the right. The problem is that the buyer does not experience coverage. A government-relations director does not open a procurement review and say "this vendor has 340,000 contact records versus 310,000." They say "which of these tells me something I did not already know, fast enough to act on it?" Coverage competes on a dimension the buyer cannot easily verify and does not emotionally weight. Worse, in a fragmented market where several established vendors each own a depth niche — federal lobbying and PAC intelligence, federal procurement, member-of-Congress research, bill tracking, narrative journalism — a generalist that is roughly seventy percent as deep as each specialist is nobody's first choice for anything. Seventy percent across five categories reads to a procurement committee as zero percent in any one.

Option B is uncomfortable because it means telling your own team that the thing they have been proud of for a decade is no longer the product. It also means the sales narrative changes from "look how much we have" to "look what we caught." That is a harder pitch to script and an easier pitch to close, because it attaches to a dated, urgent event rather than an abstract inventory.

There is a third option most turnarounds try first and should usually skip: cut price to hold volume. Discounting a comprehensiveness pitch does not fix a comprehensiveness pitch. It converts a differentiation problem into a margin problem and trains the field to lead with price, which permanently caps what the intelligence layer can ever charge. If the revenue plan has any hope of moving average contract value up, discounting has to be off the table from day one, and the RevOps team has to enforce a discount-approval floor rather than let it drift deal by deal.

The honest read is that Option A and Option B are not really symmetric. Option A has a ceiling defined by the specialists; you can spend against it forever and still arrive second. Option B has an unproven floor but a genuinely open ceiling, because timeliness is a dimension nobody in this market has fully claimed. When a market is fragmented across depth niches, the unoccupied axis is usually speed.

How to decide between broadening and narrowing

The decision should not be made on conviction. It should be made on evidence you can gather in about three weeks, using customers you already have. Four tests, in order.

Test one: the churn interview. Pull every logo that did not renew in the last eighteen months and ask a single unleading question — what were you actually trying to do when you opened our product? If the answers cluster around "look someone up," the account was always going to churn to the cheapest lookup tool, and no amount of added coverage saves it. If the answers cluster around "find out whether something was about to hurt us," you have a signal business hiding inside a database business.

Test two: the login-pattern read. Look at session frequency and depth. A reference tool gets opened when someone needs a name — spiky, shallow, unpredictable. An intelligence tool gets opened daily, first thing, as part of a morning routine. If a meaningful slice of your accounts already shows daily-habit behavior, the narrowing option has product-market evidence before you spend a dollar on it.

Test three: the loss-reason audit. Go through competitive losses and separate "lost on data depth" from "lost on we already have something." Losses on depth mean you are in a coverage war you cannot win. Losses on redundancy mean the buyer sees you as a replacement rather than an addition — which is exactly the framing the integration-layer pivot dissolves.

Test four: the save story. Ask customer success to find three specific moments where the product prevented a real problem — a rule change caught in time, a committee reassignment noticed before a filing, a hearing schedule that let a team brief its board a day early. These almost always exist and are almost never documented, because nobody was measuring them. If you cannot find three in a week, the signal thesis is weaker than you think and you should slow down. If you find a dozen, you have the entire 2026 marketing narrative already lived by real customers.

The decision tree matters more than the decision. A turnaround that narrows on instinct and then wobbles six months later does more damage than either path taken cleanly, because the field never learns a stable story and the roadmap thrashes. Write the decision down, name the evidence that produced it, and name in advance what evidence would reverse it. That last part is what keeps a repositioning from becoming a religion.

The numbers behind each path

Model both paths honestly before choosing, with ranges rather than false precision, and with the assumptions written where the board can argue with them.

The broaden path. Assume the current motion: mid-five-figure deals, sales cycles running three to four months, and a win rate suppressed by head-to-head depth comparisons. Adding coverage costs real money — data acquisition, verification staff, refresh infrastructure — and it lands as a modest win-rate improvement, not a pricing change. If you spend heavily on coverage and lift win rate a few points while holding contract value flat, revenue grows roughly in line with headcount. That is a linear business. It is not a broken one, but it never escapes the specialists, and every renewal is a fresh price fight. Net revenue retention in that world hovers near or below one hundred percent, because expansion has no natural trigger: nobody buys more seats of a lookup tool.

The narrow path. Three levers move at once, which is why the arithmetic looks different.

*Contract value.* An intelligence layer priced against avoided downside rather than against a competing directory can support several times the per-account revenue of a seat-based reference tool. The mechanism is not magic pricing — it is that the buyer changes. Reference tools are bought by a research manager with a departmental budget. Risk intelligence is bought by whoever owns the consequence of being surprised, and that person has a materially larger budget and a shorter approval path.

*Cycle time.* Cycles compress when the first meeting contains a dated event rather than a feature tour. "Here is what your committee calendar looks like for the next twenty-one days, and here are the two items that touch your license" is a discovery call, a demo, and a business case simultaneously. Realistically, a three-to-four-month cycle can be halved, not because you rush the buyer but because you skip the two meetings that existed only to establish relevance.

