G2 marketing strategy for B2B: A review site playbook from profile to pipeline

A G2 marketing strategy for business-to-business (B2B) software does two jobs: it builds vendor preference before a formal purchase starts, and it catches accounts already comparing vendors. Preference matters because Forrester found 68% of B2B buyers have a front-runner vendor in mind at the start, and that front-runner wins 80% of the time. Our five-step program runs from profile and reviews to intent routing and pipeline measurement.

What a G2 marketing strategy has to do for B2B pipeline

A G2 marketing strategy has two jobs. It has to build preference before a buyer starts a formal purchase, and it has to catch the accounts already comparing vendors. Most programs do only the second job, and in our reading that is the weaker half.

Forrester found that 82% of B2B buyers trust coworkers and management and 79% trust current vendors, while 44% trust social media influencers.

Share of B2B buyers who trust each information source

Item	Value
Coworkers and management	82%
Current vendors	79%
Social media influencers	44%

Source: Forrester, B2B Buyers Rate Their Most Trusted Information Sources, 2025

The case for the first job comes from Forrester. That gap is the problem a review program addresses.

Buyers are also doing the work alone. A Gartner survey of 646 B2B buyers conducted from August through September 2025 found that 67% prefer a rep-free experience and 45% reporting they used AI during a recent purchase.

Who buyers believe matters too. Forrester's trust research puts industry peers, analysts, and even vendor executives and customers in a middle tier, with trust levels from 66% to 72%. We read a review site as a place where customer voices reach a buyer who has not yet talked to sales.

The five-step program

This is our method, covered step by step in the sections that follow.

Steps 1 to 3 are demand creation: they shape the front-runner list. Steps 4 and 5 are demand capture. The split is laid out in demand creation vs. demand capture.

A working rule: do not judge the program on step 4 alone. If buyers arrive with their pick already made, review-site intent mostly confirms deals that steps 1 to 3 shaped. Report both halves, and compare win rates against your win rate benchmarks for accounts that did and did not touch your profile.

Do B2B buyers read software reviews

In our reading, yes, and the research places peer voices beside vendor material rather than below it. Forrester's 2025 buying research found that B2B buyers consider industry peers among their top five trusted sources. The same note lists the channels customers use to reach buyers: word of mouth, peer reviews, testimonials, communities, and reference calls. A G2 profile is one of those channels, and it is one you can manage.

Eighty-five percent of buyers also value the information they get from peers, colleagues, and customers.

Vendor content carries a problem. Buyers said much of the material they receive from vendors is extraneous or even useless (51%). In our reading, review text answers that problem because a customer wrote it, so reuse it wherever your own copy has to prove a claim.

Gartner's 2025 survey adds context on volume: buyers reported using an average of seven information sources during a recent purchase. A review site competes for one of those slots. The category you list under on G2 decides which competitors you sit beside, so we think category choice deserves the same review as the profile copy covered in Step 1.

Ways to reuse reviews outside the review site

• Product and pricing pages: place a review widget or earned badge near the pricing table, where a buyer weighs cost against proof.
• Sales decks: add two or three short quotes per persona, with reviewer role and company size, on the slide that covers the objection the quote answers.
• Paid ads: test a review quote as ad copy against your standard value proposition, and track cost per opportunity rather than click rate.
• A monthly tagging pass: read every new review, tag it by theme, and route each theme to an owner.

Monthly tagging template

Keep the tag list stable from month to month so theme counts can be compared over time. When a complaint theme falls after a fix ships, note it in the same sheet you use to track pipeline from the profile.

Step 1: complete the profile on every review site you list on

A review profile is a self-service research page that your company does not fully control. Buyers use it the way Gartner describes: they favor online self-service tools over sellers when they are gathering general information. In a survey of 632 B2B buyers run in August and September 2024, 61% preferred a rep-free buying experience overall.

The same survey points to the main risk on a half-finished profile, which is drift. 69% of B2B buyers reported inconsistencies between a vendor's website and what its sellers told them. Gartner's advice to sales leaders was to ensure synchronized company information across platforms. We think a review profile counts as one of those platforms.

The profile completeness checklist

This table is our method. Marketing operations runs it the same way on G2, Capterra and TrustRadius. Every row names a single owner and a single source of truth, so a pricing change or a feature rename only has to be made once upstream.

How to run it

Copy the table into a shared sheet with one tab per review site. Then:

1. Assign each row to a named person, not a team alias.
2. On the refresh date, the owner compares the profile field with its source of truth and writes "match" or the fix needed.
3. Marketing operations logs the date each site's profile last passed every done-when check.
4. Sales enablement reviews the pricing and feature rows each quarter, because those are where seller talk tracks and profiles tend to drift apart, in our reading.