*Retention and expansion.* Workflow embedding is what turns a good year into a durable one. A tool that fires alerts into Slack and writes tasks into the CRM gets used by people who never log into your interface at all, and usage by non-logins is the strongest renewal predictor there is. Expansion becomes structural: a second regulated business unit, a second state footprint, a second team that wants the same alerts.

What it costs and what it risks. The narrow path is not free. You are choosing to under-invest in coverage for four to six quarters, which means at least a few competitive losses you would previously have won. You need engineering for integrations and alert infrastructure, and you need product management that understands both legislative process and CRM plumbing — a genuinely scarce hire. You will lose some existing accounts that bought the old thing and do not want the new thing; budget for that in the retention model rather than pretending it away.

The honest downside case. Model the version where narrowing produces only half the projected contract-value lift and cycle compression stalls at twenty percent. If that scenario still beats the broaden path on a two-year horizon, the decision is easy. If it does not, you are betting the company on execution rather than on strategy, and the board deserves to know that before the reposition ships, not after. Most RevOps teams skip the downside model because it is unflattering; it is the single most useful slide in the deck.

One more number worth tracking that rarely makes it into a revenue plan: time-to-first-alert. How many days pass between contract signature and the moment a customer receives an alert they consider genuinely useful? If that number is under a week, retention takes care of itself. If it stretches past a month, onboarding is silently killing the thesis regardless of how good the pitch was.

Implementation and sequencing across the year

Sequencing is where most repositionings die. Teams announce the new story before they can deliver the new product, the field pitches something that does not exist yet, and the first three deals set an expectation nobody can meet. Run it in four phases and do not let a phase start early.

Phase one — evidence and instrumentation, roughly the first quarter. No external messaging changes. Run the four decision tests. Document real save stories with named customers, dates, and what was avoided. Instrument the product so you can measure alert delivery, alert open rates, and time from event to notification — you cannot sell timeliness you have never measured. In parallel, RevOps rebuilds the pipeline model: new stages that reflect an event-driven sale, a discount floor, and a segmentation cut that separates regulated-industry accounts from general research buyers.

Phase two — the wedge product, roughly the second quarter. Ship the narrowest useful version of the alert layer to a small set of design partners. Narrow means one event type, one delivery channel, one integration. Resist the urge to launch a platform. Design partners should be existing customers who already show daily-habit usage — they will tolerate rough edges and they will tell you what the alert should have said. Track a single metric: did the alert change what the customer did that day? Any alert that does not change behavior is noise, and noise is how alert products die.

Phase three — packaging and channel, roughly the third quarter. Split the offer. Keep a lower-priced data tier for the buyers who genuinely want lookup, and price it deliberately as the commodity it is. Put the alerting, scoring, and CRM integration in a higher tier and price it against consequence. Bundle at the team level rather than per seat, because the intelligence layer gets more valuable as more of the team sees the same signal — per-seat pricing actively fights your own adoption motion.

Channel work starts here, and it is the highest-leverage underused lever. Distribution partners already sitting in the buyer's workflow — procurement platforms, member associations, CRM marketplaces — reach accounts that never answer cold outreach. Partner deals take longer to negotiate than anyone plans for, so start them a quarter before you need the pipeline. Expect referral economics in the mid-teens to mid-twenties percent range depending on how much of the sale the partner actually does, and expect the first partner to take twice as long as the second.

Phase four — full go-to-market, fourth quarter. Now the story goes public. Retrain the field on an event-led first call. Rewrite the website around what the product catches, not what it contains. Publish the save stories as case studies with real numbers. Update the pitch deck so page one is a calendar of upcoming risk, not a coverage map.

The internal work nobody schedules. Two things break repositionings from the inside. The first is compensation: if the plan still pays on logo count, the field will keep chasing small fast deals and the contract-value target will never move. Comp has to change in the same quarter the packaging changes, with accelerators on the intelligence tier and a spiff on multi-team bundles. The second is customer success capacity. An intelligence product needs someone to tune each account's alert thresholds in the first thirty days, or every customer drowns in irrelevant notifications and concludes the product is noisy. That is a staffing decision made in phase one, not a problem discovered in phase four.

What the adjacent playbook looks like. This sequencing is not specific to government data. Any reference-data business facing the same squeeze — market research, compliance libraries, credentialing databases, industry directories — runs the identical four phases. The pattern generalizes because the underlying failure generalizes: reference data commoditizes, timing does not. When a buyer can get eighty percent of your content from three cheaper places, the remaining defensible product is knowing something first and delivering it into the place where work already happens.

What changes for the RevOps team specifically

The pivot rewrites the operating layer, and if RevOps treats it as a marketing exercise the numbers never follow the story.

Segmentation. The old cut was probably by company size or by federal-versus-state. The new cut is by consequence exposure: how many regulated licenses, how many jurisdictions, how expensive is a surprise. Accounts with high exposure get the intelligence pitch and a senior seller. Accounts with low exposure get the data tier and a self-serve or low-touch motion. Mixing them in one territory guarantees the reps default to the easier, smaller sale.