Track the pass date next to your other channel health metrics in the demand gen dashboard template. Once a profile is complete and stays in sync, review volume becomes the next constraint.

How to get more B2B reviews with a steady request cadence

Step 1 got the profile complete. Step 2 is volume, and volume needs an owner. We assign it to customer success, because they see every onboarding milestone and every renewal date. Marketing supplies the review links and the tracking, and the agreement goes in writing like any other service-level agreement.

What the research says about review volume

The best-known numbers come from retail, not software. The Medill Spiegel Research Center, in findings first published in 2017, reported that purchase likelihood for a product with five reviews is 270% greater than for one with none. It also found that the marginal benefit begins diminishing rapidly after the first five reviews.

Price changed the effect. Displaying reviews lifted conversion 190% for a lower-priced product and 380% for a higher-priced one. In our reading, B2B software sits at the high-consideration end, so the direction should hold, but nobody measured it on software buyers in this study. The same research found purchase likelihood typically peaks at ratings in the 4.0 to 4.7 range. A perfect score is not the target.

The monthly cadence

Run it as one batch each month. Log requests sent and reviews published.

Plan the next quarter from that rate instead of from a guess.

What the FTC rule prohibits

These are facts about the rule, not legal advice. The Federal Trade Commission (FTC) rule went into effect on October 21, 2024. It prohibits compensation or other incentives conditioned on the writing of consumer reviews expressing a particular sentiment. FTC staff add that paid or incentivized reviews would also be considered consumer testimonials under the rule.

The rule has an exception for generalized solicitations to purchasers to post reviews. Asking only happy customers is not specifically banned by the rule, but staff say this practice could violate the FTC Act. Staff also list an unusually large number of reviews in a short period as a warning sign for purchased reviews. We think a steady monthly batch sent to every eligible account keeps you clear of both patterns. Take questions about your specific program to counsel.

How to respond to negative reviews

A critical review is a sales asset if the response is fast, specific and calm. The Medill Spiegel Research Center advises brands to embrace negative reviews, because negative reviews can have a positive impact as they establish credibility and authenticity. A buyer comparing vendors reads the one-star review and then your reply. Treat that reply as part of the evaluation.

Set it up like any other service level agreement:

1. Route review alerts from each site to one shared channel.
2. Assign the owner by topic: product gaps to product marketing, support failures to the support lead, billing to customer success.
3. The owner posts the public reply and opens an internal ticket in the same session.
4. Customer success closes the loop privately and logs whether the issue was fixed.

Response template

Keep replies under 100 words and in this order:

1. Acknowledge the specific problem the reviewer named, in their terms.
2. State what you know: a fix shipped, a fix is scheduled, or the behavior is by design and why.
3. Offer a direct route: a named role and a contact path, not a generic support link.
4. Close without asking them to change or remove the review.

Here is a worked example. A reviewer gives two stars and says that report exports time out above a certain row count. The reply names the export timeout, says the limit was raised in the latest release, names the support lead who will check their account, and stops there.

That last row matters. Under the federal rule, consumer reviews include consumer ratings regardless of whether they include any text or narrative. A bare one-star rating is still a review.

What not to do

Do not answer a bad review by asking staff to post good ones.

Keep legal threats out of review disputes too. The rule defines an unfounded or groundless legal threat as one based on contentions unwarranted by existing law. Send any dispute about a review's accuracy to counsel. This section is not legal advice.

Last, check who wrote the review. Spiegel found that anonymous buyers are significantly more likely to give one- and two-star reviews, and that verified buyer badges improve the odds of purchase by 15%. Ask each review site how it verifies reviewers, and answer anonymous criticism with the same care as verified criticism.

What review site buyer intent data is and how to route it

Review sites expose three kinds of account-level activity that matter for pipeline. Treat them as different signals, because they sit at different points in a purchase. For the general concept, see our intent data explainer.

• Category visits: the account is reading about a problem space or software category. It may not know your name yet.
• Comparison visits: the account is viewing your product side by side with named alternatives. It has a shortlist.
• Profile visits: the account is reading your page, your reviews or your pricing details. It is checking you specifically.

A starting scoring and routing model

The points below are our own example weights, not a benchmark. Tune them against closed-won history in your lead scoring model template.

Worked example. A target account logs one category visit, one profile visit and one comparison visit in the same week. That scores 10 + 20 + 30 = 60. With a routing threshold of 50, it goes to the owning SDR with the competitor names attached. A second account with two category visits scores 20 and stays in nurture until a profile or comparison visit arrives.

Three routing rules keep the model honest:

1. Score at the account level, then check fit (industry, size, region) before routing. Intent from a poor-fit account is still a poor-fit account.
2. Decay points after a set window, such as 30 days, so stale research does not keep an account in the SDR queue.
3. Pass the context, not just the score. The SDR needs to know which competitor was compared and which review themes the account read.