Stage definitions. An event-driven sale does not fit stages built for feature evaluations. Replace "demo completed" with something like "risk event identified in the customer's own calendar," and "proposal sent" with "consequence quantified with the customer's own numbers." Stages should describe what the buyer has confirmed, not what the seller has performed.

Forecast hygiene. Compressed cycles distort forecasting for at least two quarters because historical stage-conversion rates no longer apply. Run the old and new models side by side through the transition and reconcile monthly rather than trusting either alone.

Attribution for the channel motion. Partner-sourced pipeline needs its own source field, its own conversion baseline, and its own margin math from day one. Retrofitting attribution after the first ten partner deals is a multi-week cleanup that could have been a fifteen-minute field configuration.

The dashboard that actually matters. Three numbers on one screen: time-to-first-useful-alert by account, percentage of accounts with a weekly-active integration, and intelligence-tier contract value as a share of total. Those three predict whether the fix is working long before bookings confirm it.

Related questions

What if the coverage gap is real and customers keep citing it?

Then buy or partner for the specific gap that shows up in loss reasons, rather than funding a general coverage program. Targeted acquisition of one dataset closes one recurring objection. Broad coverage investment closes none of them decisively.

How do you sell a higher-priced tier to customers already on the cheap one?

Do not upsell on features. Wait for a moment where the intelligence layer would have caught something they missed, then show them the timeline. Retroactive near-misses convert far better than roadmap decks, and they cost nothing to produce.

Does the free or low-cost tier cannibalize the enterprise deal?

Rarely, if the tiers are cut along value rather than volume. A cheap tier that is a smaller version of the expensive tier cannibalizes. A cheap tier that is a genuinely different job — lookup versus early warning — feeds the pipeline instead.

How long before the repositioning shows up in bookings?

Expect two to three quarters of lag. Pipeline mix shifts first, cycle time second, contract value third, bookings last. If leadership judges the pivot on bookings at the six-month mark, they will kill it right before it works.

FAQ

What was the actual revenue problem, stated plainly?

Leadership Connect was competing on comprehensiveness in a market fragmented across several specialist vendors, each deeper in its own niche. Being reasonably good at everything meant being the obvious choice for nothing, which produced long sales cycles, price-driven renewals, and contract values that could not grow. The problem was positioning expressing itself as a revenue number.

Why does "intelligence layer, not database" change the economics?

Because it changes the buyer and the trigger. A database is bought by a research function on a fixed departmental budget and evaluated against cheaper alternatives. A risk signal is bought by whoever owns the consequence of being caught off guard, on a budget sized to that consequence, with urgency supplied by a real calendar date rather than a renewal cycle.

Is integrating with existing CRMs really that important, or is it a nice-to-have?

It is the difference between a tool people log into and a tool people rely on. When alerts land in the CRM record and the messaging channel the team already lives in, the product gets consumed by people who never open your interface. Usage by non-logins is the single strongest predictor of renewal in this category.

What is the most common way this fix fails?

Announcing before shipping. The field starts pitching an intelligence layer that does not exist yet, early deals set expectations the product cannot meet, and the first churn wave discredits the whole thesis internally. Run the evidence phase quietly, ship a narrow wedge to design partners, and only then change the public story.

How do you keep alert volume from becoming noise?

Treat every alert as a behavior test. If a notification does not change what someone does that day, it should not have fired. Tune thresholds per account during the first thirty days as a staffed onboarding step, and measure alert-to-action rate as a product health metric with the same seriousness as uptime.

Does this pattern apply outside government data?

Yes. Any reference-data business under commoditization pressure — compliance libraries, market research, credentialing registries, industry directories — faces the same fork. Content commoditizes; timing does not. The durable product is knowing something first and delivering it where the work already happens.

Sources

• https://hbr.org/2016/09/know-your-customers-jobs-to-be-done
• https://www.mckinsey.com/capabilities/growth-marketing-and-sales/our-insights
• https://www.gartner.com/en/sales
• https://www.bain.com/insights/topics/customer-strategy-and-marketing/
• https://www.forrester.com/research/
• https://www.sba.gov/business-guide/manage-your-business
• https://openviewpartners.com/blog/
• https://a16z.com/enterprise/
• https://www.priceintelligently.com/blog

Related on PULSE

• [How do you coach a rep to improve their cold-call connect rate?](/knowledge/q13878)
• [What's a good cold call connect rate, conversion rate, and dials-to-meeting math in 2027?](/knowledge/q10861)
• [What's the right approach to running deal-room collaboration with buyers — Slack Connect, shared Notion, or stay on email?](/knowledge/q220)
• [How do you fix broken lead routing when sales on Outreach and leadership only reviews win rate monthly on Dynamics 365 ?](/knowledge/q10408)
• [How do you fix call recordings not tied to opps when parent-company rollup reporting and leadership only reviews win rate monthly on Dynamics 365 ?](/knowledge/q10338)
• [How do you fix broken lead routing when no dedicated RevOps hire yet and leadership only reviews win rate monthly on Dynamics 365 ?](/knowledge/q10268)