Why fast, relevant follow-up still matters

Buyers who research on review sites still want a human check before they commit. A Gartner survey of 645 B2B buyers conducted from August through September 2025 found that 69% of B2B buyers turn to sales reps to validate AI-generated insights. The same survey found that 67% prefer a sales rep-free experience.

We read those two figures together: buyers want to do the work alone, then confirm it with someone who knows the product. So the SDR touch after a comparison visit should answer the comparison. Open with how you differ from the named competitor and offer proof from your reviews. Leave the generic demo request out.

Track how many routed accounts become a sales-qualified lead (SQL) within your scoring window. If comparison-visit accounts do not convert at a higher rate than category-visit accounts, adjust the weights.

Worked example: cost per opportunity from a $36,000 review-site year

Take a paid review-site tier that costs $36,000 for the year and, in our example, produces 120 leads (buyers who reached sales from the profile or from an intent account that was worked), a cost per lead of $300. That one budget line has to be read four ways: per review, per intent account, per opportunity and as payback. The formula is the same each time: annual cost divided by the output you count, so the only real work is counting the output honestly.

On the same $36,000 year, cost per opportunity falls from $3,000 to $1,500 as the lead-to-opportunity rate rises from 10% to 20%.

Worked example: cost per opportunity from a $36,000 review-site year

Item	Value
10% of leads to opportunity	3000 US dollars per opportunity
15% of leads to opportunity	2000 US dollars per opportunity
20% of leads to opportunity	1500 US dollars per opportunity

Source: Take the Market worked example (illustrative inputs)

The table holds spend and the 120 leads fixed and changes one input: the share of those leads that become opportunities. Moving from 10% to 20% halves cost per opportunity, from $3,000 to $1,500, without a dollar of extra spend.

The four formulas

1. Cost per review = annual cost ÷ reviews published in the year that came from your own request cadence. Leave out reviews that would have arrived anyway.
2. Cost per intent account = annual cost ÷ accounts that showed intent on the site and were routed to an owner. Count accounts, not signals, or one busy buyer inflates the number.
3. Cost per opportunity = annual cost ÷ opportunities created from those routed accounts inside a fixed window, such as 90 days.
4. Payback in months = cost per opportunity ÷ (win rate × average first-year contract value × gross margin ÷ 12). Our CAC payback entry covers the full method, and the win rate benchmarks give a reference input if your own sample is thin.

What moves the conversion rate

In our reading, the lead-to-opportunity rate depends more on what happens after routing than on the tier itself. Gartner found that 73% of B2B buyers actively avoid suppliers who send irrelevant outreach. An intent account dropped into a generic sequence is likely to stay in the 10% row. A rep who opens with the comparison or category the buyer was reading on the site has a better chance of reaching the 20% row.

The decision rule

Set the paid tier's cost per opportunity next to the same figure for your other demand capture channels, using the same opportunity definition and window. If it sits at or below them, renew. If it sits above them, fix the routing and the first touch for two quarters before you cut. Then rerun the table with your actual rate. The demand gen budget allocator lets you test where the $36,000 would sit against other lines.

How to measure pipeline from review sites, with a 90-day timeline

Review sites feed pipeline through paths a click tracker misses. A buyer reads a comparison grid, asks an AI assistant for a shortlist, then types your URL two weeks later. Gartner's 2025 survey found 45% reporting they used AI during a recent purchase. We think that is one more reason last-touch reports undercount review sites.

So we measure with six metrics and one form field. The definitions below are our method.

Keep sourced and influenced in separate columns and never add them together. Our sourced versus influenced method explains why.

The self-reported attribution field

Add a required free-text field to every demo form that asks where the buyer first heard about you. Free text catches answers a dropdown would flatten, such as a named review site or an AI assistant. Marketing operations codes the answers weekly. A review-site answer counts as sourced when the CRM first touch is blank, direct or organic search. The full setup is in our self-reported attribution playbook.

The 90-day timeline

Ninety days is short against most B2B cycles. Read pipeline created and stage progression at day 90, not closed revenue. Check the gap against our sales cycle length benchmarks before anyone calls the channel a failure.

What this page does not cover

This page does not compare review sites or their paid tiers, and it does not give actual G2 pricing: the $36,000 figure is a worked example. The notes on the FTC rule describe the rule as facts and are not legal advice, and the cost and payback math is not financial advice.

Related on Take the Market: Self-reported attribution: how to design the field, code the answers and reconcile them with software attribution; ABM playbook for one-to-few programs: the account list, the plays, the metrics and a 90-day plan; Google Ads for B2B SaaS: campaign structure, match types, negatives and offline conversion import.